Download:
pdf |
pdfTANF
SAMPLING
AND
STATISTICAL
METHODS
MANUAL
October 1999
TANF
SAMPLING AND STATISTICAL METHODS MANUAL
TABLE OF CONTENTS
Page
1100
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1110
Purpose of the Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1200
BASIC STATISTICAL CONCEPTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1210
Sampling and Non-Sampling Errors . . . . . . . . . . . . . . . . . . . . . . . 5
1220
Common Types of Scientific Sampling Techniques . . . . . . . . . . . 5
1221
Simple Random Sampling . . . . . . . . . . . . . . . . . . . . . . 6
1222
Systematic Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1223
Stratified Random Sampling . . . . . . . . . . . . . . . . . . . . 6
1224
Allocation of Stratified Sample . . . . . . . . . . . . . . . . . . 8
1224.1
Proportional Allocation . . . . . . . . . . . . . . . 8
1224.2
Optimal Allocation . . . . . . . . . . . . . . . . . . . 8
1230
Validity and Reliability of Statistical Data . . . . . . . . . . . . . . . . . . 9
1231
Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1232
Precision -- Computation of the Confidence Interval 10
1232.1
Computation of Sample Size to Obtain a
Desired Precision . . . . . . . . . . . . . . . . . . . 13
1232.2
Computation of Levels of Precision for
Stratified Samples From State (Tribal)
Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1232.3
Relative Efficiency of Stratified Random
and Simple Random Sampling . . . . . . . . 16
1300
SAMPLING PLAN REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . .
1310
Criteria for Plan Approval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1320
Sample Frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1330
Sample Selection Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . .
-i-
17
17
18
19
1400
SAMPLE SIZES AND PROCEDURES FOR SELECTING SAMPLE
CASES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1410
Annual Sample Size Requirements . . . . . . . . . . . . . . . . . . . . . . . 21
1411
Sample Size Requirements for the TANF Active Sample
21
1412
Sample Size Requirements for the TANF Sample of
Closed Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1413
Sample Size Requirements for the SSP-MOE Active
Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1414
Sample Size Requirements for the SSP-MOE
Sample of Closed Cases . . . . . . . . . . . . . . . . . . . . . . . 24
1415
Adjustment to the Sample Size for States and
Tribes with Small Caseloads . . . . . . . . . . . . . . . . . . . 24
1416
Average Monthly Sample Size . . . . . . . . . . . . . . . . . . 25
1420
Sample Frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
1421
1430
1440
Sampling Frame for the TANF Active
Case Sample . . . . . . . . . . . . . . . . . . . . . . .
1422
The Treatment of Special Groups With Respect to
TANF Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1422.1
Newly Approved Applicant (a.k.a., Initial
Assistance Cases) . . . . . . . . . . . . . . . . . . .
1422.2
Non-Custodial Parents . . . . . . . . . . . . . . .
1422.3
Members of Indian Tribes Not Receiving
Assistance under a Tribal Family
Assistance Plan . . . . . . . . . . . . . . . . . . . . .
1422.4
Members of Indian Tribes Receiving
Assistance under a Tribal Family
Assistance Plan . . . . . . . . . . . . . . . . . . . . .
1422.5
Cases Selected For More Than One
Sample Month . . . . . . . . . . . . . . . . . . . . . .
1422.6
Cases Receiving Assistance Under the
State's TANF Program and Separate
State Programs for the Same Month . . . . .
1422.7
Cases With a Child Not Living With a
Parent or Adult Caretaker Relative . . . . . .
1423
Sample Frame for the Sample of Closed
TANF Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1424
Sample Frame for the Sample of Active SSP Cases .
1425
Sample Frame for the Sample of Closed SSP Cases .
Procedures for Selecting Sample Cases . . . . . . . . . . . . . . . . . . . .
Procedures for Selecting Sample Cases Using a Simple
Random Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
- ii -
27
29
29
29
30
30
30
30
31
31
32
32
32
36
1450
1500
1600
1700
Retention of Sampling Records . . . . . . . . . . . . . . . . . . . . . . . . . . 38
CORRECTION FOR UNDERSAMPLING AND EXCESSIVE
OVERSAMPLING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1510
Standard Method to Correction for Undersampling or
Oversampling when Sample Selected Using
Systematic Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1510.1
Correction for Oversampling . . . . . . . . . . . . . . . . . . .
1510.2
Correcting for Undersampling . . . . . . . . . . . . . . . . . .
1520
Alternate Method of Correcting for Undersampling or
Oversampling When Sample Selected Using Systematic
Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1530
Correcting for Undersampling Using a Reserve Sample Pool . . .
1531
Procedure for Setting Up a Reserve Sample Pool . . .
1532
Procedure for Obtaining Cases from a Reserve
Sample Pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1540
Correction for Undersampling or Oversampling When
Sample Was Selected Using Simple Random Sampling . . . . . . .
1540.1
Correcting for Undersampling . . . . . . . . . . . . . . . . . .
1540.2
Correcting for Oversampling . . . . . . . . . . . . . . . . . . .
WORK PARTICIPATION RATES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1610
Work Participation Rate Standards and Caseload
Reduction Credit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1620
Definitions of Annual and Monthly Work Participation Rates
...................................................
1630
Countable Work Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1640
Required Hours of Work to be "Engaged in Work" . . . . . . . . . .
1641
Deemed Engaged In Work . . . . . . . . . . . . . . . . . . . . .
1642
The Thirty (30) Percent Limit . . . . . . . . . . . . . . . . . .
1650
Waiver Inconsistencies as Related to the Work
Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1660
Tribal Work Participation Rates . . . . . . . . . . . . . . . . . . . . . . . . . .
1661
Calculation of the Monthly Work Participation Rate
from Universe Data . . . . . . . . . . . . . . . . . . . . . . . . . .
1662
Calculation of the Monthly Work Participation Rate
from Sample Data . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1663
Calculation of the Monthly Work Participation Rate
from Stratified Sample Data . . . . . . . . . . . . . . . . . . . .
1664
Adjusting the Monthly Work Participation Rate for
Exceeding the 30% Limit . . . . . . . . . . . . . . . . . . . . . .
41
42
42
43
44
46
47
49
49
50
51
53
53
56
58
60
61
61
61
63
65
67
70
74
STATISTICAL METHODS IN DATA ANALYSIS . . . . . . . . . . . . . . . . 77
- iii -
1710
1720
1730
Statistical Tests of Significance . . . . . . . . . . . . . . . . . . . . . . . . . . 77
1711
Testing the Representativeness of the Sample
with the Caseload . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
1711.1
Comparison of Sample and Total Caseload
When Proportions Are Not Used . . . . . . . 78
1711.2
Comparison of Sample and Total Caseload
When Proportions Are Used . . . . . . . . . . . 79
1711.3
One Sample Chi-Square ( Χ2 ) Test . . . . 80
1712
Testing Differences of Proportions Between Samples 85
1713
Testing Differences Within the Same Sample -- ChiSquare ( Χ2 ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
1721
Moving Averages . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
1722
Individual Monthly Rates . . . . . . . . . . . . . . . . . . . . . . 93
1723
Computation of a Regression Line by "Least Squares"
Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
1723.1
Practical Uses of Trend Line and Trend
Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
1723.2
Testing Trend for Statistical Significance 99
1723.3
Relationship Between Time Sequence and
Error Rates . . . . . . . . . . . . . . . . . . . . . . . 100
Statistical Procedures for Developing Profiles of
Error-Prone Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
1731
Criteria for Setting Up Error-Prone Profile Models . 103
INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
- iv -
FIGURES
Figure 1. - Normal Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Figure 2. - Six-Month Moving Averages of Completed Sample Cases
By the Month of Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Figure 3. - Six-Month Moving Averages of Completed Sample Cases
Regardless of Month of Review . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Figure 4. - Trend Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
APPENDICES
APPENDIX A - Table of Random Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
APPENDIX B - Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
APPENDIX C - Standard Error of Percentages Based on Selected Sample Sizes . . . . 137
APPENDIX D - TANF SAMPLE PLAN Guidance . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
APPENDIX E - TRIBAL CODES FOR THE TANF PROGRAM . . . . . . . . . . . . . . . 147
APPENDIX F - FIPS County Codes - Alphabetical List . . . . . . . . . . . . . . . . . . . . . . . 159
-v-
- vi -
TANF
SAMPLING AND STATISTICAL
METHODS MANUAL
1100
INTRODUCTION
Title I of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996
(PRWORA) establishes the Block Grants for Temporary Assistance for Needy Families
(TANF) Program by amending Titles IV-A and IV-F of the Social Security Act. The
purpose of this welfare reform initiative, which replaced the Aid to Families with
Dependent Children Program and the Jobs Opportunity and Basic Skills Program, is to
increase the flexibility of States and Tribal grantees in operating a program designed to:
1.
Provide assistance to needy families (cases) so that children may be cared
for in their own homes or in the homes of relatives;
2.
End the dependence of needy parents on government benefits by promoting
job preparation, work, and marriage;
3.
Prevent and reduce the incidence of out-of-wedlock pregnancies and
establish annual numerical goals for preventing and reducing the incidence
of these pregnancies; and
4.
Encourage the formation and maintenance of two-parent families.
While the TANF provisions allow States and Tribal grantees discretion as to the
mechanisms used in meeting these goals, they place on States and Tribal grantees a
responsibility for measuring, tracking, and reporting on their reform initiatives.
The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 requires
States and Tribes to collect on a monthly basis and report to the Secretary of the
Department of Health and Human Services (DHHS) on a quarterly basis a wide variety of
-1-
disaggregated case record information on the families receiving assistance, families no
longer receiving assistance, and families applying for assistance from programs funded
under the TANF program. State or Tribal grantee may comply with this requirement by
collecting and submitting case record information for its entire caseload or by collecting
and submitting the case record information for a portion of the caseload which is
obtained through the use of scientifically acceptable sampling methods.
1110
Purpose of the Manual
Sampling is the selection of a part of a whole for the purpose of drawing conclusions
about the population, or universe. It permits the administrator to cut costs; reduce
manpower requirements; gather vital information more quickly; obtain data not available
otherwise; obtain more comprehensive data; and, in some instances, actually increase
statistical accuracy. The manual explains statistical techniques in sufficient detail for
careful observance of sound sampling procedures and other basic statistical principles.
Theory is included to the extent necessary to provide working rules for application of the
more commonly used techniques as well as for recognizing the limitation of such
techniques. Because many users of the Manual are not statisticians, mathematical
exposition and technical language have been kept to a minimum.
This sampling manual contains the broad framework and procedures to be used by each
State or Tribes, that opts to file its TANF Data Report (or Tribal TANF Data Report)
based on a sample of its caseload. In developing its more specific sampling plans, States
and Tribal grantees have considerable latitude in designing samples that are consistent
with the principles described herein. The manual should provide the user with a basic
understanding of the TANF program sample requirements and statistically valid sampling
methods, which are essential to the successful reporting on the TANF program.
Section 1200 describes common types of sample designs (e.g., simple random sampling
and systematic random sampling) and basic statistical concepts, which are applicable in
any sample survey setting. It is intended to provide a general background to nonstatisticians who use the manual. Section 1300 contains sampling plan requirements: a
State or Tribal sampling plan must include a detailed description of the sample frame and
the procedures that are to be employed in constructing the sample frame, i.e., the list from
which the sample is to be selected. Also, the plan must describe in detail the sample
selection procedures for identifying the sample cases (families) for which data are to be
reported. For guidance on developing a sampling plan, see Appendix D. Section 1400
contains the sample size requirements, sample selection procedures for systematic
random sampling and simple random sampling, and special sampling problems
associated with the TANF program. Section 1500 describes procedures to be using in the
event an adjustment to the sample size is needed. Section 1600 describes the
methodology for calculating the monthly and annual work participation rates. Section
-2-
1700 contains general information on basic statistical techniques that can be used for an
effective analysis of the TANF program data. States and Tribes should use the sampling
plan requirements specified in Sections 1300, 1400, and 1500, along with the outline
contained in Appendix D (Page 141), to develop their detailed sampling plans. If they
need further assistance to develop sampling plans, they can contact the Administration
for Children and Families (ACF) Regional Administrator for assistance.
-3-
-4-
1200
BASIC STATISTICAL CONCEPTS
Probability sampling is an acceptable alternative to providing 100% counts of the TANF
caseload each month. Probability sampling has two properties: (1) every unit in the
entire population has a known, non-zero chance (called a probability) of being selected in
the sample, and (2) there is an element of "randomness" used to select the particular
sample for which data are to be collected. These two principles --measurability and
randomness -- distinguish probability samples from haphazard, judgment, or quota
samples.
1210
Sampling and Non-Sampling Errors
When a sample is selected through a random procedure, the estimates of a population
characteristic from that sample will generally be different from the true value of the
population characteristic simply because the estimates are based on a sample. Sampling
error is defined as the difference between the value of the characteristic as estimated from
the sample and the true population value of the characteristic. Although such errors
cannot be avoided, they can be controlled and measured (in probability samples).
Non-sampling errors, on the other hand, are generally not measurable (except by the use
of special auxiliary sample checks). Examples of non-sampling errors include: (1)
careless errors in coding responses, (2) errors attributable to the imperfect design of
measurement tools, e.g., I.Q. tests are only an approximate measure of intelligence, and
(3) errors due to inability to obtain relevant information for all sample members, i.e.,
non-response bias.
The design of any study should be examined carefully in order to determine the presence
and impact of such errors.
1220
Common Types of Scientific Sampling Techniques
It is impossible to specify a single sampling procedure that would be best suited to all
State agencies for all samples. There are many different ways of selecting scientific
(probability) samples from populations with items of equal importance. The simplest and
most widely used methods are: simple random sampling, systematic sampling, stratified
simple random sampling, and stratified systematic sampling. These four widely used
methods are acceptable methods of sampling for the purpose of collecting and reporting
-5-
the disaggregated TANF and separate State program - maintenance of errort (SSP-MOE)
data.
1221
Simple Random Sampling
Simple random sampling is a method of selecting a sample in such a way that each unit
of the frame has an equal and independent chance of being included in the sample. For
samples of any given size (n) from a population of size N, all possible combinations of n
units that could form samples of that size must have the same probability of selection. A
table of random numbers (see Appendix A, page 107) or a computer program with a
random number generator is generally used to choose the sample units. This method is
relatively easy to administer and is responsive to variations in caseload size over the
course of the sample period.
1222
Systematic Sampling
Systematic sampling method provides a system or pattern of selection of individual units
from a sample frame (which may be a hardcopy list or computer file of all the individual
units in the population) at equally spaced intervals (such as every 10th, 140th, 850th, etc.,
as required to obtain the total of a given sample size) with the starting point within the
first interval being determined by random selection.
In using the systematic sampling method, one needs to be aware of a major pitfall that
exists when the cases on the sample frame are arranged in some kind of repetitive or
cyclical pattern. In such an ordered list, the sample interval might sometimes be the same
as the cycle and could, therefore, yield a sample of cases with similar characteristics
which may not be typical of the caseload. It is, therefore, important not to use a
systematic sample with a listing that is cyclical in nature.
1223
Stratified Random Sampling
Stratified random sampling is random sampling of a population that is divided into a
number of sub-populations according to some pre-determined criterion (geographic
location, characteristic, etc.). In order to produce estimates with a given precision while
minimizing the total sample size required, the population is divided into several
homogeneous groups so that the units in the same group are more alike than the units in
different groups. Each group is called a "stratum" and the process of dividing the
population into groups is referred to as "stratification." The strata do not overlap and
together comprise the entire population. Sample cases can be selected independently
from each stratum using either systematic sampling, simple random sampling, or an
-6-
alternative approved sampling procedure. If the percent of the sub-population selected
from each sub-population are equal, i.e., proportional sampling, no weighting is required.
The sample is "self-weighting." Otherwise, individual weighting factors for each subpopulation must be taken into account before the sub-population sample results can be
combined.
There are various purposes for stratification. It may be that information is desired on the
strata separately; that more accurate estimates of the population parameters are needed
than can be obtained by a non-stratified sample; or that costs and administrative
constraints must be considered. To achieve these purposes, optimum allocation of the
sample size among the strata is usually required. Because a disproportionate number of
cases can be drawn from particular strata, some strata may be sampled more intensively
than others. For example, a State may find it administratively efficient to give a higher
probability of being sampled to urban areas than to rural areas.
The following points should be considered in using the stratified sampling method:
5.
Stratified sampling requires advance knowledge of the proportion of the
population in each stratum;
6.
Stratification by one characteristic does not ensure an efficient stratification
by other characteristics that may be of interest;
7.
Gains in precision for population estimates will be negligible unless it is
known that there are substantial differences among the strata and relatively
small differences within each stratum;
8.
The cost and effort of creating the strata may outweigh the potential gains
in precision;
9.
The weighting procedures required for calculating population estimates and
confidence levels for stratified samples in which the strata units are
disproportionately allocated can be complex and time consuming (see
Section 1232.2 (page 14); and
10.
Over stratification (i.e., creating too many strata) for a given size of sample
can result in some small strata that may adversely affect the precision of
estimates.
-7-
1224
Allocation of Stratified Sample
If a State selects a stratified sample, the State must decide how to allocate the sample
among the strata and describe the allocation procedures in the sampling plan. Two
common methods for sample allocation are allocation proportional to stratum caseload
size and optimal allocation with respect to an important program characteristic (e.g.,
participation rate).
1224.1
Proportional Allocation
Proportional allocation means that the size of the samples from the different strata are
proportional to the size of the caseload for the strata. In general, this allocation method is
desirable because it produces a self-weighting sample. For proportional allocation
calculate the stratum sample size by multiplying the total sample size by the ratio of the
stratum's caseload to the total caseload.
Nh
nh ' n
1224.2
N
Optimal Allocation
Optimal allocation of a given size sample means that the sizes of the samples from the
different strata are determined so that the overall variance is minimized. This is done by
taking into consideration several characteristics, e.g., caseload size as well as the
estimated standard error for the value of the program characteristic of interest to the
program administrator. Because strata differ in both caseload size and the program
characteristic, it is reasonable to take larger samples from the strata with greater value of
the program characteristic of interest and smaller samples from the strata with less value
of the program characteristic of interest. Optimal allocation with respect to the program
characteristic produces a disproportionate stratified sample that minimizes the estimated
standard error of the program characteristic. Because the sample is disproportionally
allocated, the sample results will have to be weighted to generate State program
characteristics. The equation for the optimal allocation of a sample is:
nh ' n
NhSh
' (NhSh)
H
h'1
-8-
where:
nh
is the sample size for the h th stratum;
n ' n1 % n2 % . . . % nh is the total State sample size;
h ' 1, 2, . . ., H
1230
represents the H strata, in which the State's
caseload is grouped for sampling;
Sh
is the estimated standard error of program characteristic for the h th
stratum; and
Nh
is the TANF caseload for the h th stratum.
Validity and Reliability of Statistical Data
Sampling and statistical procedures, by themselves, cannot assure validity (or freedom
from bias) of the collected data -- that is, that case record information is actually correct
and is reported correctly. The validity of the statistical data depends upon the adequacy
of the coding schedule in relation to the scope, detail, and significance of the data
collected; the accuracy and completeness of the data in the case record; and the degree to
which case record reviews are carried out effectively.
Sound sampling procedures can assure a known degree of reliability (also referred to as
precision) of statistical data. If sampling procedures are soundly based, the results
obtained from one sample taken from the total caseload will be the approximate results
obtained if the whole caseload was reviewed.
The TANF sample is designed so that the reliability of the sample results is measurable
and can be shown to be relatively high. These results can be made more reliable through
proper application of statistical methods, as well as through an increase in sample size.
Because of their importance, examples of sources of bias (which affect validity) and
explanations of the formulas involved in measuring precision (reliability) are discussed in
some detail.
1231
Bias
A biased sample is one that does not represent the population from which it was selected,
i.e., an infinite number of selected samples would not yield the characteristics of the
-9-
population from which they were selected. For example, suppose that an opinion survey
was conducted in the middle of the day by interviewing everyone on a busy street willing
to stop for ten minutes for the interview. If 90 percent of those persons interviewed had a
favorable opinion on the issue involved, it would not necessarily follow that about 90
percent of the city residents have a favorable opinion. People on a particular street at a
particular time of day would more than likely be unrepresentative of the total city
population. Also, the fact that the sample consisted only of individuals who could spare
ten minutes in the middle of the day makes the sample even more unrepresentative. Such
a sample could contain bias.
One source of bias deals with cases for which data cannot be collected. "Data not
collected" or non-response cases fall into several categories. Such cases should have
been included in the sample but could not be for reasons such as the case record could not
be located or contains incomplete information.
If the number of non-response cases is small, the bias resulting from their non-response
will generally also be small. If the number of such cases is large, a considerable bias may
be introduced. In effect, a segment of the total caseload is unrepresented if the sample
cases for that segment are not reviewed. If a substantial number of sample cases are not
included, there is no assurance that conclusions drawn from the sample apply to the total
caseload. The number of such cases can be anticipated and should be compensated for
by oversampling. Even if the correct number of cases is compensated by oversampling,
non-response bias may still be present.
1232
Precision -- Computation of the Confidence Interval
Population values, which can normally be estimated from a sample, are often referred to
as population "parameters." A single valued estimate of a population parameter is called
a "point estimate." In order to predict the actual proportion of the population with a
given caseload characteristic (i.e. the proportion of the caseload with an adult
participating in a work program) with any degree of certainty, a range of possible values
(confidence interval) is computed. The first step is to compute the "variance" (also called
the "mean square deviation") of the point estimate. Variance is the quantity that is used
to measure the extent of fluctuations around the mean (simple average) while mean
square deviation is used to measure the dispersion around the mean or some arbitrary
origin.
- 10 -
For systematic samples, when simulating simple random selection, the estimated variance
of a proportion is computed approximately by the following equation 1:
Vp '
where:
p (1 & p)
N &n
×
n
N
p
=
estimated proportion (for item being estimated) in the sample, and
n
=
sample size
The precision of a sample estimate is measured by the standard error of the
estimate, Sp , which is the square root of the variance. The standard error, like the
variance, is normally unknown, and can be estimated from the sample.
Sp ' Vp
or
p (1 & p)
N&n
×
n
N
If n is small relative to N , then (N & n) / N can be ignored.
The precision specification consists of two elements. First, the administrative decision on
the desired degree of reliability determines the sample size necessary to meet the
specified probability level and precision range. For example, the administrator might
specify that the estimate of the proportion of two-parent families in the caseload is to be
within 1 percentage point of the figure that would be obtained by a complete review of
the entire caseload. This is called the tolerance specification or limit.
Secondly, since the administrator is dealing with a sample, a certain degree of risk is also
assumed. Thus, in the example given above, if the sampling error had been computed so
that the estimate plus or minus 1 percent would include the true value in 95 out of 100
samples selected from the same population, the estimate plus or minus 1 percent would
be called the 95 percent confidence interval.
1/ It can be shown that, if the units are randomly ordered, the variance of a systematic
sample is equivalent to the variance of a simple random sample.
- 11 -
In general, the 95 percent confidence interval is equal to the point estimate plus or minus
1.96 times the standard error of the normal distribution (or its approximation) and is
expressed as follows:
95% CI ' p ± 1.96 Sp
This confidence interval will cover the true value of "p" about 95 percent of the time
when sampling repetitively. Expressed in another way, we can be reasonably confident
that about 95 percent of the sample proportions will be within 1.96 standard errors of
their corresponding population proportion. A visual representation of this statement is
shown in the following figure. (The standard normal deviate, 1.96, is associated with the
exact 95 percent confidence interval. In practice, however, 2 is sometimes conveniently
used to replace 1.96 for constructing a 95 percent confidence interval. The actual
probability is 95.46 percent if 2, instead of 1.96, is used.)
Figure 1.
Normal Distribution
If " p " is the sample proportion, then there is a 95 percent probability that the
population value lies between p & 2 Sp and p % 2 Sp . Thus, the population value is
within 2 standard errors. (If 99.7 percent confidence was desired, the appropriate
- 12 -
universe value would be within 3 standard errors.) This is called two-tailed
probability and is used when interest is in both the upper and lower limits of an
estimate.
If however, only one limit is of interest, a one-tailed limit can be used. The
standard error (SE) units and probabilities are different for one-tailed limits. The
95 percent confidence interval for the one-tailed lower limit is p & 1.65 Sp . If p
represents the sample estimate of the participation rate, there is a 95 percent
probability that the true participation rate is greater than p & 1.65 Sp . Similarly,
the 95 percent confidence interval for the one-tailed upper limit is p % 1.65 Sp .
There is a 95 percent probability that the true participation rate is less than
p % 1.65 Sp .
1232.1
Computation of Sample Size to Obtain a Desired Precision
By algebraic rearrangement, it is possible to compute the minimum sample size
needed to obtain a desired precision. For example, to obtain the sample size
required for 95 percent confidence, that a sample proportion " p " will be within
plus or minus 2 percent of the true proportion " P " when " p " is assumed to be 50
percent. The computation is as follows:
e ' 1.96
p (1 & p)
n
(1.96)2 p (1 & p)
or n '
e2
where " e " is the desired precision level (2 percent in this example).
Substituting:
n '
(1.96)2 (.50) (1&.50)
(.02)2
- 13 -
n
=
2,401 or approximately 2,400 cases 2
It should be noted that, for a proportion, precision is primarily a function of sample
size. Larger samples will generally yield more precise estimates. In many cases,
the size of the population from which the sample is drawn is not important. As the
population size increases, and the ratio (N & n)/N approaches 1.00 (where " N " is
the population size and " n " is the sample size), the effect of population size on
precision diminishes and can usually be disregarded.
The specification of precision and confidence are both administrative decisions that
are generally the responsibility of those who will use the data. The uncertainty
associated with sampling can be reduced by taking larger samples or using superior
measurement techniques, but only at some expense. Therefore, these decisions
also must take account of the resources available to collect the sample data.
1232.2
Computation of Levels of Precision for Stratified Samples From State
(Tribal) Sample
In a stratified sample, population and variance estimates are computed from
information in each stratum or group, appropriately weighted and combined.
Precision for Proportions
If in each stratum ( h ) a systematic sample (approximating a simple random
sample) is selected, the equations for estimating the overall proportion (ps) and its
variance are as follows:
2/
The 2,400 figure is based on the assumption that the population rate is 50 percent and
that the sample is a small fraction of the caseload so that the finite population factor
can be ignored. If the same fraction is large, the finite population factor should be
included; the sample size can be modified using the equation
n1 '
n
1 % (n & 1) / N
where N is the population size.
- 14 -
ps ' j Whph ' j
H
Nh
h'1
N
ph
and
3
Vp
s
' j
H
Nh
h'1
N
2
ph (1 & ph)
nh
where:
H
= number of strata;
Nh
= population size in stratum h ;
N
= 'Nh = total population size;
Wh
=
nh
= sample size in stratum h ;
n
= 'nh = total sample size of all strata; and
ph
= proportion in stratum h .
Nh
N
= stratum weight;
The standard error of ph is estimated by the square root of its estimated variance
and, as mentioned earlier, is used in the calculation of confidence intervals. These
intervals are calculated in the same manner as for a non-stratified sample.
For example, assume a sample is drawn from three strata. The population sizes in
each stratum are 1,000, 2,000, and 4,000; the sample sizes are 50, 200, and 200;
and the stratum proportions are .05, 0.1, and 0.2 respectively. The overall
proportion is estimated as:
3/ If the finite population factor is included, the equation is given as follows:
Vp ' j
s
H
Nh
h'1
N
2
Nh & nh
ph (1 & ph)
Nh & 1
nh
- 15 -
Ps '
1000
7000
.05 x
2000
7000
0.1 x
%
%
0.2 x
4000
7000
' .15
and the variance of the proportion is:
Vp
s
'
1000
7000
2
(.05)(.95)
50
x
4000
7000
%
2
x
%
(0.2)(.08)
200
2000
7000
2
x
(0.1)(0.9)
200
' 0.00032
The standard error of the proportion is:
Sp
s
'
Vp
s
'
.00032 ' .018
The 95 percent confidence interval of the proportion is:
CI ' ps ± 1.96 Sp
1232.3
s
' .15 % 1.96(.018), or from .115 to .185
Relative Efficiency of Stratified Random and Simple Random Sampling
The frequently adopted definition of relative efficiency ( E ) of an estimator having
a variance, for example, of V1 to another having a variance of V2 is:
E ' V2 / V 1
Thus, the smaller the variance of an estimator, the more efficient the estimator. If a
State proposes to change its sample design, e.g., from a systematic sample to a
stratified random sample, it should check to see if the estimator (for a variety of
characteristics being measured) based on the new sample design has a variance that
is equal or smaller than that of the present sample design.
- 16 -
1300
SAMPLING PLAN REQUIREMENTS
The sampling plan serves as the foundation for the Administration for Children and
Families (ACF) review of the integrity of the State agency's and Tribal grantee's TANF
sampling procedures and SSP-MOE sampling procedures. The State or Tribe that elects
to submit case record information for a sample of families (also known as, cases) must
select its TANF sample (and, if applicable, SSP-MOE sample) for data reporting
purposes under a sampling plan approved by the ACF Regional Administrator. All
sampling procedures used by the State agency or Tribal grantee, including frame
composition and construction, must be fully documented and available for review by the
ACF Regional Office. This requirement includes all data processing specifications and
automated routines used to select the samples.
The sampling plan documentation must describe the list(s) of families from which the
samples are selected, the sample selection procedures, and the methodology for
estimating caseload characteristics and sampling errors. Referencing sub-sections of this
manual in the sampling plan does not constitute acceptable compliance with the
requirements set forth for sampling plan documentation without further explication of the
specific procedures the State or the Tribe will use. Detailed descriptions of the sample
frames, sample selection, and estimation procedures used by the State or Tribe must be
included in the sampling plan documentation.
If a State or Tribe opts to report the required case record information for a sample of
families (as opposed to for the entire caseload), a State shall have an approved sampling
plan in effect for a full sample period. A State or Tribe may not implement a new sample
design without prior approval. A revised sampling plan must be submitted to the ACF
Regional Administrator with specific documentation of any substantive modification of a
previously approved sample design at least 60 days before the start of the annual sample
period, i.e., no later than August 1. The State is not required to resubmit the sampling
plan if it is unchanged from the previous year. Changes in random start numbers, sample
intervals, or caseload estimates are not to be submitted as a revision of the sampling plan.
They should, however, be sent to the ACF Regional Office.
1310
Criteria for Plan Approval
The sampling plan must meet the following criteria:
- 17 -
11.
Conformance to principles of probability sampling, i.e., each case
(family) in the population must have a known, non-zero probability
of selection and computational methods of estimation must lead to a
unique estimate;
12.
Documentation of methods for constructing and maintaining the sample
frame(s), including assessment of frame completeness and any potential
problems associated with using the sample frame(s);
13.
Documentation of methods for selecting the sample cases from the sample
frame(s); and
14.
Documentation of methods for estimating case characteristics and their
sampling errors, including the computation of weights, where appropriate.
1320
Sample Frame
Samples are selected from a list of families called a "sample frame." The sampling plan
must describe in detail the master file, the payroll file, or other list(s) from which the
sample of families is actually selected. The plan must explicitly describe the following
sample frame characteristics:
1.
Date(s) when the sample cases (both regular and supplemental, if
applicable) for the sample month are selected, e.g., first workday of the
month following the sample month);
2.
Source, components, accuracy, and completeness of the sample frame in
relation to the total caseload; if not accurate or complete, explanation of
why not and how State (Tribe) plans to correct for the problems with the
sample frame;
3.
Procedures for ensuring that the sample frame contains complete coverage
of the applicable caseload (e.g., the active TANF sample frame includes all
families receiving assistance under the State's or Tribe's TANF Program,
including all newly approved applicants for the sample month, and the
closed TANF sample frame includes all families no longer receiving
assistance under the State's TANF Program, i.e., assistance terminated
effective for the sample month);
4.
Whether or not the frame is constructed by combining more than one list (if
more than one list, explanation of how lists are identified and how
duplication of cases on lists are prevented);
- 18 -
5.
Whether the frame is compiled entirely in the State office, entirely in local
offices, in the State office based on information supplied by local offices,
etc.;
6.
Form of the frame, e.g., a computer file, microfilm, or hard copy; if parts of
the frame are in different forms, specifications for each part;
7.
Frequency and length of delays and method used in updating the frame or
its sources;
8.
Procedures for estimating the proportion of sample cases for which the
State (Tribe) will not be able to collect and report case record information
(e.g., dropped as "listed-in-error" because the family (case) did not receive
TANF assistance for the reporting month);
9.
Methods of locating and deleting "listed-in-error" cases from the frame;
10.
Structure of the frame, i.e., the order of cases within each list and the data
elements on the frame, including definitions of coded values;
11.
Treatment of special populations under TANF (e.g., individuals under a
tribal family assistance plan, a non-custodial parent who participates in
work activities); and
12.
Criteria for stratifying sample (if applicable).
1330
Sample Selection Procedures
The sampling plan must describe in detail the procedures for selecting the sample cases.
The plan must explicitly describe the following characteristics:
1.
Procedures for estimation of caseload size, if applicable to sampling
method;
2.
Procedures for determination of an appropriate allowance for sample cases
for which the review may not be complete because the sample case was
"listed-in-error" (e.g., family did not receive TANF assistance for the
sample month);
3.
Procedures for determining the required monthly sample size and indication
of the sample size;
- 19 -
4.
If stratified sample design is used, procedures for sample allocation;
5.
Procedures for the computation of sample intervals and the determination
of random starts if the State (Tribe) used systematic sampling or stratified
systematic sampling;
6.
Application of selection procedures to identify sample cases;
7.
Procedures to compensate for excessive oversampling or undersampling;
and
8.
Time schedule for each step in the sampling procedure.
- 20 -
1400
SAMPLE SIZES AND PROCEDURES FOR SELECTING
SAMPLE CASES
1410
Annual Sample Size Requirements
State agencies and Tribal grantees should consider their own management information
needs relative to desired reliability of characteristic data broken out for specific
groupings, geographic areas, or by monthly or quarterly time periods in deriving the
TANF and the SSP-MOE sample sizes. While this section of the manual specifies the
minimum required annual sample sizes, States and Tribes are encouraged to select larger
size samples in order to increase the precision of the resulting estimates and to meet their
own information needs.
For TANF data collection and reporting purposes, there are two sampling frames from
which cases are to be sampled. The sampling frames are for families receiving assistance
(i.e., active cases, including all newly approved applicants) and families no longer
receiving assistance (i.e., closed cases).
If a State has one or more SSP-MOE, it must collect and report a limited amount of data
on TANF families receiving assistance, as defined in Appendix B (page 119) and no
longer receiving assistance under the SSP-MOEs if it wishes to: receive a high
performance bonus or qualify for work participation caseload reduction credit (see
section 261.40 through 261.44 of the final rule). The SSP-MOE data reporting
requirements are not applicable to Tribes. For the SSP-MOE data collection and
reporting purposes, there are two sampling frames from which cases are to be sampled.
The sampling frames are for families receiving assistance (i.e., active SSP-MOE cases,
including all newly approved applicants) and families no longer receiving assistance (i.e.,
closed SSP-MOE cases).
1411
Sample Size Requirements for the TANF Active Sample
The minimum required annual sample size for the active TANF sample is 3000
completed cases, of which approximately 2400 are ongoing cases and 600 are newly
approved applicants. Of the 2400 ongoing cases, approximately 600 cases are two-parent
TANF families. The minimum required sample sizes are designed to provide reasonably
precise estimates for such proportions as the work participation rates for all families (e.g.,
a precision of about plus or minus 2 percentage points at a 95% confidence level) and for
- 21 -
two-parent families (e.g., a precision of about plus or minus 2.3 percentage points at a
95% confidence level), as well as for demographic and case characteristics of newly
approved TANF families and all TANF families. In addition, these sample sizes will
permit us to detect real changes in certain proportions over time (e.g., changes in the
proportion of child-only cases).
The midpoint estimate (from which the confidence limits are constructed) of the overall
and two-parent work participation rates will be used in determining if States have met the
statutory requirements. If the State is unwilling to accept the precision levels obtained
from the minimum required annual sample sizes for the purpose of assessing penalties for
failing to met the work participation rates, it is the State's responsibility to increase its
sample size to what the State determines is an acceptable level of precision for this
purpose.
To meet these sample size requirements, States and Tribes may select one of the
following options:
9.
Use a simple or systematic sampling methodology (or other acceptable
method) and use an overall sample size that is sufficiently large enough to
obtain the 600 cases needed to meet the two-parent family required sample
size, the 600 required to meet the newly approved application sample size,
and the 3000 cases required to meet the overall sample size.
10.
Stratify the sample by newly approved applications; two-parent families;
and all other families. Uuse a random sampling method within each
stratum to select the sample. Taking into consideration the fact that twoparent families are included in the calculation of the all family work
participation rate, compute the sample size for each stratum based on the
600-case requirement for the two-parent stratum and 600 for the newly
approved applications stratum, and 1800 cases for the remaining families.
Under option 2, each stratum is sampled separately, and the monthly all families work
participation rate is a weighted rate, reflecting the representation of two-parent families
and other families with at least one adult or a minor child head-of-household to the total
all family population. If a State or Tribe uses a stratified sample design, the State (or
Tribe) must submit the monthly caseload for each stratum. These monthly caseload sizes
by stratum are due 45 days after the close of each quarter (i.e., the same due dates as for
the quarterly TANF Data Report).
If a State or Tribe does not have enough newly approved applicants or two-parent
families to meet the required annual sample sizes of 600 families (i.e., the average
monthly sample size of approximately 50 newly approved applicant families or 50 twoparent families), the State or Tribe must select 100% of such families and select from the
- 22 -
other ongoing stratum enough additional cases to meet the overall required annual sample
size of 3000 families. If a State or Tribe does not have enough families to meet the
overall sample requirement (i.e., 3000 families for the active TANF sample for an
average monthly sample of 250 families), the State or Tribe must report on 100% of their
families each month.
States and Tribes are not limited to these two methods for meeting the sample size
requirements. However, alternative methods should be discussed with Regional
statistical staff to ensure the reliability of the work participation rates and any other
statistic used to award a bonus or assess a penalty is not severely affected.
1412
Sample Size Requirements for the TANF Sample of Closed Cases
The minimum required annual sample size for the sample of closed cases is 800 cases.
An 800-case sample will permit us to obtain a precision of plus or minus 3.5 percentage
points for an attribute of 0.50 at a 95% confidence level. This result is obtained from the
formula in Section 1232.1 of this manual.
If a State or Tribe does not have enough closed cases to meet the required minimum
annual sample size of 800 families (i.e., an average monthly sample size of
approximately 67 families), the State or Tribe must collect data for and report on 100%
of the closed cases.
1413
Sample Size Requirements for the SSP-MOE Active Sample
The minimum required annual sample size for the active SSP-MOE sample is 3000 cases,
of which approximately 2400 are ongoing cases and 600 are newly approved applicants.
Of the 2400 ongoing SSP-MOE cases approximately 600 cases are two-parent families.
The minimum required annual sample sizes are designed to provide reasonably precise
estimates for such proportions as the work participation rates for all families (e.g., a
precision of about plus or minus 2 percentage points at a 95% confidence level) and for
two-parent families (e.g., a precision of about plus or minus 2.3 percentage points at a
95% confidence level), as well as for demographic and case characteristics of State SSPMOE families. In addition, these sample sizes will permit us to detect real changes in
certain proportions over time (e.g., changes in the proportion of child-only cases).
If a State does not have enough newly approved applicants or two-parent families to meet
the required annual sample size of 600 newly approved applicant families and 600 twoparent families (i.e., the average monthly sample size of approximately 50 newly
approved applicant families and 50 two-parent families respectively), the State must
select 100% of such families and select from the other ongoing stratum enough additional
- 23 -
cases to meet the overall required annual sample size of 3000 families. If a State does not
have enough families to meet the overall sample requirement (i.e., 3000 families for the
active SSP-MOE sample for an average monthly sample of 250 families), the State must
collect data for and report on 100% of its families.
1414
Sample Size Requirements for the SSP-MOE Sample of Closed Cases
The minimum required annual sample size for the SSP-MOE sample of closed cases is
800 cases. An 800-case sample will permit us to obtain a precision of plus or minus 3.5
percentage points for an attribute of 0.50 at a 95% confidence level. This result is
obtained from the formula in Section 1232.1 of this manual.
If a State does not have enough closed cases to meet the required annual SSP-MOE
sample size of 800 families (i.e., an average monthly sample size of approximately 67
families), the State must collect data for and report on 100% of the closed cases.
1415
Adjustment to the Sample Size for States and Tribes with Small Caseloads
If a State or Tribe has a small average monthly caseload, it may use the following
procedures in applying the finite correction factor to adjust the minimum annual sample
size. The formula for obtaining an adjusted sample size using the finite correction factor
is:
n1 '
where
1.
n
1 % (n & 1) / N
N
=
Total number of case months for the annual sample period (i.e., the
average monthly caseload times twelve months)
n
=
minimum required annual sample (e.g., active case sample is 3000
cases and closed case sample is 800 cases)
Compute the estimated number of case months for the annual reporting
period.
For example, it a State or Tribe has an estimated average monthly active
TANF caseload of 1,000 cases, then the total number of case months is 12,000
case months (i.e., N = 1,000 cases per month times 12 months = 12,000 case
months).
- 24 -
2.
Use the above formula and round up to determine the adjusted overall sample
size requirement.
For our example, the adjusted overall minimum required active TANF sample
size would be:
n1 '
3.
3000
' 2400
1 % (3000 & 1) / 12000
In computing the adjusted minimum annual sample size for the State's or
Tribe's active TANF sample or the active State SSP-MOE sample, prorate the
overall adjusted sample size to determine the required number of two parent
families, the required number of newly approved applicants and the required
number of other ongoing cases.
For our example, the sample size requirement for two-parent families is 480 cases (i.e.,
600 times 2400 divided by 3000 ), for newly approved applicant families is 480 (i.e., 600
times 2400 divided by 3000) and for other ongoing cases is 1440 cases (i.e., 1800 times
2400 divided by 3000).
1416
Average Monthly Sample Size
A State agency or Tribal grantee must select approximately one-twelfth of its annual
sample size each sample month. The average monthly sample size is determined by
dividing the required annual sample size by 12 and rounding the result up to the nearest
whole number. For the active TANF sample and SSP sample, the average monthly
sample sizes are 250 cases, of which 50 are two-parent families, 50 are newly approved
applicants, and 150 are other ongoing cases. For TANF and SSP samples of closed
cases, the average monthly sample sizes are about 67 cases. The following additional
procedures apply to the TANF samples and to the State's SSP samples:
1.
State agencies and Tribal grantees should select additional cases (use the
rate for "listed-in-error" cases based on historical data or, if unknown, use
five percent) of each sample to compensate for cases that may be reported
as "listed-in-error";
2.
A State or Tribe may increase its sample size above the minimum (and we
encourage them to do so), but may not reduce its sample size below the
minimum; and
3.
A State or Tribe has the option of collecting and reporting data for the
entire TANF population and a State has the option of collecting and
- 25 -
reporting data for its entire SSP population. However, we encourage States
and Tribes to take advantage of their option to use sampling, when
appropriate. Sufficiently large samples can produce reasonably precise
estimates, while saving substantial administrative staff resources and funds.
1420
Sample Frame
Creating a frame or list of cases from which the monthly samples are to be selected and
determining the sample size are preliminary steps applicable to any probability sample
design. Careful study of the structure of the sample frame is always essential in
probability sampling, especially in systematic sampling. The choice of a frame depends
upon the criteria of timeliness, completeness, and administrative burden. The structure of
the sample frame should provide for an unduplicated list of cases comprising the target
population or otherwise allow for all units to have a known, non-zero chance of selection
into the sample. In systematic sampling, cases should be randomly ordered with respect
to the variables being measured, e.g., case characteristics data, earnings, participation in
work activities, etc. This random order is usually achieved if cases are arranged by case
number or by county and then alphabetically within county, or by any other file
organization that is not directly related to the measurement of critical variables. The
structure of the sampling frame must be fully documented in the sampling plan and may
not be changed without an approved revision of the sampling plan.
- 26 -
1421
Sampling Frame for the TANF Active Case Sample
The monthly TANF sample frame consists of all families who receive assistance under
the State (Tribal) TANF Program for the sample month by the end of the sample month.
The term "assistance", defined in §260.31 of the final rule, includes cash, payments,
vouchers, and other forms of benefits designed to meet a family's ongoing basic needs
(i.e., for food, clothing, shelter, utilities, household goods, personal care items, and
general incidental expenses). It includes such benefits even when they are provided in
the form of payments by a TANF agency, or other agency on its behalf, to individual
recipients and conditioned on their participation in work experience, community service,
or other work activities (i.e., under §261.30).
Except where excluded as indicated in the following paragraph, it also includes
supportive services such as transportation and child care provided to families who are not
employed.
The term "assistance" excludes:
1.
Nonrecurrent, short-term benefits (such as payments for rent deposits or
appliance repairs) that:
a.
Are designed to deal with a specific crisis situation or episode of
need;
b.
Are not intended to meet recurrent or ongoing needs; and
c.
Will not extend beyond four months.
2.
Work subsidies (i.e., payments to employers or third parties to help cover
the costs of employee wages, benefits, supervision, and training);
3.
Supportive services such as child care and transportation provided to
families who are employed;
4.
Refundable earned income tax credits;
5.
Contributions to, and distributions from, Individual Development
Accounts;
6.
Services such as counseling, case management, peer support, child care
information and referral, transitional services, job retention, job
advancement, and other employment-related services that do not provide
basic income support; and
- 27 -
7.
Transportation benefits provided under an Access to Jobs or Reverse
Commute project, pursuant to section 404(k) of the Act, to an individual
who is not otherwise receiving assistance.
The exclusion of nonrecurrent, short-term benefits under (1) of this paragraph also covers
supportive services for recently employed families, for temporary periods of
unemployment, in order to enable continuity in their service arrangements.
The TANF active case sample frame could be a master file; a payroll file; an eligibility,
activity, or other caseload file; or a combination of such files depending on how the State
or Tribe defines its range of benefits/assistance. If such a list cannot be constructed based
on the above definition of the sampling universe, it may be necessary to use a special
procedure to ensure that all families receiving assistance have a known, non-zero chance
of being included in the sample. The sampling plan should contain the State or Tribe's
objective criteria for the delivery of assistance and determination of eligibility as set forth
in the State or Tribe's family assistance plan. State agencies and Tribal grantees should
verify the receipt of assistance for all selected cases, and all such cases discovered not to
have received assistance for the reporting month should be reported as "listed-in-error."
For all other cases selected into the sample, the data collection must be completed and the
data must be submitted to ACF by the specified time frames.
States or Tribes that use regular first-of-the-month payroll or eligibility listings as the
frame for selection of sample cases must extend that frame at the end of the report month
and continue sampling all cases for which assistance was initiated during the report
month that were not on the first-of-the-month payroll/eligibility listing. Care must be
taken to ensure that the sample frame consists of unduplicated cases. A distinction is
made between cases already receiving TANF and cases in which assistance is initiated
during the month. For example, a case receiving a regular payment on October 1 and a
supplemental payment on October 12 should only be subject to selection once for the
month of October. Procedures for accomplishing this must be specified in the sampling
plan. Normally, this will be accomplished by running a computer sort/merge routine at
the end of the report month in order to establish the list of supplemental cases to be added
to the frame.
States or Tribes that use simple random sampling should form the sample frame at the
end of the sample month, ensuring all families that received assistance for the month by
the end of the month are on the sample frame. Then the sample is selected after the end
of the sample month.
- 28 -
1422
The Treatment of Special Groups With Respect to TANF Reporting
- 29 -
There are a number of family circumstances that merit special attention. These are
described below.
1422.1
Newly Approved Applicant (a.k.a., Initial Assistance Cases)
A newly-approved applicant or an "initial payment/assistance" case for a sample month
means the family is newly added to the TANF caseload and the current reporting month
is the first month in which the TANF family receives TANF assistance (and thus has had
a chance to be selected into the TANF sample). This may be either the first month that
the TANF family has ever received assistance or the first month of a new spell on
assistance. The initial payment/assistance case should be included on the sample frame
for the initial month in which it received assistance and for all subsequent months for
which assistance is issued. For States that provide assistance back to the date of
application, these cases may, at State option, be included on the frames for prior months,
as assistance was not received by the end of such months.
A family that moves back and forth between receipt of assistance to receipt of only nonassistance in a subsequent month while remaining in the TANF program will be a newly
approved applicant each time it moves to receipt of assistance for a reporting month.
1422.2
Non-Custodial Parents
A non-custodial parent is defined in §260.30 as a parent of a minor child who: (1) lives in
the State and (2) does not live in the same household as the minor child. The State must
report information on the non-custodial parent if the non-custodial parent: (1) is receiving
assistance as defined in §260.31; (2) is participating in work activities as defined in
section 407(d) of the Act; or (3) has been designated by the State as a member of a family
receiving assistance. In reporting non-custodial parents, States or Tribes should not treat
the non-custodial parent as a separate case. Rather, when the family unit containing
his/her child(ren) is selected into the sample, code the type and amount of assistance
received by the non-custodial parent as part of that case. The non-custodial parent's
person level data must also be provided. States and Tribes have the option to include or
exclude the non-custodial parent from the work participation rate on a case-by-case basis.
If an individual is both a custodial parent for a TANF family receiving assistance and a
non-custodial parent for another TANF family receiving assistance, the State or Tribe
should report the individual only with the family for which (s)he is the custodial parent.
- 30 -
1422.3
Members of Indian Tribes Not Receiving Assistance under a Tribal Family
Assistance Plan
The State sample frame must include each member of an Indian tribe otherwise meeting
the definition of the sampling unit who is domiciled in the State and is not receiving
assistance under a Tribal family assistance plan.
1422.4
Members of Indian Tribes Receiving Assistance under a Tribal Family
Assistance Plan
The State should not include members of an Indian tribe receiving assistance under a
Tribal family assistance plan, even if the State selected the option to include such
families in the calculation of its participation rate as provided for in section 407(b)(4) of
the Social Security Act.
1422.5
Cases Selected For More Than One Sample Month
If a family is selected into the sample for more than one month during the annual
reporting period, the State or Tribe should collect data for and report on the family for
each month for which it is selected.
1422.6
Cases Receiving Assistance Under the State's TANF Program and Separate
State Programs for the Same Month
A TANF eligible family may receive some form of assistance under both the State's
TANF Program and its SSP during the reporting month. If this occurs, the family should
be included on the active sample frame for both the TANF and the SSP. If such a family
is selected into the sample, the State should collect data for and report on the family for
each program for which it was selected.
- 31 -
1422.7
Cases With a Child Not Living With a Parent or Adult Caretaker Relative
Many activities are covered under section 401(a) of the Social Security Act (Act) (the
purposes of the TANF program). However, some activities are not permissible under the
purposes of the TANF program, but had been included in a State's approved AFDC plan,
JOBS plan, or Supportive Services plan as of 9/30/95, or at State option, 8/21/96.
Section 404(a)(2) "grandfathers in" States whose prior programs had such expenditures.
Thus, this section allows States to use Federal TANF funds for specific activities that had
been previously authorized based on an approved plan, using the same financial
eligibility criteria contained in the approved prior plan. Examples of such activities are
juvenile justice and foster care activities that were included in some States' approved
plan.
The legislative history makes it clear that the State may elect to continue to provide the
service or benefit under section 404(a)(2) of the Act, notwithstanding the prohibitions in
section 408 of the Act. For example, if a State's approved AFDC plan enabled it to
provide "assistance" or services to children in the juvenile justice system that does not
constitute TANF "assistance", then it may continue to use TANF funds for such activities
even though the child is not living with his parent or other adult caretaker relative.
Nonetheless, if the child is receiving "assistance" funded under the State TANF program,
the child is a child-only family for data collection and reporting purposes. For a State
that reports on its entire caseload, the State must collect data on and report data for all
such child-only families for each month that the families receive assistance. For the State
that reports its data for a sample of families, the State must include all such child-only
families on its monthly sample frame for each month that the family receives assistance.
If the child-only family is selected in a monthly sample, the State must collect data for
and report data on the child-only family for that month.
1423
Sample Frame for the Sample of Closed TANF Cases
For closed cases, the monthly TANF sample frame must consist of all families whose
assistance under the State TANF Program was terminated for the reporting month (do not
include families whose assistance was temporarily suspended), but received assistance
under the State's TANF Program in the prior month. A family that moves from receipt of
assistance under the TANF program for a month to receipt of benefits that are not
assistance under the TANF program for the subsequent month is a closed case for
reporting purposes. Also, a TANF eligible family that is transferred to the State's SSP or
Tribal TANF program is usually closed for the State TANF Program.
1424
Sample Frame for the Sample of Active SSP Cases
- 32 -
The monthly active SSP sample frame must consist of all families who receive assistance
under the separate State programs for the reporting month by the end of the reporting
month. The term "assistance" for separate State programs has the same meaning as for
TANF Programs. See Section 1421 (page 27) for the definition.
1425
Sample Frame for the Sample of Closed SSP Cases
For closed cases, the monthly SSP sample frame must consist of all families whose
"assistance" under the SSP was terminated for the reporting month (do not include
families whose assistance was temporarily suspended), but received assistance under the
SSP in the prior month. A family that is transferred to a State's TANF Program is usually
a closed case for the SSP.
1430
Procedures for Selecting Sample Cases
States and Tribal grantees have flexibility to choose from a wide variety of sampling
methods, including systematic sampling, simple random sampling, and stratified
(systematic or simple) random sampling. For illustrative purposes, the following
procedures are based on the systematic sampling design and, if used, are repeated each
month during the annual sample period. In illustrating the procedures, a State or Tribe
with an estimated average monthly active TANF caseload of 42,600 is used. These same
procedures could be used to select the sample of closed TANF cases or the sample of the
active and closed SSP cases. Note, these procedures could be used to select a sample
within each stratum for a stratified systematic sample design.
1.
Estimate Caseload Size
The TANF average caseload is an estimate of the average monthly number
of cases that will receive assistance for the forthcoming annual sample
period. The average caseload size should be estimated on the basis of past
caseload sizes and trends. Any known circumstances, such as policy
changes that would appreciably change caseload sizes, also should be taken
into account in making the estimate.
Since the average monthly caseload must be estimated before the beginning
of the annual sample period, unanticipated changes can result in the need
for adjusting the sample interval. Recognizing the difficulty of forecasting
caseloads over a 12-month period, States and Tribes should re-evaluate the
estimated caseload before the end of each quarterly reporting period. If the
caseload estimate is changed, a new sample interval for the 12-month
period and adjustments to the number of sample cases already selected may
- 33 -
be needed. The procedures in Section 1510 (page 42), or 1520 (page 44,)
can be used depending on whether the sample requires correction for
oversampling or undersampling. If no correction is required for the
remaining quarterly reporting period(s), using these procedures will result
in a self-weighting annual sample.
2.
Determine Sample Size
The minimum required annual sample sizes of completed cases are shown
in Section 1410 (page 21). In our illustration, the minimum sample size is
used.
n ' 3,000
An estimate of the percent of cases that may possibly be reported as listedin-error during the TANF data collection process will need to be made in
order to arrive at the required completed sample size. For example,
assuming that 5 percent of the selected cases will be reported as listed-inerror, the number of cases to be selected can be computed as follows:
3,000 ÷ (1 & .05) ' 3,158
3.
Establish Frame
As mentioned in Section 1421 (page 27), a listing of all TANF cases that
received assistance for the sample month by the end of the month
(including initial assistance cases and cases that are reinstated) comprises
the frame from which the sample is selected.
4.
Establish Average Monthly Sample Size
The average monthly sample size is obtained by dividing the sample size
for the sample period by the number of months in the period. In our
illustration, the average monthly sample size is 3,158 ÷ 12, which is
263.167 cases.
5.
Compute Sample Interval
The sample interval is obtained by dividing the estimated average caseload
in the annual sample period (Step 1) by the unrounded average monthly
sample size (Step 4). In our example, the sample interval is 42,600 ÷
- 34 -
263.167, or 161 (rounded down). This means that each month, data will be
collected for 1 out of every 161 TANF cases.
6.
Select Random Start Number
The random start number can be as large as the number of cases contained
in the sample interval and is used only to determine the first selected
sample case for each month's sample. Since the sample interval in our
example is 161, we must select a random start number between 001 and
161 (inclusive). Assume the number selected is 103.
7.
Select Monthly Sample
The sequential position of the first selected sample case on the frame is the
starting point for selection of all subsequent cases. (If the frame is in
several parts, it will be desirable to assemble the parts so that one
continuous list is created. Every " k th " case will then be selected from a list
in which all cases are present.)
It is important in selecting the monthly sample to apply the same sample
interval to the entire list of cases each month. This is an important part of
the sampling design and should not be violated in order to obtain a specific
number of cases each month.
In our illustration, if the sample interval was a whole number, the 103rd
case on the list would be selected and every 161st case thereafter, i.e.,
103rd, 264th, 425th, etc. In each of the remaining eleven months of the
sample period, assuming no adjustment in estimated caseload size is
necessary after the sixth month, new random starts would be obtained as
the first case of each month and then multiples of 161 added to obtain the
other cases for data collection and reporting.
It should be understood that the numbers selected for the sample cases
relate to specific cases; substitutions or approximations are not acceptable.
For example, only the 103rd case must be selected, not the 102nd, or 104th,
etc. Once the random start and sample interval are determined, the specific
cases to be selected are identified.
There are several methods of selecting sample cases when the sample
interval is not a whole number. In one method, the sample case to be
selected is determined by rounding the number obtained after the sample
interval is added to the previous sample interval. For example, since in our
illustration the sample interval number was actually 161.87 instead of 161,
- 35 -
the following sample cases would be selected from the sample frame of
eligible cases (assuming a random start number of 163):
Selected Cases
Selection Procedure
# 103
- random start case
# 265
103
+ 161.87
264.87 = 265
- random start case
- interval
- rounded
# 427
264.87
+ 161.87
426.74 = 427
- previous total
- interval
- rounded
# 589
426.74
+ 161.87
588.61 = 589
- previous total
- interval
- rounded
# 750
588.61
+ 161.87
750.48 = 750
- previous total
- interval
- rounded
etc.
etc.
For TANF purposes, an acceptable method for selecting sample cases when
the sample interval is not a whole number is to round down to the next
lower whole number and use that number in selecting the sample cases.
For example, using the same sample interval of 161.87 and random start
number of 103, the interval would be rounded down to 161 and the sample
cases selected would be the 103rd, the 261th, the 425th, the 586th, 747th,
etc.
- 36 -
8.
Submission of Caseload Size, Sample Interval and Sample Cases Selected
If a State or Tribe opts to use systematic sampling or stratified systematic
sampling, the State or Tribe should send the estimated average monthly
caseload and the computed sample interval(s) to be used for the 12-month
sample period to the ACF Regional Administrator thirty (30) calendar days
before the October sample selection.
If a State or Tribe uses a stratified sample design, it must submit the
monthly caseload sizes by stratum for each month of the quarter within 45
days after the end of the quarter. These data are needed for weighting
purposes.
Regardless of the method used to select the sample cases, each State and
Tribe that opts to collect data for and report on a sample of cases must
submit the monthly list of selected sample cases (including reserve pool
cases, if applicable, under Section 1531 page 47), within 10 days of the
date of selection specified in the State or Tribe sampling plan.
1440
Procedures for Selecting Sample Cases Using a Simple Random Sample
States and Tribal grantees may want to use simple random sampling or stratified simple
random sampling because there are a number of computer software packages that contain
programs that use this method of sampling. For illustrative purposes, the following
procedures are based on the simple random sampling design and, if used, are repeated
each month during the annual sample period. These same procedures could be used to
select the sample of closed TANF cases or the sample of the active and closed SSP cases.
Note, these procedures could be used to select a sample within each stratum for a
stratified simple random sample design.
1.
Establish the Monthly Sample Frame
As mentioned in Section 1421 (page 27), a listing of all TANF cases that
received assistance for the sample month by the end of the month
(including initial assistance cases and cases that are reinstated) comprises
the frame from which the sample is selected.
2.
Determine the Number of Families on the Sample Frame
Many automated simple random sampling routines need to know the
number of sampling units on the sample frame and the number of units to
be selected prior to execution of the sample selection routine. For the
- 37 -
TANF active sample, the sampling units are the families receiving TANF
assistance. If a stratified simple random sample is used, the State must
determine the number of families in each stratum for the sample month.
3.
Determine Sample Size
The minimum required annual sample sizes of completed cases are shown
in Section 1410 (page 21). In our illustration, the minimum sample size is
used.
n ' 3,000
An estimate of the percent of cases that may possibly be reported as "listedin-error" during the TANF data collection process will need to be made in
order to arrive at the required completed sample size. For example,
assuming that 5 percent of the selected cases will be reported as "listed-inerror", the number of cases to be selected can be computed as follows:
3,000 ÷ (1 & .05) ' 3,158
4.
Establish Average Monthly Sample Size
The average monthly sample size is obtained by dividing the sample size
for the sample period by the number of months in the period. In our
illustration, the average monthly sample size is 3,158 ÷ 12, which is
263.167 cases or 263 cases.
5.
Select Monthly Sample
The most practical way of selecting a sample of TANF cases using a simple
random sample is with the use of automated routines. These routines use a
random number generator to select n (the number of units to be selected)
out of N (the number of units on the sample frame). The n sample cases
should be selected without replacement. To illustrate using a monthly
sample frame with 42,600 families and a monthly sample size of 263
sample cases, the automated sampling routine would select 263 numbers
between 1 and 42,600 inclusive. If the random numbers generated include
20, 175, 183. 500, etc., then the 20th, 175th, 183rd, and 500th case on the
sample frame would be drawn into the sample.
- 38 -
6.
Submission of Caseload Size, Sample Interval, and Sample Cases Selected
If a State or Tribe uses a stratified sample design, it must submit the
monthly caseload sizes by stratum for each month of the quarter within 45
days after the end of the quarter. These data are needed for weighting
purposes. States and Tribes that use non-stratified sample designs report
their total monthly caseload numbers on the TANF Data Report - Section
Three. These figures are used to weight the State data.
Regardless of the method used to select the sample cases, each State and
Tribe that opts to collect data for and report on a sample of cases must
submit the monthly list of selected sample cases (including reserve pool
cases, if applicable) within 10 days of the date of selection specified in the
State or Tribe sampling plan.
1450
Retention of Sampling Records
Each State and Tribe shall retain all sampling records for an annual sample period for a
minimum of three years after the end of the sample period. These materials shall include
the
1.
original monthly sample frames from which the sample was selected;
2.
computer programs used to construct the sample frames and select the
sample cases;
3.
caseload estimate worksheets;
4.
sample intervals and random start numbers;
5.
sample size;
6.
lists of selected cases, including supplemental and reserve pool cases if
applicable; and
7.
audit trail tracking logs.
In addition, the State and Tribe shall retain the approved sampling plan until a revised
plan is approved and implemented. When the revised approved sampling plan is
implemented, the previously approved sampling plan should be retained for three years.
These materials are to be made available to the Regional staff upon request.
- 39 -
- 40 -
1500
CORRECTION FOR UNDERSAMPLING AND EXCESSIVE
OVERSAMPLING
When using systematic sampling, imprecise caseload projections or an unexpected drop
rate will result in the State or Tribe not obtaining its target sample size. If the actual
universe is larger than the estimated size, oversampling may occur. If the actual universe
is smaller than the estimated size, undersampling may occur. A State agency and Tribal
grantee must correct for undersampling to the extent necessary to meet sample size
requirements for TANF reporting and a State agency must correct for undersampling to
the extent necessary to meet sample size requirements for reporting of separate State
programs. A State agency or Tribal grantee has the option as to whether or not to correct
for excessive oversampling. However, we encourage States and Tribes to select larger
than the minimum required annual sample size in order to increase the precision of
statistics that are estimated from the sample data.
In correcting the TANF or SSP sample size, care must be taken to assure that the
statistical principles of "randomness" and measurability are not violated. The selection of
additional families for the TANF and SSP samples or deletion of units from the samples
must be done in a manner that assures all cases in the population have a known, non-zero
probability of selection into the final sample. In addition, techniques of stratification
should not be employed in such a way that small additional strata are created for which
computed estimates may be unreliable, resulting in a loss of precision in population
estimates.
The procedures that a State (Tribe) uses to correct for excessive oversampling or correct
for undersampling will depend partly on the procedures the State (Tribe) used to select its
original sample cases. States and Tribes may choose from a wide variety of sampling
methods. State agencies or Tribal grantees that select their TANF samples or State
agencies that select their SSP samples using the systematic sampling method can use the
procedures in Sections 1510, 1520, or 1530 of this manual to adjust sample sizes. State
agencies or Tribal grantees that select their TANF samples or State agencies that select
their SSP samples using the simple random sampling method can use the procedures in
Sections 1540 to adjust sample sizes. For State agencies or Tribal grantees that use
another method to select their TANF samples, ACF Regional Office staff will be happy
to provide technical guidance on procedures to correct for excessive oversampling or
undersampling to ensure that the principles of probability sampling are retained.
- 41 -
Monthly sample sizes should be monitored throughout the reporting period and
correction should be made only when it becomes clear that target samples will not be
met. It is good practice to re-estimate caseloads at the end of each quarterly reporting
period. Waiting to the end of the annual period to make necessary corrections could
create difficulties in collecting the information and adversely affect the State's (Tribe's)
ability to submit data in a timely manner.
The following procedures allow State agencies and Tribal grantees to make corrections in
all months starting with the first month of the reporting period. A consideration for a
State in selecting this method is that, in certain circumstances, it may be difficult to
obtain accurate information for past months. This method does not involve the creation
of additional strata.
1510
Standard Method to Correction for Undersampling or Oversampling when
Sample Selected Using Systematic Sampling
1510.1
Correction for Oversampling
8.
Using the procedure described in Section 1430, Step 1 (page 32),
re-estimate the caseload size, adding on the expected number of cases to be
dropped as listed-in-error, and compute a revised sample interval.
For each month in which the sample cases have already been selected:
9.
Divide the size of the monthly sample frame by the revised sample interval
(Step 1) to obtain the revised estimate of the number of sample cases that
should have been selected.
10.
Subtract the number of cases obtained in Step 2 from the number of sample
cases that have been selected. This is the number of sample cases to be
eliminated.
11.
Divide the number of sample cases that have been selected by the number
of cases to be eliminated (Step 3) to obtain the secondary sample interval to
be used in identifying the cases to be eliminated.
12.
Use a random start and apply the secondary sample interval obtained in
Step 4 to select cases from the list of sample cases already selected. The
cases so identified are to be eliminated regardless of whether or not data
had already been collected.
- 42 -
For months in the annual period for which sample cases have not yet been
selected:
13.
1510.2
Use the corrected sample interval for the period obtained in Step 1 to select
sample cases from the monthly frames.
Correcting for Undersampling
1.
Using the procedure described in Section 1430, Step 1, (page 32)
re-estimate the caseload size, adding on the expected number of cases to be
dropped as listed-in-error, and compute a revised sample interval.
For each month in which the sample cases have already been selected:
2.
Divide the size of the monthly sample frame by the revised sample interval
(Step 1) to obtain the revised estimate of the number of sample cases that
should have been selected.
3.
Subtract the number of sample cases already selected from the number
obtained in Step 2. This is the number of additional sample cases to be
selected from the monthly frame.
4.
Divide the total monthly sample frame size by the number identified in Step
3 to obtain the secondary sample interval to be used in selecting additional
cases from the monthly sample frame.
5.
Use a random start and apply the secondary sample interval obtained in
Step 4 to the monthly sample frame from which cases have already been
selected. (If correction for undersampling is required only for the third
and/or fourth quarters of the annual period, the State has the option of
applying the secondary interval either to the first month of the sample
period (October) or the first month of the applicable quarter (April or
July)). Add the specific cases identified to the cases already selected for
the same month as the month of the sample frame from which they were
selected. If a case previously selected in the sample is again selected and
identified for the same month as previously selected, an alternate case is to
be selected by using a table of random numbers.
For months in the annual period for which sample cases have not yet been
selected:
- 43 -
6.
1520
Use the corrected sample interval for the period obtained in Step 1 to select
sample cases from the monthly frames.
Alternate Method of Correcting for Undersampling or Oversampling When
Sample Selected Using Systematic Sampling
An alternate method involves no adjustment for the months for which cases were already
selected, however it does result in stratification of the sample by time. The alternative
method entails the computation of a new sample interval that will either (1) undersample
the remaining months of the 12-month sample period to meet sample size requirements if
the earlier months had been oversampled, or (2) oversample the remaining months of the
annual period to meet sample size requirements if the earlier months had been
undersampled.
Because two different sample intervals will have been used, results of cases selected by
each sample interval cannot be directly added to obtain State-wide (Tribal-wide)
estimates as the proportions of the monthly frames sampled are different, i.e., the total
sample is not a self-weighting sample. 4 The alternate method will require all data to be
weighted at the end of the 12-month period. The procedure involves inflating the various
frequencies (e.g., number of families with an adult working, the number of families with
a minor parent head of household, cases with earned income, etc.) in cases obtained using
each sample interval, to their representation in the caseload and dividing the result by the
caseload. 5 This gives the weighted rate for the State (Tribe). In order to make each of
the frequencies (number of families with an adults working, the number of families with
a minor parent head of household, etc.) comparable with those of other States (Tribes), it
is necessary to multiply the weighted rate by the total sample size. The equation for this
procedure is as follows:
4/
It should be noted that a self-weighting sample, except for rounding, must
possess the following characteristic:
Sample cases selected in a specific
month
Total sample cases selected in
sample period
5/
=
Cases in sample frame for same
month
Total cases in all sample frames in
sample period
"Caseload", for the purpose, is defined as the completed sample size multiplied by the
sample interval.
- 44 -
Weighted State Rate
'
'(xm)(SIm)
'(nm)(SIm)
where:
j
=
the sum of . . . all strata ("stratum" is defined as part of the annual
period using the same sample interval);
m
=
the m th stratum (m is the stratum index);
xm
=
"characteristic of interest" in the m th stratum;
nm
=
completed sample size in the m th stratum; and
SIm
=
sample interval used in the m th stratum.
For example, assume that a State originally had estimated that its caseload would average
80,000 cases for the annual sample period. Assuming a 5 percent drop rate, the State
used a sample interval of 303. Actual experience after 10 months resulted in the State
revising its average caseload to 75,000, making no change in its drop rate. If the State
made no corrections, the final completed sample size for the period would be short
approximately 188 cases.
Assume that the State decides to obtain the additional 188 cases by using a revised
sample interval of 219 for the last 2 months of the sample period. Also assume for the
first 10 months of the sample period (m ' 1) that the -Number of cases completed (n1) = 2,350
Number of cases with "characteristic of interest" (x1) = 112
and for the last 2 months of the sample period (m ' 2) that the -Number of cases completed (n2) = 650
Number of cases with "characteristic of interest" (x2) = 37
Using the definition of "caseload" as defined earlier, i.e., sample cases completed
multiplied by the sample interval, the weighted proportion of the case with the
characteristic of interest would be computed as follows:
- 45 -
=
(112 x 303) % (37 x 219)
'
(2,350 x 303) % (650 x 219)
=
.0492
42,039
854,400
The State case proportion for the "characteristic of interest" would be .0492. The
reported number of cases with the characteristic of interest for the 12-month period, for
comparability with other States, would be 148, i.e., .0492 x 3,000.
Note that each frequency of occurrence or proportion of the total sample must be
calculated in the same way, e.g., number of families with an adults working, the number
of families with a minor parent head of household, the number of child only cases, the
number of cases with earned income, etc. Caseload weights are to be used in computing
State-wide (Tribal-wide) characteristics.
Note, it is important that the appropriate code be entered on the coding schedule to
identify the stratum from which the case was selected.
1530
Correcting for Undersampling Using a Reserve Sample Pool
Correcting for undersampling using the sample interval (see Section 1520, page 44)
involves resampling the original frame using a new sample interval. A State (Tribe) may
find this to be difficult and/or costly. The same result can be achieved by selecting a
reserve sample pool at the time of original sample selection. The designated reserve
sample cases are to be used only if correction for undersampling is required. Properly
selected reserve pool cases retain the self-weighting property of the final sample.
However, careful attention to the controls is necessary to ensure that cases are properly
selected. Any number of cases may be designated as a reserve pool -- a good number
could be 10 or 15 percent of the required sample size.
The State (Tribal) sampling plan must describe in detail the procedures for setting up a
reserve sample pool. If a random number generator is used, the type of generator and
seed number is to be specified.
1531
Procedure for Setting Up a Reserve Sample Pool
Procedures for setting up a reserve sample pool are similar to those outlined in
Section 1430 (page 32). To illustrate the procedures, the example in Section 1430, is
used, i.e., a State (Tribe) uses the systematic sampling method, elects the standard sample
size, has an estimated average monthly caseload of 42,600 and estimates a 5 percent drop
- 46 -
rate for the 12-month sample period. In addition, the State (Tribe) specifies 15 percent of
its selected sample as reserve pool cases each month.
1.
Determine Average Monthly Sample Size
Divide the number of sample cases for which data is to be collected in the
annual sample period by (1 - 0.15) to obtain the estimated total number of
sample cases to be selected. In our example, according to Section 1430, Step
2, (page 33) the number of sample cases (completed and dropped cases) is
3,158. The number of cases to be selected would be 3158 ÷ (0.85), or 3,715,
or an average of 309 cases per month. The average number to be placed in a
reserve pool each month is 15% of 309 cases, or 46 cases (rounded down).
Note that the reserve pool is only to be used to correct for undersampling; it is
not to be used to replace dropped cases.
2.
Select Monthly Sample
Using the monthly sample size from Step 1, 309 cases, and the procedures
outlined in Section 1430, Steps 5, 6 and 7, (page 34) compute the sample
interval, determine a random start and select monthly sample cases from the
sample frame.
3.
Compute Secondary Interval for Selection of Reserve Pool Cases
Compute a secondary sample interval to be applied to the list of sample cases
selected each month. This is obtained by dividing the estimated average
monthly sample size by the average estimated number of cases designated for
the reserve pool. In our illustration, the sample interval is 309 ÷ 46, or 6.72.
4.
Select and Identify Monthly Reserve Pool Cases
Since the interval obtained in Step 3 above is not a whole number, the
acceptable method is to round up to the next higher number. (Note that
rounding up is recommended to ensure that the basic sample will have a
sufficient number of cases.) In our example, 1 out of every 7 cases on the
monthly list of selected sample cases would be identified for the reserve pool,
using a random start number between 1 and 7 inclusive. It is important in
selecting monthly reserve pool cases to apply the same sample interval to the
entire list of selected cases each month. This is an important part of the
sample design and should not be violated in order to obtain a specific number
of reserve pool cases each month.
- 47 -
5.
Submission of Sample Cases Selected
The estimated average caseload, the specified percentage of monthly selected
sample cases for the reserve pool, the computed sample intervals, manually
generated random start and seed numbers to be used in the 12-month sample
period for selection of total sample cases and reserve pool cases should be sent
to the ACF Regional Administrator thirty (30) calendar days before the
October sample selection. The monthly list of selected sample cases, with
reserve pool cases identified, and computer generated random start and seed
numbers should be submitted within 10 days of the date of selection specified
in the State sampling plan.
If random numbers are used to identify cases for the reserve pool, it is absolutely
essential that the total number of sample cases selected each month is known.
The following procedures are to be used and repeated each month if random numbers are
used.
1.
Determine the total number of sample cases selected. In our example,
assume that the number selected for October was 309.
2.
Multiply the number obtained in Step 1 by the percentage of selected
sample cases specified for the reserve pool. In our illustration, the number
of cases to be placed in a reserve pool for October is 309 x .15, or 46 cases
(rounded down). Note that the same percentage must be applied each
month.
3.
Randomly select and identify reserve pool cases. In our example, 46
random numbers between 1 and 309, inclusive, would be selected.
If a Table of Random Numbers is used (see Appendix A, page 107), a
photocopy of the page(s) used, showing start number, direction, and all
selected numbers circled, is to be submitted each month along with the total
list of selected sample cases. Sample cases corresponding to the circled
random numbers are to be identified on the total list.
1532
Procedure for Obtaining Cases from a Reserve Sample Pool
States (Tribes) with reserve pools must use the same procedures in correcting for
undersampling as outlined in Section 1520 (page 44) and 1530 (page 46). A revised
estimate of the number of sample cases that should have been selected (excluding reserve
pool cases) is to be computed. The difference between the number that should have been
- 48 -
selected and the number that was selected is the number of additional sample cases that
will need to be selected from the reserve pool.
If a State (Tribe) uses a disproportionate stratified sample design, the State must maintain
a separate reserve sample pool for each stratum. If a State with a disproportionate
stratified sample design undersamples, the State must use the allocation procedures
specified in its sampling plan to determine in which stratum (or strata) the State has
undersampled. The State must correct for undersampling in each stratum in which
undersampling occurred.
The same primary sample interval as determined in Section 1430, Step 5 (page 34), Step
2, must be used to select sample cases for months in the annual period for which such
cases have not yet been selected. However, a new secondary sample interval to be
applied to the monthly lists of selected sample cases must be used in identifying cases for
the reserve pool.
The revised secondary sample interval is the product of the original secondary sample
interval and the number of cases in the reserve sample pool prior to selection of cases
from the reserve sample pool divided by the number of cases remaining in the reserve
sample pool after selection of cases from the reserve sample pool.
1540
Correction for Undersampling or Oversampling When Sample Was Selected
Using Simple Random Sampling
As described in Sections 1510 and 1520 (pages 42 and 44), there are two basic
approaches to correcting the annual samples. The first approach is to correct the sample
for both the months for which the sample has already been selected and the months for
which the sample has not been selected. This approach provides an annual sample with
approximately one-twelfth of the sample selected each month. The second approach is to
make the entire adjustment in the months for which the sample has not yet been selected.
Monthly samples selected using simple random sampling are less likely to need large
adjustments for undersampling than for samples selected using systematic sampling.
This is true because under simple random sampling a fix number of sampling units is
selected each month regardless of the monthly caseload. Under systematic sampling, a
fixed proportion of the caseload is selected each month. However, caseloads can vary
from month to month. This variation of the monthly caseload results in variation in the
monthly sample size.
- 49 -
1540.1
Correcting for Undersampling
- 50 -
If a small correction (e.g., less than 50 cases) is needed to ensure the State agency or
Tribal grantee will meet its minimum required annual sample size and no month is
substantially short of the approximate one-twelfth of the annual sample, then the State or
Tribe should correct for undersamping by adjusting the sample size in months for which
the sample has not yet been selected. If the sample for all months have been selected,
then the adjustments should be made for the months in the last quarter of the fiscal year.
On the other hand, if a large correction is needed for months in which the sample have
already been selected, the State or Tribe should consider making adjustments to all
monthly samples. To correct a monthly sample for undersampling , use the following
procedures:
1.
Retrieve the Original Monthly Sample Frame
As required under Section 1460, States and Tribes must to retain their
original monthly sample frames. The State or Tribe should locate the
original month sample frame for use in selecting the additional sample
cases.
2.
Review Original Determinations of Total Monthly Caseload and Average
Monthly Sample Size
Review the original application of the sample selection procedures from
Section 1440 Steps 2, 3, and 4 (Page 37) to identify the reason for
undersampling (e.g., under estimated the number of listed-in-error cases).
3.
Determine the Number of Additional Sample Cases Needed
Subtract the number of completed cases for the sample month from the
required number of sample cases for the month to determine the short fall.
Allowing for some additional listed-in-error cases (using the procedures in
Section 1440, Step 4 (Page 37)), determine the number of additional sample
cases to be selected from the original sample frame.
4.
Select the Additional Sample Cases
Using the same procedures as in Section 1440, Step 5 (page 37), select the
addition sample cases from the sample frame and forward the sample
selection list to the ACF Region Office.
- 51 -
1540.2
Correcting for Oversampling
- 52 -
States are not required to correct for excessive oversampling. If correction is desired, the
procedures to correct for excessive oversampling are similar to correcting for
undersampling. However, instead of using the original monthly sample frame, the State
or Tribe would use its monthly sample selection list and apply the following procedures:
1.
Determine the Number of Cases on the Original Monthly Sample Selection
List
This is the total number of sample cases, including cases that were listed-inerror.
2.
Determine the Number of Excess Cases
In determining the number of excess cases, make allowances for the
number of listed-in-error cases in the original sample. A proportion of
these cases will be selected as cases to be removed. For example, if the
original monthly sample had 325 cases of which 25 were listed-in-error,
then the listed-in-error cases represent about 7.7% of the total sample and
the total number of case of the sample frame. In reducing the sample so
that there are at least 250 completed cases, the State or Tribe could expect
about 20 listed-in-error cases. Therefore, the number of excess cases is 325
- 270 or 55 cases.
3.
Select the Excess Cases to be Removed from the Sample
Using the same process as in Section 1440, Step 5 (page 37), select the
number of excess cases from the original sample and forward the sample
selection list of excess cases to the ACF Regional Office.
- 53 -
- 54 -
1600
WORK PARTICIPATION RATES
The purpose of the chapter is to provide States and Tribes with the methodology for
calculating the monthly and annual work participation rates. In applying the
methodology it is necessary to understand the statutory and regulatory provisions on the
work participation requirements, the policies associated with waiver inconsistencies, and
the TANF Data Report - Section One data elements used to capture the information
needed for these calculations. It is for this reason that we have included, as background,
certain mandatory work requirements from the law and the final regulation, which States
must adhere to in administrating their TANF programs. These provisions include:
1610
4.
Establishing the minimum all families and two-parent work participation
rate requirements;
5.
Defining the monthly and annual work participation rate calculations,
including families that are to be included in or excluded from the
calculation;
6.
Identifying countable work activities, including limitation on certain
activities;
7.
Specifying the hourly requirements for engaged in work for the all families
(also known as “overall”) and two-parent families work participation rates
and related special circumstances (e.g., deemed engaged in work); and
8.
Applying waiver inconsistencies as related to the work requirements.
Work Participation Rate Standards and Caseload Reduction Credit
For each fiscal year, the statute specifies the all families and two-parent families
minimum work participation rate standards that States must meet in administering their
TANF Programs. These standards are shown in the following table:
- 55 -
Minimum Required Work Participation Rate
Fiscal Year
All Families Rate
Two-Parent
Families Rate
1997
25
75
1998
30
75
1999
35
90
2000
40
90
2001
45
90
2002 and thereafter
50
90
States that are successful in moving welfare recipients from welfare to self-sufficiency or
otherwise reducing their welfare rolls are given credit for their efforts with respect to
these standards. If the average number of cases receiving assistance, including assistance
under a separate State program, for the State in the preceding fiscal year was lower than
the average number of cases receiving assistance in FY '95, then the minimum work
participation rate standard that the State must meet for the fiscal year will decrease by the
amount of percentage points the caseload has fallen in comparison to the FY '95 caseload.
This reduction in the minimum work participation rate standard is referred to as the
caseload reduction credit. The caseload reduction credit will not include changes that are
required by Federal law or that are as a result of changes in State eligibility criteria.
The minimum two-parent families participation rate the State must meet for the fiscal
year decreases, at State option, by either:
1.
The number of percentage points the prior-year two-parent caseload,
including assistance under a separate State program (as provided in
§261.42(b)), fell in comparison to the FY 1995 two-parent caseload; or
2.
The number of percentage points the prior-year overall caseload, including
assistance under a separate State program (as provided in §261.42(b)), fell
in comparison to the FY 1995 overall caseload.
These calculations must disregard the net caseload reduction (i.e., caseload decreases
offset by increases) due either to requirements of Federal law or to changes that a State
has made in its eligibility criteria in comparison to its criteria in effect in FY 1995.
- 56 -
We will determine the total and two-parent caseload reduction credits that apply to each
State based on the information and estimates reported to us by the State on eligibility
policy changes, application denials, and case closures. In order to receive a caseload
reduction credit, a State must submit a Caseload Reduction Report to us containing the
following information:
1.
A listing of, and implementation dates for, all State and Federal eligibility
changes, as defined at §261.42, made by the State since the beginning of
FY 1995;
2.
A numerical estimate of the positive or negative impact on the applicable
caseload of each eligibility change (based, as appropriate, on application
denials, case closures or other analyses);
3.
An overall estimate of the total net positive or negative impact on the
applicable caseload as a result of all such eligibility changes;
4.
An estimate of the State's caseload reduction credit;
5.
Total prior year caseload;
6.
The number of application denials and case closures for fiscal year 1995
and the prior fiscal year;
7.
The distribution of such denials and case closures, by reason, for fiscal year
1995 and the prior fiscal year;
8.
A description of the methodology and the supporting data that the State
used to calculate its caseload reduction estimates;
9.
A certification that it has provided the public an appropriate opportunity to
comment on the estimates and methodology, considered their comments,
and incorporated all net reductions resulting from Federal and State
eligibility changes; and
10.
A summary of all public comments.
We will calculate the caseload reduction credit that applies to the work participation
rate(s). However, we will not calculate a caseload reduction credit unless the State
reports case-record data on individuals and families served by any separate State
program, as required under §265.3(d). A State may only apply to its participation rate a
caseload reduction credit that we have calculated. If a State disagrees with the caseload
- 57 -
reduction credit, it may appeal the decision as an adverse action in accordance with
§262.7.
A State must report the necessary documentation on caseload reductions for the
preceding fiscal year by December 31.
1620
Definitions of Annual and Monthly Work Participation Rates
The statute defines the overall annual participation rate and the overall monthly
participation rate as follow:
Overall Annual Participation Rate is the average of the State's overall participation
rates for each month in the fiscal year.
Overall Monthly Participation Rate is: (1) the number of families receiving TANF
assistance that include an adult or a minor head-of-household who is engaged in
work for the month (the numerator), divided by (2) the number of families
receiving TANF assistance during the month that include an adult or a minor
head-of-household minus the number of families that are subject to a penalty for
refusing to work in that month (the denominator). However, if a family has been
sanctioned for more than three of the last 12 months, we will not exclude it from
the participation rate calculation.
Other circumstances for which a family may be disregarded from the overall monthly
work participation rate calculation are:
1.
A State has the option of not requiring a single custodial parent caring for a
child under age one to engage in work. If the State adopts this option, we
will disregard such a family in the participation rate calculation for a
maximum of 12 months;
2.
At State option, a family that is participating in a Tribal Work Program may
be included or excluded from the work participation rate calculation. If the
State has opted to exclude all Tribal Work Program participants from its
work participation rate, such families will be excluded from the calculation;
3.
A family that is part of an ongoing research evaluation (as a member of a
control group or experimental group) approved under Section 1115 of the
Social Security Act will be excluded from the participation rate calculation;
and
- 58 -
4.
Some States have an approved 1115 waiver, which contains provisions that
exempt certain groups of recipients from participating in work activities.
Based on these approved 1115 waiver inconsistencies, we will exclude such
families from the participation rate calculation.
The statute defines the two-parent annual participation rate and the two-parent monthly
participation rate as follow:
Two-parent Family Annual Participation Rate is the average of the State's twoparent participation rates for each month in the fiscal year.
Two-parent Family Monthly Participation Rate is: (1) the number of two-parent
families receiving TANF assistance in which the adults or minor child head-ofhousehold and other parent meet the requirements set forth in §261.32 for the
month (the numerator), divided by (2) the number of two-parent families receiving
TANF assistance during the month minus the number of two-parent families that
are subject to a penalty for refusing to work in that month (the denominator).
However, if a family has been sanctioned for more than three of the last 12
months, we will not exclude it from the participation rate calculation.
Other circumstances for which a family may be disregarded from the two-parent monthly
work participation rate calculation are:
1.
At State option, a family that is participating in a Tribal Work Program may
be included or excluded from the work participation rate calculation. If the
State has opted to exclude all Tribal Work Program participants from its
work participation rate, such two-parent families will be excluded from the
two-parent participation rate calculation;
2.
A two-parent family that is part of an ongoing research evaluation (as a
member of a control group or experimental group) approved under Section
1115 of the Social Security Act will be excluded from the participation rate
calculation;
3.
Some States have an approved 1115 waiver, containing provisions that
exempt certain groups of recipients from participating in work activities.
Based on these approved 1115 waiver inconsistencies, we will exclude such
two-parent families from the two-parent participation rate calculation; and
4.
If a two-parent family includes a disabled parent, we will not consider the
family as a two-parent family for the purpose of calculating the two-parent
work participation rate.
- 59 -
For the purpose of calculating the two-parent work participation rate, the two-parent
families must include, but is not limited to, any family with two natural or adoptive
parents (of the same minor child) receiving assistance and living in the home, unless both
are minor and neither are head-of-household. This is a minimal definition. At State
option, a broader definition of two-parent families may be used. For example, a State
may want to include step-parents and/or non-custodial parents.
A State may opt to include a noncustodial parent as part of the eligible family receiving
assistance. If the State does so, the noncustodial parent may receive assistance or other
services and may participate in work activities. The included noncustodial parent must
live in the State, but may not live with his/her child(ren). A noncustodial parent may
participate in work activities funded under the State TANF Program. In addition, the
State must report the noncustodial parent as part of the TANF family. However, the State
may choose whether a two-parent family with a noncustodial parent as one of the two
parents is a two-parent family for the purposes of calculating the two-parent work
participation rate. If a State chooses to exclude a two-parent family with a noncustodial
parent as one of the parents from the two-parent work participation rate, the State must
code the TANF Data Report data element "Type of Family for Work Participation" (data
element #12) with a "2," code “Family Affiliation” (data element #30) with a “1,"code
“Noncustodial Parent Indicator” (data element #31) with a “1,“ code “Parent with a
Minor Child in the Family” (data element #39) with a “1,” and code the data element
"Work Participation Status" for the noncustodial parent with a "99."
1630
Countable Work Activities
The statute requires that adults and minor child heads-of-household participate in certain
work activities. Countable work activities include the following:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Unsubsidized employment;
Subsidized private sector employment;
Subsidized public sector employment;
Work experience;
On-the-job training (OJT);
Job search and job readiness assistance;
Community service programs;
Vocational educational training;
Job skills training directly related to employment;
Education directly related to employment, in the case of a recipient who has
not received a high school diploma or a certificate of high school
equivalency;
- 60 -
11.
12.
Satisfactory attendance at secondary school or in a course of study leading
to a certificate of general equivalence, if a recipient has not completed
secondary school or received such a certificate; and
Providing child care services to an individual who is participating in a
community service program.
Each adult (or minor child head-of-household) has a life-time limit on countable hours of
participation for vocational educational training. Vocational educational training may
only count as a work activity for a total of 12 months.
There are four limitations on job search and job readiness training. These are:
1.
Job search and job readiness assistance only count for 6 weeks in any fiscal
year;
2.
An individual's participation in job search and job readiness assistance
counts for no more than 4 consecutive weeks;
3.
If the State's (Tribe's) total unemployment rate for a fiscal year is at least 50
percent greater than the United States' total unemployment rate for that
fiscal year or the State is a "needy" State (within the meaning of Section
403 (b)(6)), then an individual's participation in job search or job readiness
assistance counts for up to 12 weeks in that fiscal year; and
4.
A State may count 3 or 4 days of job search and job readiness assistance
during a week as a full week of participation, but only once for any
individual.
Some States have an approved 1115 waiver. These waivers may include provisions that
allow the State to count certain additional work activities (e.g., extending participation in
vocational education and/or job search and job readiness training beyond statutory
limitations). The average hours per week of participation in these additional work
activities is captured on the TANF Data Report by the data element "Additional Work
Activities Permitted Under Waiver Demonstration."
1640
Required Hours of Work to be "Engaged in Work"
A family counts as participating in work for the overall work participation rate for a
month in which an adult or minor child head-of-household is engaged in work. That is,
the adult or minor child head-of-household participates in countable work activities
during the month for at least the minimum average number of hours per week as
specified in the following table:
- 61 -
If the fiscal year is:
Then the minimum average hours
per week is:
1997 .................
20
1998 .................
20
1999 .................
25
2000 or thereafter ...
30
At least 20 of the above hours per week must come from work activities (1) through (8)
and (12), as shown in Section 1630. (Hereafter, we will refer to these work activities as
"core" work activities.) Hours above the 20 hours per week may also come from work
activities (9), (10), and (11) as shown in Section 1630. A State with an approved 1115
waiver may be able to count additional work activities (see Section 1630 work activity
(13) and Section 1650) as a core work activity and count the hours of participation in
these additional work activities toward the 20 hours and the hours above the 20 hours.
A two-parent family counts as engaged in work for the month in determining the
two-parent rate, if one of the following is applicable:
1.
If the family does not receive federally-funded child care and an adult (or
minor child head-of-household) and the other parent in the family are
participating in work activities for an average of at least 35 hours per week
during the month, and, at least 30 of the 35 hours per week come from
participation in the core work activities, (1) through (8) and (12) listed in
Section 1630. The family counts as engaged in work. Above the 30 hours
per week, countable hours may also come from work activities (9), (10),
and (11) from those work activities listed in Section 1630.
2.
If the family receives federally-funded child care, an adult in the family is
not disabled or caring for a severely disabled child, and an adult (or minor
child head-of-household) and the other parent in the family are
participating in work activities for an average of at least 55 hours per week
during the month, and, at least 50 of the 55 hours per week come from
participation in the core work activities, (1) through (8) and (12) listed in
Section 1630. The family counts as engaged in work. Above the 50 hours
per week, countable hours may also come from work activities (9), (10),
and (11) from those work activities listed in Section 1630.
- 62 -
1641
Deemed Engaged In Work
For purposes of the overall work participation and two-parent work participation rate, a
family with single minor child head-of-household or married teen parent is deemed
engaged in work in a month if (s)he maintains satisfactory attendance at a secondary
school or the equivalent during the month or participates in education directly related to
employment for an average of at least 20 hours per week during the month.
A single custodial parent or caretaker relative with a child under age six will count as
engaged in work if (s)he participates for at least an average of 20 hours per week. Hours
of participation in all twelve work activities specified in Section 1630 (not just the core
work activities) apply toward meeting this 20 hour requirement. If the State has an
approved 1115 waiver, the additional hours of work activities permitted under the waiver
also count toward the 20 hours.
1642
The Thirty (30) Percent Limit
In counting families for each monthly participation rate, not more than 30 percent of
families with individuals engaged in work in a month may be included in the numerator
because the individuals are: (1) participating in vocational educational training; or (2)
beginning in fiscal year 2000 and thereafter, [individuals] deemed to be engaged in work
by participating in work activities (10) and (11) as listed in Section 1630. For each
month in which the State exceeds the 30% limit, its overall and two parent work
participation rates will be adjusted by decreasing the number of participating families
until the 30% limit is not exceeded.
1650
Waiver Inconsistencies as Related to the Work Requirements
Section 415 of the Social Security Act states that any waiver granted to a State under
Section 1115 of the Act would delay implementation of the amendments made by
PRWORA to the extent that such amendments are inconsistent with the waiver.
The regulations provide States (not applicable to Tribes) with guidance on waiver
inconsistencies as related to the work participation rates and the time limits. The
definitions of "waiver" and "inconsistent" are:
Waiver - consists of the work participation or time-limit component of the State's
demonstration project under section 1115 of the Act. The component includes the
revised AFDC requirements indicated in the State's waiver list, as approved by the
Secretary under the authority of section 1115, and the associated AFDC provisions
that did not need to be waived.
- 63 -
Inconsistent - means that complying with the TANF work participation or sanction
requirements at section 407 of the Act, or the time-limit requirement at section
408(a)(7) of the Act, would necessitate that a State change a policy reflected in an
approved waiver.
In order for the State's alternative waiver requirements to be considered in the calculation
of the work participation rates, the Chief Executive Officer must certify in writing to the
Secretary:
1.
The specific inconsistencies that the state is choosing to continue; and
2.
Consistent with the waiver, the policies that the State will use to:
a.
Exempt individuals from requirements to participate in work
activities;
b.
Assign individuals to an alternative work activities or to an
alternative number of work hours; and
c.
Determine exemptions from or exceptions to the time limits.
The TANF Data Report has four data elements that capture data related to the State
waiver. These data elements are:
1.
Waiver Evaluation Experimental and Control Group;
This data element is used to identify families that are members of a control
group or experimental group of a Welfare Reform Demonstration Project
for the purpose of continuing an impact analysis. Such families are
excluded from the calculation of the work participation rates.
2.
Work Participation Status;
This data element contains the following two codes that identify families
that are disregarded from the work participation rate based on approved
1115 welfare reform waivers:
03
=
Disregarded, family is part of an ongoing research evaluation
(as a member of a control group or experimental group)
approved under Section 1115 of the Social Security Act
- 64 -
04
3.
=
Disregarded from the work participation rate based on an
approved waiver inconsistency that exempts the family from
participation
Additional Work Activities Permitted Under Waiver Demonstration
This data element is used to code the average hours per week of
participation in work activities that are permitted and countable under a
Federally-approved 1115 waiver, but not countable under the statute.
4.
Required Hours of Work.
Some States have a Federally-approved 1115 waiver that permits the State
to set on an individual basis the number of hours of participation for an
individual to be counted as engaged in work. The data element is used to
capture the required hours of work for such an individual.
1660
Tribal Work Participation Rates
Under each Tribal Family Assistance Plan, the Tribe and the Federal Government have
negotiated and agreed upon a number of operating procedures that effect the calculation
of the work participation rates, including the following:
a.
Minimum required hours of participation in work activity to be considered
engaged in work .
b.
Activities which count toward meeting the minimum hours of work .
c.
The participation rate standard to be met.
d.
Groups for which participation rate are to be measured (e.g., one-parent
families, all adults).
e.
The year by which the participation rate standard is to be achieved.
1670 Methodology Used in Calculating the Monthly Work Participation Rate
The monthly TANF population consists of all families who receive assistance under the
State TANF Program for the reporting month. For the all family (and two parent family)
work participation rate, we are interested in a portion of these families. This smaller
- 65 -
grouping is referred to as a subpopulation or subdomain. For the overall work
participation rate, the subpopulation of interest is all TANF families with an adult or
minor child head-of-household, except those families that are disregarded due to:
1.
Single custodial parent with child under 12 months;
2.
Sanctioned for the reporting month, but not sanctioned for more than 3
months within the preceding 12-month period;
3.
Participating in a Tribal Work Program, State has opted to exclude all
Tribal Work Program participants from its Work Participation rate;
4.
Family is part of an ongoing research evaluation (as a member of a control
group or experimental treatment group) approved under Section 1115 of the
Social Security Act; and
5.
An approved waiver inconsistency that exempts the family from
participation.
Similarly, for the two parent work participation rate, the subpopulation of interest is all
two parent TANF families with an adult or minor child head-of-household, except those
that are disregarded due to:
1.
Sanctioned for the reporting month, but not sanctioned for more than 3
months within the preceding 12-month period;
2.
Participating in a Tribal Work Program, State has opted to exclude all
Tribal Work Program participates from its Work Participation rate;
3.
Family is part of an ongoing research evaluation (as a member of a control
group or experimental treatment group) approved under Section 1115 of the
Social Security Act; and
4.
An approved waiver inconsistency that exempts the family from
participation.
The standard statistical methodology for estimating means (proportions are special cases
of means) over subpopulations from universe data, non-stratified samples and stratified
samples are shown below.
- 66 -
1661
Calculation of the Monthly Work Participation Rate from Universe Data
For a State that reports the TANF Data Report for its entire caseload, the monthly work
participation rate (R) is the total number of families participating from the subpopulation
(Yj) divided by the total number of families in the subpopulation (Nj) and is calculated as
follows:
' Yi
N
R '
where i
Yi
Nj
Yj
Nj
'
i'1
Nj
=
1, 2, ..., N
=
1, if the i th family is participating in j th subpopulation
=
0, if the i th family is not participating in j th subpopulation
=
the number of cases in the j th subpopulation
For example, a State with a monthly caseload of 42,600 families reports its entire
caseload as follows:
- 67 -
The number of families that are:
All Families
Two-Parent
Families
1.
Reported (i.e., total caseload)
42.600
3,000
2.
No Parent Families
11,500
-
3.
Listed-in-error
50
-
4.
Disregarded: single custodial parent with
child under 12 months
2,200
-
5.
Disregarded: sanctioned for the reporting
month, but not sanctioned for more than 3
months in the preceding 12-month period
1,775
245
6.
Disregarded: Participating in a Tribal
work program
25
0
7.
Disregarded: family is part of an ongoing
research evaluation (as a member of a
control group or experimental group)
approved under Section 1115 of the Act
0
0
8.
Disregarded (#5): Based on an approved
waiver inconsistency
2,000
295
9.
Required to Participate (item #1 minus
item 2 through item 8)
25,050
2,460
10.
Participating
8,338
1,225
11.
Counting toward the 30% limit
1,731
120
The ratio estimator for the all family work participation rate is:
' Yi
N
R '
Yj
Nj
'
i'1
Nj
'
8,338
' 0.3329
25,050
The ratio estimator for the two-parent work participation rate is:
- 68 -
' Yi
N
R '
1662
Yj
Nj
'
i'1
1,225
' 0.4980
2,460
'
Nj
Calculation of the Monthly Work Participation Rate from Sample Data
For a State that samples but does not stratify, the estimated monthly work participation
rate is calculated using the ratio estimator.
The ratio estimator is:
Rˆ '
yi
xi
N y¯j
N x¯j
'
N n
' y
n i'1 i
N n
'x
n i'1 i
N j
' y
n k'1 jk
n
'
N j
' x
n k'1 jk
n
=
1, if the i th family is participating in j th subpopulation
=
0, if the i th family is not participating in j th subpopulation
=
1, if the i th family is in j th subpopulation
=
0, if the i th family is not in j th subpopulation
The estimated variance for the ratio estimator is:
2
' (yi & Rˆ xi)
n
ˆ '
v(R)
(1 & f)
n x¯2
i'1
n&1
where:
f '
n
N
- 69 -
R '
Yˆ
Xˆ
For example, a State with a monthly caseload of 42,600 families reports based on a nonstratified sample as follows:
The number of families that are:
All Families
Two-Parent
Families
42.600
3,000
1.
Total Caseload
2.
Reported (sample size)
255
51
3.
No Parent Families
60
-
4.
Listed-in-error
2
-
5.
Disregarded: single custodial parent
with child under 12 months
10
-
6.
Disregarded: sanctioned for the
reporting month, but not sanctioned
for more than 3 months in the
preceding 12-month period
19
4
7.
Disregarded: Participating in a Tribal
work program
0
0
8.
Disregarded: family is part of an
ongoing research evaluation (as a
member of a control group or
experimental group) approved under
Section 1115 of the Act
0
0
9.
Disregarded (#5): Based on an
approved waiver inconsistency
10
5
10. Required to Participate (item #2
minus item 3 through item 9) in the
sample
154
42
11. Participating in the sample
56
21
12. Counting toward the 30% limit in the
sample
12
3
- 70 -
For the all family rate:
The estimated total number of families required to participate from the total
caseload (i.e., the denominator of the participation rate) is:
42,600 × 154
' 25,727.0588
255
The estimated total number of families that are participating from the total
caseload (i.e., the numerator of the participation rate) is:
42,600 × 56
' 9,355.2941
255
The estimated monthly all family work participation rate is:
Rˆ '
N y¯j
N x¯j
'
N n
' y
n i'1 i
N n
' x
n i'1 i
N j
' y
n k'1 jk
n
'
N
' x
n k'1 jk
nj
'
42,600 ( 56
255
42,600 ( 154
255
' 0.3636
The estimated number of participating families that count toward the 30% limit is:
42,600 × 12
' 2,004.7059
255
The number of participating families due to vocational education (and after 1999 due to
deemed engaged in work based on work activities 10 and 11 from Section 1630) is less
than 30% of total participating families. Therefore, no adjustment is necessary.
For the two-parent work participation rate:
The estimated total number of two-parent families required to participate from the
total caseload (i.e., the denominator of the participation rate) is:
3,000 × 42
' 2,470.5882
51
- 71 -
The estimated total number of two-parent families that are participating from the
total caseload (the numerator of the participation rate) is:
The estimated monthly two-parent work participation rate is:
Rˆ '
N y¯j
N x¯j
'
N n
' y
n i'1 i
N
' x
n i'1 i
n
N j
' y
n k'1 jk
n
'
N
' x
n k'1 jk
nj
3,000 ( 21
51
3,000 ( 42
51
'
' 0.5000
the estimated number of participating two-parent families that count toward the
30% limit is:
3,000 × 3
' 176.4706
51
The number of two-parent participating families due to vocational education (and after
1999 due to deemed engaged in work based on work activities 10 and 11 from Section
1630) is less than 30% of total number of two-parent families that are participating.
Therefore, no adjustment is necessary.
1663
Calculation of the Monthly Work Participation Rate from Stratified Sample
Data
For a State that selects a stratified sample the monthly work participation rate as
estimated with the ratio estimator is:
' Nhj y¯hj
H
Rˆ '
Yˆ j
Xˆ j
'
H'1
' Nhj x¯hj
H
h'1
where k
=
1, 2, ... nhj
h
=
1, 2, ... H
- 72 -
'
' yhjk
H
j
Nh
n hj
h'1
nh
k'1
H
j
Nh
n hj
h'1
nh
k'1
' xhjk
nhj
=
the number of families in h th stratum and
the j th subpopulation
H
=
the number of strata
1, if the i th family from stratum h is participating in
the j th subpopulation.
yhjk =
0, if the i th family from stratum h is not
participating in the j th subpopulation.
=
The estimated variance for the ratio estimator is :
2
j Nh
1
ˆ Ñ
v(R)
2
¯2
N X
H
1 & fh
h'1
nh
fh '
ShZ2
nh
Nh
where:
ShZ2 ' ShY2 % Rh2ShX2 & 2Rh phXY ShX ShY
phXY '
ShX
2
ShY
2
ShXY
ShX ShY
Nh
' j
(xhi & X¯h)2
i'1
Nh & 1
Nh
(yhi & Y¯h)2
i'1
Nh & 1
' j
- 73 -
For example, a State with a monthly caseload of 42,600 families reports based on a
stratified sample, in which the two-parent families are in stratum 02 and all other families
are in stratum 01, as follows:
Two-Parent
Families
The number of families that are:
All Families
Strata
01
02
02
39.600
3,000
3,000
2. Reported (sample size)
204
51
51
3. No Parent Families
60
0
-
4. Listed-in-error
2
0
-
5. Disregarded: single custodial parent
with child under 12 months
10
0
-
6. Disregarded: sanctioned for the
reporting month, but not sanctioned
for more than 3 months in the
preceding 12-month period
15
4
4
7. Disregarded: Participating in a Tribal
work program
0
0
0
8. Disregarded: family is part of an
ongoing research evaluation (as a
member of a control group or
experimental group) approved under
Section 1115 of the Act
0
0
0
9. Disregarded (#5): Based on an
approved waiver inconsistency
15
5
5
10. Required to Participate (item #2 minus
item 3 through item 9) in the sample
112
42
42
11. Participating in the sample
35
21
21
12. Counting toward the 30% limit in the
sample
9
3
3
1. Total Caseload
- 74 -
For the all family rate:
The estimated total number of families required to participate from the total caseload
(i.e., the denominator of the participation rate) is:
39,600 ( 112
3,000 ( 42
%
' 24,211.7647
204
51
The estimated total number of families that are participating from the total
caseload (i.e., the numerator of the participation rate) is:
39,600 ( 35
3,000 ( 21
%
' 8,029.4118
204
51
The estimated monthly all family work participation rate is:
R '
39,600 ( 35
3,000 ( 21
%
204
51
39,600 ( 112
3,000 ( 42
%
204
51
'
8,029.4118
' 0.3316
24,211.7647
The estimated number of participating families that count toward the 30% limit is:
39,600 ( 9
3,000 ( 3
%
' 176.4706
204
51
In this example, the two-parent work participation rate is based on the data in stratum 02
and the result are the same as in the previous example.
1664
Adjusting the Monthly Work Participation Rate for Exceeding the 30% Limit
If, in the example from Section 1662 for the all family work participation rate, the
number of participating families that count toward the 30% limit is 20 sample cases
(instead of 12 sample cases), then the 30%limit is exceeded. In this instance, the
estimated total number of participating families that count toward the limit is:
- 75 -
42,600 ( 20
' 3,341.1765
255
The estimate number of participating families that counts toward the 30% limit
(3,341.17) exceeds the 30% limit (3,341.1765 / 9,355.2941 = .3571.) To make the
adjustment, first determine the number of participating families that do not count toward
the 30% limit (9,355.2941 - 3,341.175 = 6,014.1176). This group represents the 70% of
the total adjusted number of participating families. Thus, the total adjusted number of
participating families is calculated by dividing the number of participating families that
do not count toward the 30% limit by 0.7 (i.e., 6,014.1176 / 0.7 = 8,591.5966). The
adjusted all family work participation rate is
8,591.5966
' 0.3340
25,727.0588
The adjustment from 0.3636 to 0.3340 is a decrease in the participation rate of 0.296 or
2.96%.
1680
TANF Data Reporting Elements Used in Calculating the Monthly Work
Participation Rate
The overall and two parent work participation rates are calculated based on data provided
on the TANF Data Report - Section One and, for States that do not use a stratified
sample, the TANF Data Report - Section Three, data element #8, the total number of
families. For States that use a stratified sample design, the State must submit for each
month the number of families in each stratum. The TANF Data Report - Section One
data elements used in the calculation are listed below:
Item Number
Data Element
1
State FIPS code
4
Reporting Month
5
Stratum
9
Disposition
12
Type of Family for Work Participation
17
Receives Subsidized Child Care
- 76 -
Item Number
Data Element
27
Waiver Evaluation Experimental and Control Group
30
Family Affiliation
31
Non-custodial Parent
32
Date-of-Birth
37
Marital Status
38
Relationship to Head-of-Household
39
Parent with a Minor Child
48
Work Participation Status
49
Unsubsidized employment
50
Subsidized private sector employment
51
Subsidized public sector employment
52
Work experience
53
On-the-job training (OJT)
54
Job search and job readiness assistance
55
Community service programs
56
Vocational educational training
57
Job skills training directly related to employment
58
Education directly related to employment, in the case of a
recipient who has not received a high school diploma or a
certificate of high school equivalency
59
Satisfactory attendance at secondary school or in a course of
study leading to a certificate of general equivalence, if a
recipient has not completed secondary school or received such
a certificate
60
Providing child care services to an individual who is
participating in a community service program
61
Additional Work Activities Permitted Under Waiver
- 77 -
Item Number
Data Element
63
Required Hours of Work
67
Date-of-Birth (Child)
- 78 -
1700
STATISTICAL METHODS IN DATA ANALYSIS
A State or Tribal grantee may comply with the reporting requirements of TANF by
reporting on the entire TANF caseload or by using data collected through scientifically
acceptable sampling methods approved by the Secretary. In addition to information
necessary to compute participation rates, the sample will provide demographic and
financial characteristics of families, including age, race, sex, education, income, and type
and amount of assistance of family members. Together with a sample of closed cases,
States will be able to generate data on families applying for assistance, families receiving
assistance, and families that have become ineligible. By carefully analyzing the data,
States will be able to examine trends in employment and earnings of families with minor
children. (If the sample is sufficiently large enough, the State will be able to produce
accurate and reliable information on the number of hours of participation in different
activities such as, education, subsidized employment, unsubsidized employment, job
search, etc.)
The following subsections outline some of the more common statistical techniques that
can be used in the statistical analysis process. States are encouraged to do their own
research and develop statistical methodology to meet their own special needs in data
analysis.
1710
Statistical Tests of Significance
Because sample results will normally be in error by some amount simply because they
are based on a sample, inferences from sample results must take into account sampling
error. The means for doing this is known as testing statistical hypotheses and estimation
(including confidence interval construction) for statistically significant differences. The
"difference" may be between two or more samples or between a sample and the
population. The hypothesis used in testing differences (called the null hypothesis) is that
there is no "true" difference between the observed results, i.e., that the observed
difference is only due to random errors or chance. When the observed difference is
sufficiently larger than the sampling error, it can be stated that there is a statistically
significant difference, i.e., that a "true" difference most likely exists.
This section is concerned with various statistical procedures that test null hypotheses.
The tests that follow are appropriate for the systematic random or simple random
sampling methods.
- 79 -
1711
Testing the Representativeness of the Sample with the Caseload
There are several statistical techniques that can be used to ensure that the sample is
acceptably representative of the caseload from which it is drawn. These techniques
involve the comparison of sample case findings with known caseload information. The
two statistical methods that are discussed are: (1) the confidence interval estimate of
population parameters for averages and proportions, and (2) the one-sample chi-square
test for distribution of sample findings.
All States and Tribal grantees collect information on their entire caseload on an ongoing
basis -- monthly, quarterly, or annually. Caseload data closest to those of the sample
period should be used in making the comparisons. If the test reveals significant
differences in results, the method of sample selection and sample sizes should be reexamined to provide assurance that no errors have occurred in the sample selection
process.
Sections 1711.1 and 1711.2 below illustrate the methods using the confidence interval to
estimate representativeness of the sample when proportions are not used and when
proportions are used.
1711.1
Comparison of Sample and Total Caseload When Proportions Are Not Used
In order to determine whether the sample average dollar amount of assistance is
representative of the caseload, use the following procedure:
For our example, assume that the average dollar amount of assistance in the total
caseload is $90.20 and in the sample, $95.35 with a standard error of $5.48.
The equation for a 95 percent confidence interval in this calculation is approximately as
follows:
x¯ ± 1.96
s
n
where:
n
=
number of sample cases for which a review was completed;
- 80 -
=
x¯
mean dollar amount of assistance per sample case for which data
was collected = ' xi ;
n
j
=
the sum operator;
s
=
estimated standard deviation =
xi
=
actual dollar amount of assistance for a sample case; and
=
estimated standard error of x¯.
s
n
' (xi & x¯)2
n & 1
;
If, in our example, the estimated standard error of the sample average dollar amount of
assistance is $5.48, then 1.96 times the standard error is $10.74. Therefore, the 95
percent confidence limits are $95.35 ± $10.74, or $84.61 to $106.09. Since the
confidence interval in this case includes the "true" or total caseload average dollar
amount of assistance of $90.20, there is no evidence that the sample is not representative
of the caseload from which it is drawn.
1711.2
Comparison of Sample and Total Caseload When Proportions Are Used
If information on the proportion of the entire caseload having certain characteristics is
available, a similar test can be conducted. For example, if the proportion of 2-parent
families in the entire caseload is known, the sample proportion can be compared to this
figure. In this situation, a confidence interval is calculated around the total caseload, or
population value, to see if the sample value is included.
The sample proportion of 2-parent families should fall within the following interval:
P ± 1.96
P (1 & P)
n
where:
P
=
proportion of 2-parent families in the caseload; and
- 81 -
n
=
number of completed sample cases
It should be noted that the best estimate of a standard error uses the most complete data
readily available. Theoretically, total caseload data, if available, should be used to
calculate the standard error wherever findings are compared between sample cases and
the total caseload. Calculation of the standard error from total caseload data is a
relatively simple process where proportions are being compared. However, where
proportions are not used, as in comparing average dollar amount of assistance,
calculating the best estimate of the standard error from the total caseload is a very lengthy
process. In such circumstances, the standard error is calculated from the sample data.
One Sample Chi-Square ( Χ2 ) Test
1711.3
This method for testing the representativeness of samples compares the distribution of
sample cases by certain characteristics with that of the total caseload. The assumption is
that a certain amount of information is available based upon universe counts of the entire
caseload.
The most readily available characteristic that can be compared is the distribution of cases
by county, or other geographic areas. If cases in the sample have been drawn with each
case having an equal chance of selection, they would be distributed among the counties
or other geographic areas in the same proportions as cases in the total caseload. To
determine if the county (or other geographic area) variations in sample cases are large
enough to support a possible suspicion of bias, the chi-square test of significant
differences can be computed. In the chi-square test, theoretical, i.e., expected values are
computed. If the observed values differ greatly from these expected values, a significant
concentration is in evidence.
The equation for computing the chi-square statistic is as follows:
X2 ' j
(O & E)2
E
where:
j
=
the sum of all categories;
O
=
observed number of cases in each category (or case characteristic);
and
- 82 -
E
=
expected number of cases in each category (or case characteristic)
which when calculated is as follows:
number of population cases in category
x sample size
total cases in population
The following example will illustrate the method.
Comparison of Distributions of Cases by County Groups
County
Groups
Total Caseload ( U )
Observed Number
of Cases in
Sample ( O )
Expected Number of Cases
in Sample
U
E '
x (' O)
'U
#1
1,000
11
1,000
x 200 ' 10
20,000
#2
3,000
33
3,000
x 200 ' 30
20,000
#3
5,000
58
5,000
x 200 ' 50
20,000
#4
4,000
57
4,000
x 200 ' 40
20,000
#5
2,000
13
2,000
x 200 ' 20
20,000
#6
2,000
13
2,000
x 200 ' 20
20,000
- 83 -
Comparison of Distributions of Cases by County Groups
County
Groups
Total Caseload ( U )
#7
Observed Number
of Cases in
Sample ( O )
Expected Number of Cases
in Sample
U
E '
x (' O)
'U
15
3,000
x 200 ' 30
20,000
' O ' 200
' E ' 200
3,000
' U ' 20,000
X2 ' j
(O & E)2
E
=
(11 & 10)2
10
+
(33 & 30)2
30
+
(58 & 50)2
50
+
(57 & 40)2
40
+
(13 & 20)2
20
+
(15 & 30)2
30
=
21.31
To show significance, the computed value must exceed the critical value in the following
table.
Critical Chi-Square ( Χ2 ) Values
Degrees of Freedom
Critical Value of Χ2 Statistic
1
3.84
2
5.99
3
7.81
- 84 -
Critical Chi-Square ( Χ2 ) Values
Degrees of Freedom
Critical Value of Χ2 Statistic
4
9.49
5
11.1
6
12.6
7
14.1
8
15.5
9
16.9
10
18.3
11
19.7
12
21.0
13
22.4
14
23.7
15
25.0
16
26.3
17
27.6
18
28.9
19
30.1
20
31.4
21
32.7
22
33.9
23
35.2
24
36.4
25
37.7
26
38.9
27
40.1
- 85 -
The critical value is dictated by the number of "degrees of freedom." Problems of this
type have degrees of freedom equal to the number of categories minus "1", in this
example, 7 & 1 ' 6 . The critical value of 12.6 is clearly exceeded. Thus, a suspicion
of possible bias in the sample is given greater validity and observed variation in such
categories is more than can reasonably be attributed to chance. (The table of values is set
at 95 percent, i.e., when a computed value exceeds the table value, there is less than 5
chances out of 100 that the large observed differences are due to chance. This predefined
statistical probability, in this table, set at alpha = .05 is called a Type I error.)
Note that the chi-square test is inapplicable, i.e., serious distortions of results may appear,
when 20 percent or more of the groups have expected values of less than "5" or any
group has an expected frequency of less than "1." Under these circumstances, groups
must be combined until the requirements are satisfied. When practical, such
combinations should be made before obtaining or looking at the sample results, in order
to avoid biases in the test. The combinations should be meaningful, e.g., rural counties,
northern counties, etc.
If there are only two groups, each expected value must be "5" or more. In such tables, the
preferred calculation of chi-square is as follows:
Χ2 ' j
(* Observed & Expected * & 0.5)2
Expected
6
6/ The parallel bars, **, indicate absolute value of the term, i.e., ignore the sign and
assume positive. The 0.5 figure is called the Yates Correction for Continuity.
- 86 -
1712
Testing Differences of Proportions Between Samples
Repeated sampling from a given population should not differ from each other by more
than chance fluctuations.
The equations used to determine the statistical significance the of difference in
proportions, such as participation rates, between two reporting periods and using a
predefined probability (Type I error, or alpha = .05) are as follows:
p '
nA pA % nB pB
nA % nB
where:
p
=
weighted participation rate for reporting periods A and B combined;
pA
=
participation rate for reporting period A;
pB
=
participation rate for reporting period B;
nA
=
number of sampled cases in reporting period A; and
nB
=
number of cases reviewed in reporting period B.
The equation for the statistic is as follows:
7
z '
* pA & pB *
p (1 & p) × (
1
1
%
)
nA
nB
If the computed value of " z " is greater than 1.96, a significant difference exists between
pB and pA .
7/ The " z " test is satisfactory only if both n A and nB are large.
- 87 -
For example, assume a participation rate of 33.2 percent based on 1,573 sample cases in
sample period A is compared with an participation rate of 25.7 percent based on 1,495
sample cases in sample period B. The test of significance would be computed as follows:
(1,573) (0.332) % (1,495) (0.257)
' 0.295453 or 0.295
1,573 % 1,495
p '
* .332 & .257 *
z '
.295 (1 & .295) ×
' 4.55
1
1
%
1,573
1,495
Since the computed value of z (4.55) is larger than 1.96, the difference between the
participation rates is statistically significant. If the computed value was less than 1.96,
the difference would not have been statistically significant. It is, therefore, reasonable to
deduce that the observed difference in the participation rate is not attributable to chance
fluctuations.
Testing Differences Within the Same Sample -- Chi-Square ( Χ2 )
1713
A test of statistical significance can be used to determine if the characteristics of one
group vary significantly from the characteristics of another. For example, this test can be
used to compare the distribution of participants in one county versus another.
In testing this hypothesis, the chi-square test uses "column" and "row" groupings.
Although the expected values are computed differently than in Section 1711.3 (page 80),
the overall equation is computed the same way:
Χ2 ' j
(O & E)2
E
where:
O
=
observed number of cases in each grouping; and
- 88 -
E
=
proportional number of cases expected, if no sampling variation was
present, computed as follows:
E '
row total x column total
grand total
To determine whether the distribution of participants by activity varies between two
counties, assume the following data were observed from the sample:
Number of Sample Cases
Type of Activity
County A
County B
280
160
120
Education
45
30
15
Subs. Employment
40
15
25
Unsubs. Employment
22
12
10
Public Sector Empl.
17
7
10
Job Search
55
35
20
Job Skills
54
34
20
Voc. Training
47
27
20
Total
Total
- 89 -
Expected Values ( E )
County A
County B
Education
160 x 45
' 25.7
280
120 x 45
' 19.3
280
Subs. Empl.
160 x 40
' 22.9
280
120 x 40
' 17.1
280
Unsubs. Empl
160 x 22
' 12.6
280
120 × 22
'
280
9.4
Public Sector
Empl.
160 × 17
'
280
120 × 17
'
280
7.3
9.7
Job Search
160 × 55
' 31.4
280
120 × 55
' 23.6
280
Job Skills
160 × 54
' 30.9
280
120 × 54
' 23.1
280
Voc. Training
160 × 47
' 26.9
280
120 × 47
' 20.1
280
Χ2 ' j
(O & E)2
E
=
(30 & 25.7)2
%
25.7
(15 & 19.3)2
%
19.3
(15 & 22.9)2
%
22.9
(25 & 17.1)2
%
17.1
- 90 -
(12 & 12.6)2
%
12.6
( 10 & 9.4)2
%
9.4
( 7 & 9.7)2
%
9.7
(10 & 7.3)2
%
7.3
(35 & 31.4)2
%
31.4
(20 & 23.6)2
%
23.6
(34 & 30.9)2
%
30.9
(27 & 26.9)2
%
26.9
Χ2
=
(20 & 23.1)2
%
23.1
(20 & 20.1)2
20.1
.72 + .96 + 2.73 + 3.65 + .03 + .04 +
.75 + 1.00 + .41 + .55 + 0 + 0
= 10.29
To determine if the computed chi-square value is significant, i.e., the concentrations of
error can not be reasonably regarded as due to sampling variation, the table of Critical
Chi-Square Values in Section 1713 (page 86), should be used. The appropriate number
of degrees of freedom (DF) for examples of this type (with any number of rows or
columns) is computed by the following equation:
DF ' (number of rows & 1) x (number of columns & 1)
Again, if the computed chi-square value exceeds the table value, the value is significant,
i.e., participation in different activities varies from County A to County B. In this
example, DF = 6. Since 10.67 does not exceed 12.6, the data is not significant at the .05
level.
- 91 -
Interpretation of significant data is a somewhat more complex task. Briefly, the analyst
must look to the source of the greatest variation, noting whether the observed value was
larger or smaller than expected. If this test had shown statistical significance, the analyst
would need to further examine the subsidized employment category, where County A
had a smaller than expected number of cases while County B had a greater than expected
number of cases.
The restrictions on the use of this table are the same as in Section 1711.3 (page 80) -- that
the test is inapplicable, i.e., serious distortions of results may appear, when 20 percent or
more of the cells have expected values of less than "5" or any cell has an expected value
of less than "1." Under these circumstances, rows and/or columns must be combined
until the requirements are satisfied.
As indicated in Section 1711.3 (page 80), in a 2 x 2 table, each expected value must be
"5" or more. (In such tables, the preferred method for computing the chi-square is by the
use of the equation given in Section 1711.3.)
1720
Trends
It may be important in a State for the TANF system to have feedback on apparent
changes over time for a variety of statistics (e.g., changes in caseload, in participation
rates, in out-of-wedlock births, in error rates). The general direction of change in data
over time is called the "trend" and can be used, for example, to assess the effectiveness of
State policies or of corrective actions in reducing error rates. Throughout this section, we
are using the error rate, however the methodology is applicable to other proportions.
Trends can be based on moving averages of error rates or on individual monthly error
rates.
1721
Moving Averages
Trends based on a moving average involve taking the averages calculated over a fixed
number of months and progressively dropping data for the earliest month and adding data
for the latest month. In this way, the composition of each fixed time period average
remains approximately the same because any given average covers early, middle, and late
months of the fixed period. Monthly aberrations are smoothed because these fixed
groupings are not particularly sensitive to any given monthly rate. Thus, the long term
trend can be judged visually.
For TANF purposes, a six-month moving average is recommended. Six-month moving
averages can be computed on reviewed sample cases by either month of review or by
month of completion (see Figures 2. and 3. below). The advantage of computing six- 92 -
month averages by month of review is that the effectiveness of corrective actions for
which results are expected at a given point in time can be observed more clearly than if
computed by month of completion. On the other hand, averages computed by month of
completion have the advantage of timeliness, i.e., there is no delay of several months for
cases to be completed before a trend can be observed. (It should be noted,
however, unless cases are completed on a more continuous flow basis than is generally
true at the present time, moving averages based on month of completion can lead to
spurious peaks and valleys in the data.)
- 93 -
Figure 2.
Six-Month Moving Averages of Completed Sample Cases
By the Month of Review
- 94 -
Figure 3.
Six-Month Moving Averages of Completed Sample Cases
Regardless of Month of Review
1722
Individual Monthly Rates
Individual rates are generally examined for short-term time periods. Because each
month's sample is small, the monthly error rates tend to fluctuate much more than sixmonth moving averages. The classic way of measuring this trend is to fit a mathematical
trend line, called a regression line, estimated by the method of "least squares." While a
trend line could be drawn by inspection, such a line probably would be inaccurate and
- 95 -
would be graphed differently by different people, depending on who was drawing the
line. The advantages to the regression line are: (1) the sum of squares of monthly error
rate deviations from the trend line is minimized; (2) all analysts fit the same line;
(3) different measures, e.g., degree of relationship, can be computed; and (4) future
estimates can readily be extrapolated from the line.
1723
Computation of a Regression Line by "Least Squares" Method
It is best to fit the line after all sample cases for the annual sample period have been
completed. In our example, we are using the error rate; however, the methodology is
applicable to other proportions. If a regression line is to be fitted for shorter or longer
periods, the overall error rate for the shorter or longer period must be used in the
computation. The form of the equation used is as follows:
pˆm ' a % b m
where pˆm is the estimated error rate for a given month. The equation for "b" is:
b '
¯
' nm (pm & pt) (m & m)
¯2
' n m (m & m)
where:
b
=
error rate change (increase or decrease) per unit month advance;
nm
=
number of sample cases completed for the m th month;
pm
=
actual proportion of error cases in sample for the m th month;
pt
=
actual proportion of error cases in sample for annual sample period;
m
=
"1" for first month; "2" for second month; etc.; and
- 96 -
a ' pt & b m
m
¯
' (m) (nm) .
=
' nm
The equation for "a" (the y intercept) is:
To illustrate the "least-squares" method of fitting a trend line, data for a six-month period
are used. Assume the number of sample cases completed and the case error rate for each
month to be as follows:
Number of Cases
Case Error
Reviewed ( nm )
Proportions( pˆm )
1
203
.082
May
2
201
.088
June
3
197
.065
July
4
194
.049
August
5
202
.080
September
6
204
.063
' n m = 1,201
pt = .071
Month
m
April
Total
Step 1. Compute m
¯ :
m
¯ '
j (m)(nm)
j nm
- 97 -
m
¯ '
1 (203) % 2 (201) % 3 (197) % 4 (194) % 5 (202) % 6 (204)
1,201
m
¯ '
4,206
1,201
m
¯ ' 3.5
Step 2.
Compute b:
b '
¯
' nm (pm & pt) (m & m)
¯2
' n m (m & m)
Month
(m)
nm
(pm & pt)
(m & m)
¯
1
203
(.082-.071)
(1-3.5)
-5.583
1,268.75
2
201
(.088-.071)
(2-3.5)
-5.126
452.25
3
197
(.065-.071)
(3-3.5)
.591
49.25
4
194
(.049-.071)
(4-3.5)
-2.134
48.50
5
202
(.080-.071)
(5-3.5)
2.727
454.50
6
204
(.063-.071)
(6-3.5)
-4.080
1275.00
3 = -13.605
3 = 3,548.25
b '
Step 3.
&13.605
3,548.25
=
nm(pm & pt)(m & m)
¯
&.0038 or &.38%
Substitute and solve for " a " ("the y intercept"):
- 98 -
nm(m & m)
¯2
a ' pt & bm
Step 4.
=
(.071) & (&.0038)(3.5)
=
.0846 or 8.46%
Substitute the equation for the line into the general form and solve for pˆm
values using m = 6 and m = 0:
pˆm ' a % b m
When m
=
6, then
pˆm ' .0846 & .00383 (6) ' .0616 or 6.16% .
When m
=
0, then
pˆm ' .0846 & .00383(0) ' .0808 or 8.08% .
Step 5.
Draw a trend line on a graph (Figure 4.) using the values of pˆm and m from
step 4.
- 99 -
Figure 4.
Trend Line
Error Rate
(pˆm)
Month (m)
The graph shows an inverse relationship between month sequence and error rates, i.e., the
error rates decrease as the months progress. The trend line would be more accurate if
twelve months of data were used instead of six months.
Once the trend line is established, it is possible to compute from it what the estimated
error rate would be each month if only the factor of trend affected the rate; in other
words, what the error rate would have been if there were no unpredictable or cyclical
factors affecting it.
- 100 -
1723.1
Practical Uses of Trend Line and Trend Values
The differences between the actual and trend values of the error rates show whether the
actual values are above or below the values they would have been if only trend affected
the rates. These differences may reveal the combined effect of such factors as policy
changes and staff turnover on the eligibility and payment process in the TANF program.
The trend line also provides a basis for estimating probable error rates in future periods.
The accuracy of such estimates will depend on the number of points used in the time
series and the assumptions made regarding the future effects of unpredictable factors on
the error rates. (It should be noted, however, that the line of best fit is an average line,
and predicting error rates beyond the range of values used to compute the line assumes
the same scattergram beyond the range.)
The regression line p ' a % b m discussed above is restricted to linear regression
only, i.e., fitting a straight line to the data. If the scatter diagram from the data indicates
non-linearity (e.g., no pattern or curvature), the model given in Section 1723 (page 94) is
not applicable. Other appropriate methods or models should be considered.
1723.2
Testing Trend for Statistical Significance
Testing for a significant trend is actually a test of the null hypothesis, i.e., b = 0 in the
equation p ' a % b m . The test statistic used is again the chi-square test. The
following equation is not in the form shown earlier for chi-square but it can be shown
that this statistic is distributed as chi-square with one degree of freedom when the number
of months is large. Therefore, the critical value is 3.84 (see Section 1711.3 (page 80),
Critical Chi-Square ( Χ2 ) Values).
Χ2 '
[' nm (pm & pt) (m & m)]
¯ 2
[pt (1 & pt) 'nm (m & m)
¯ 2]
All of the terms in the equation have been previously computed for the regression line
itself. Thus, substituting in the equation using the data for the example in Section 1723
(page 94):
Χ
2
'
(&13.605)2
185.096
'
' 0.789
(.0712) (1 & .0712) (3,548.25)
234.039
- 101 -
Since the computed Χ2 value (0.791) is less than 3.84, the trend is not statistically
significant.
Note that in this example, there are only 6 months available. This may not be large
enough to ensure the satisfactory use of the chi-square test. The example is used only to
illustrate the computation. Basing predictions on a linear fit that is not statistically
significant is highly questionable.
1723.3
Relationship Between Time Sequence and Error Rates
In comparing the error rates over the months of the sample period, it is frequently
desirable to measure the degree of relationship. One way of looking at this relationship is
to determine how similar or the closeness of the relationship between the error rates and
time.
The statistic usually used to determine the mutual relationship between two variables is
called the coefficient of correlation ( r ). It ranges from +1 to -1. If a perfect
relationship exists as rates rise over the period, the coefficient of correlation equals +1. If
a perfect relationship exists as rates decline over the period, the coefficient equals -1. If
no relationship exists, the computed value equals zero. Rarely are there situations where
r ' ± 1 .
The following equation is used to compute the coefficient of linear correlation:
r ' b
¯2
'nm (m & m)
'nm (pm & pt)2
The only term that has not been computed for our example in Section 1723 (page 94) is
'nm (pm & pt)2 . This computation is as follows:
Month (m)
nm
(pm & pt)
nm(pm & pt)2
1
203
(.082 - .071)
.025
- 102 -
Month (m)
nm
(pm & pt)
nm(pm & pt)2
2
201
(.088 - .071)
.058
3
197
(.065 - .071)
.007
4
194
(.049 - .071)
.094
5
202
(.080 - .071)
.016
6
204
(.063 - .071)
.013
' = .213
Substituting all the computed terms in the above formula, the coefficient of correlation is:
r ' & .0038
3,548.25
' &.49
.213
Thus, the degree of relationship on the scale of -1 to +1 is -0.49.
It should be emphasized at this point, that there is no direct or proportional comparison
between different values of r . For example, when the coefficient of correlation (r)
between two variables is +0.8, it does not mean that the association is twice as good as
that shown by a value of r = +0.4.
Assume that in our example, the State wants to know how much of the variation in the
error rate is associated with or explained by the time sequence. A simple method of
measuring this explained variation in terms of a percentage of the total variation has been
developed through the use of the coefficient of determination (r 2) :
Coefficient of determination = 100 r 2
(explained variation)
From this formula, the percentage of unexplained variation can also be calculated:
Unexplained variation = 100 (1 & r 2)
- 103 -
In our example, the coefficient of correlation was -0.49. Therefore, only 24 percent, or
(100) (0.49)2 , of the total variation in error rates is accounted for by the time sequence.
Conversely, it can be determined that the time sequence fails to account for 76 percent of
the total variation in error rates, or (100) (1 & 0.492) . Obviously, other factors play a
more important role in the decrease in error rates and must be brought into the analysis.
As noted in Section 1723.2, (page 99), since in this example the regression line is not
statistically significant, neither the coefficient of correlation nor the coefficient of
determination is statistically significant. The example only serves to illustrate the
computations. It should be recognized that when r is based on a sample, it is subject to
chance variation, just as is any other statistic based on a sample. Thus, before assuming a
strong or weak correlation, consideration should be given not only to the value of r ,
but also to the size of the sample. Furthermore, sample correlation analysis has some
basic limitations. A common-sense approach is needed to tell whether two variables (in
this example, error rates and time) are, in fact, casually related or the apparent
relationship is just a coincidence.
1730
Statistical Procedures for Developing Profiles of Error-Prone Cases
The purpose for developing profiles of error-prone or high risk cases or characteristics is
to facilitate the identification of those particular types of cases or characteristics that
should be singled out for special consideration, review, or treatment. For example, cases
with a particular combination of factors might be redetermined for eligibility more
frequently than other cases; particular elements might require more verification; or cases
more likely to be in error might be emphasized in training.
In determining the kind of statistical method to be used in developing error-prone
profiles, a State should consider sample size, whether the error rate is high or low, and
whether it wants the profile to have limited or broad error-prone groups. Resource
demands and statistical availability should also be considered. Demands upon State
resources will vary with the procedure selected.
Various statistical procedures are used in analyzing and predicting the risk and the
expected amount of error of cases possessing a specific type of error. One predictive
technique used with quantitative or numerical data is called multiple regression. Another
technique, known as discriminant analysis, uses multivariate quantitative information.
Multiple regression techniques can be used to predict the expected dollars in error in
cases possessing certain characteristics. Corrective action can then be focused on cases
possessing characteristics associated with the highest average dollars in error.
Discriminant analysis can be used to determine the likelihood of a case being in error.
- 104 -
This predictive technique tries to define a functional relationship for assigning certain
types of cases to various groups.
Most of the procedures that have been used in the TANF program establish specific
characteristics from the sample by which a case is determined to belong to a certain
group. They are generally case-driven procedures that take one of two approaches. Either
a search is conducted for characteristic combinations that have a high concentration of
case errors or a procedure is developed to rank cases from most error-prone to least errorprone. (It should not be too difficult to make these procedures dollar-driven. In the
former procedure, the search criteria can become a high concentration of dollar errors. In
the latter, the definition of error can be modified so that most error-prone implies most
prone to high dollar error. This might be accomplished by defining an error case as one
in which (1) the amount in error exceeds a certain amount, such as the median amount of
error; or (2) the percent of the amount in error exceeds a certain percentage, perhaps of
the total payment. Techniques of regression analysis would be well suited for developing
a procedure that predicts the amount in error for a given case.)
All procedures used in the TANF program are based on a prior quality control sample. If
the conditions under which the sample was reviewed remain constant, the sample can be
used to predict cases most likely to contain errors. However, if these conditions change,
so must the procedures.
1731
Criteria for Setting Up Error-Prone Profile Models
Cases selected and reviewed as error-prone should have the highest likelihood of being in
error and should produce the highest cost savings to a State. The error-prone model
should meet the following specifications:
13.
Cases are ranked by error proneness so that resources are expended more
efficiently;
14.
Screening models are easy to use so that extensive time is not required to
train supervisors;
15.
Criteria used are quick to apply so that extensive time is not needed to
identify error-prone characteristics in the case file;
16.
Models can be incorporated into the existing case processing system;
17.
Models include a monitoring component that informs the agency of success
rates;
- 105 -
18.
System is easily updated so that staff can adjust the model to reflect
changes in caseload; and
19.
System is cost-effective and feasible.
- 106 -
APPENDICES
- 107 -
- 108 -
APPENDIX A
Table of Random Numbers
A table of random numbers is a compilation of numbers whose frequency and sequence
of occurrence have been determined by chance. Since the position that any digit occupies
is a result of chance, any number formed by a combination of these digits, in any
sequence, by any progression, systematic or random, in any direction from any starting
point, may be regarded as a random grouping or selection.
The only requirement is that all of the items from which a random selection is to be made
have, or were assigned, individual identifying numbers. The entire group of numbered
items may be regarded, for certain purposes, as a statistical population. A selection of
any part of that statistical population by means of a table of random numbers may be
regarded as a random sample of the population.
For example, if the population to be sampled consists of 84 cases, numbered from 1
through 84, random numbers of two digits are required. If the population to be sampled
consists of 796 cases, random numbers of three digits are required. To obtain a two-digit,
three-digit, seven digit or other size number from the table, combine adjacent digits as
needed. It makes no difference where in the table one begins or in which direction one
moves in selecting random numbers. However, each time the table is used, select a
different starting point.
Example: If the highest consecutively numbered case in the population is 7,543, assume
that a randomly selected location starts with the four digits in line 49, column 1. Assume
also that it is decided in advance that the numbers to be used in drawing the sample will
be consecutive numbers obtained by reading across the columns from left to right on each
consecutive line in the table until a sample of the desired size has been accumulated. If
the first four digits of each number in each five-digit column are used, the sample would
consist of cases identified as 6837, 7076, 1059, 0454, 5432, 0234, 1724, 2886, 1477,
6273, 1566, and so on until the desired sample size is obtained. The numbers 9501,
9352, 7646, 9227, as well as any other number larger than 7,543 that may later be
encountered are not usable for this universe and are, therefore, rejected.
- 109 -
- 110 -
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
11048015011015360201181647916466917914194625903620720969995709129190700
22236846573255958539330995891982798253402939653409552666191743961599505
32413048360225279726576393648091517924830493403208130680196556334858629
44216793093062436168007856163763944053537713415700400849749179775816379
53757039975818371665606121917826046881305496846067214110069270126354613
67792106907110084275127756534981860270659906551505321916818254439442880
79956272905564206999498872310167119418738440134884063213210691063412952
89630191977054630797218876209229459556869690146004518425849034250832307
98957914342636611028117453181035774084378253311256658678449470558556941
108547536857433425398853060595333886762300081581798316439114581859364952
112891869578882313327670997799365686505859901063159501547855909161078188
126355340961482350342749626694451866372695521802084712234905113370390322
130942993969526369273788974334883632017617300150827284115271563061374952
141036561129875298568948237522676768993394015112635885104202852997589868
150711997336710480817877233139164756481056977358597729372744612855190707
165108512765518215125977452163086075692144494425390070960639907560140719
170236821382524046026889368198855532244819011886525564835449190594455157
180101154092333629490431273041461859429852715858503051132019159274764951
195216253916463695858623216145138314998736234956435094738177523515635749
200705697628337870999842698066917698813602518514610488916195092562558104
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
214866391245858281434609172301689022904734591932217830421616669990432812
225416458492224217410347070253067646826384581510664621524152279690944592
233263932363055972420013363380059434228728358060691217012641611829622851
242933427001876378730858731002564583415398465574113510367076843618818510
250248833062288340735119731924206095261280500016765832586866795072094953
268152572295048399642324878826516656614778767971478013300870747966695725
272967620591680862643246901208498976881536866451265992259571028042825280
280074257392390646643284673400273283261362989479606764760645849609698253
290536604213256692642244407440483793763904457666613475470665203469390449
309192126418641179430526766259403997222209715006456891402424160784469618
310058204711879177734142206351267408799547818174260743808766556202876630
320072569884627975617086324880727622236086846379316176038658557791988006
336901165797958765529318988273542657508625408015992029841801501277748501
342597657948298888860467917487081891282271654246977433611542628596303547
350976383473735771290830883183172829035797059984168834952378883891788050
369156742595279583013404024863852988099730555368485529080092507965673211
371795556349909994912720044599310611520542180590200873708835173610342791
384650318584188454961802304510382065558727281681547556942533892056287338
399215789634948247817184610828340992225417441374841325555212463550920468
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
401457762765356058126339667473585687356307616074951889656201037749018062
419842707523333626427001638924776696998420048804558546565041024688045709
423491463976887208276534476170328758940836324277000270663888637777569348
437006028277394754647323219534169497025832699759488419661728280010266794
445397654914069906724568350829481139842878802878826747363466340654197809
457607229515409800739158745257742298780059399119618941151142226069759583
469072552210839742999265831388575049083765556571436131720573755622841546
476436467412333393192614883244135974492351974738928635931041102372651900
480896200358316622538861642340728124935648568916935248373455787854781788
499501268379935267076510593045427646354328023491724728865147776273092277
501566410493204923839191132219995951681652271954822346751229233226185653
511640881899041535338179401214388303592350366933123859649917547277202338
521862981953055209196204739130929766224822947300649635090048228677298289
537311535101474988763799016710608882471013187352028623153729243516543040
545749116703231674932345021331321254441035807804539344812125159893191202
553040583946237921442215059457992271619792099837435368668304297073525499
561663135006859009827532388523901681569298827323848073817325234196144437
579677320206425597898505300221642436954224350831968711052914916038319746
583893564202143498267466523441330069735552359701912463318296860338759846
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
593162476384174035336344167644866475875366765543160112614330726033292325
607891919474236322788947914025843768020801721523933934806089308500187820
610393133309570477421163445173616282539908056079128468833255703881846920
627442633278439721011989917156655287273823731448866288970744925180599378
630906600903207959545292648454540955288815165535112579375975961629666092
644223812426870251426720979045086453531355860642947247689059745246816834
651615308002265044174481959656427424056302000336710777510706252872534191
662145740742298209678329400218401503534537333100611695240159571657206004
672158157802020508972817937376214707542080974034862668995438053338621597
685561278095831973373205810248138690260397164890326488525427860526992532
694465766999993245128184463605637931293454688762547193911256501268273572
709134084979469498197337949610234399715263806444394289203717959953350501
719122721199319352702284067054623521614486298916860741867149519169685065
725000138140663211992472163095381215106878919031874934405560878279070925
736539005224729582860981406391472554948542426274523357202946172377207896
742750496131839444157510573086196448273923361520518494142252998438734925
753716994851391178963200959164876553649071397821709502330743010027548280
761150870225511113835119444664997194505422134427867584081669389365459894
773744930362066945469004052531156275795348786621116381651502453497152924
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
784651570331859223832957015157659716117869453496179666345810734910679860
793098681223424165835321532305023230586482051740790154339588617481846942
806379864995465830976544160781288399142865925208353180377359098125054238
818248684846992546763243218500762136164816512028812441870526895127583556
822188532906924310906064297516746412662570261230515559194527992822585762
836033698782074085345813564590892644529789852054100112535121331464523541
844393746891240102556086355339412578654990718991547595434982272182419585
859765663175893031627507100920632194218611473482020318534038627809550136
860329901221054183898255758922372675986367212169844208303566139151175928
877962606486035741766807785760207992425651833258842885076728112271750585
888563668335475390312965651119770251026113994476864534327151525523093448
891803914367613370617712143466093298974014647080053335398584081326147908
900836215656606273647865648167645341209013078324157417639821636085975567
917955629068041421626815387128566622738358224787337388732094438255805250
929260882674270723253417075276989820463863119513464888022561483492557031
932398225835400556700612293027531482722235350719970437543116013550385171
940991596306059089790128395141860082180703704267564776310887173789040129
955093733300266956224769927761235084243834866547095979725938722811719233
964248878077698826165734136791809752643092040987357180799765367125564239
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
974676486273630039301731204366924020235275573065554353203180984762588684
980323745430554176328290816173498829890183366007840606216957874257990730
998659181482526676158314972900538953476036491994371697548043794637028672
1003853401715949648728865680437723956012918865376273819636511322573956947
1011328416834741519202724670366650077022878021795160207270765179727545960
1022122400370304200388396648894284158317564273956390441548491978227724120
1039905247887810856493366279864326579383287341421324130590977603584891983
1040019950993986033845287890946246972157484675017763844331112577113111059
1056057806483287333786707936987109853927186312378061244488978197040195419
1069124018312174410192918163692013121154288392963731865724904017901762077
1079745814229120635961132249904663321619358025915426388449019120743650813
1083524938646344757241760514692571248951924868719244636607114583044052639
1093898046600117591190046743278607794039298978389514532378680388935137005
1101075052745387498736558959537318929559062394041319859960704082981283126
1113624727850739582067337800638357105184724524922234278071174569610418327
1127099466986997447243801174421591139220724543223692370009232336543859685
1139963894702114631814881386804319062852506020168515188598478210026582525
1147205515774438579980510419769392599303544215608347143989907702296544247
1152403865541857885583538835593991379035112013243952076210224678327532286
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
1167497614631359082822139470915481285430166090737588736782002689712157676
1173555371628701892643663407911789034855359803924101236270777868957821059
1183567612797514348297642010263449292092155588075464458581953317862973344
1197481567523729852318302446635949892420633588428596107648701643499467662
1204524688048651735098991060898943606332819685599922149475505583469871800
1217650947069863784179711910496728857597966324661008354728819725897530761
1221968990332043152135897248111883906263312524960734979178336925735272862
1234275135318975136153754955081590033780778275079547821252127463755497775
1241194622681450451396457517594195804544067587165884045557963453327153464
1259651848688209961109048396571778386786464143422154546717723648695455580
1263572658643768698462239098360837250592265231076027805822467604429407672
1273973742750489687053684864649523840494317654021358901055790441930883623
1289702566492561770404980312480282640843591755286534149044954958125653214
1296281408075097885635076787515915450949295858305986030883896609614218354
1302557822950152278329141737795999619171845868997069424290015518009282118
1316876369576889914966246704633625662500481733239142715264069695704854149
1321790000813643616072588974610059970930666264511152844323347786034260388
1337194460227635517110905624438365825426160321166340335404571461090907346
1345468493691851326439929182443241449155226787933410730374484295137609559
1352594627623112586520452832508802227305554995217379185744292767032660251
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
1360135339318449614497291766902625607306606518261889383448319159776475091
1379908388191276629911357174355719988413951710575396161448749090732280960
1385202145406379457523424327869782264487779237539992663898548861805196314
1397875547744437768309803225142818363755984133005221258781149054650204472
1402528269106591801625722810436091222425643898843114985423325813437470873
1411195994202027438684779725518111299876844053205423653891702263863284776
1421164413792981900142430078281975558305197477146844022016792040686294451
1430630797912681105981295448432443126288880130401645843813894164248233939
1447628575714895859929652640465185548690754889321993757119232515561923679
1455532207589396006086663007200076681984164611318142960676428077828629015
1467801790928902209250383375269867439930885885672916972816533571542886932
1474476843342206962633143140697448292824988942374613877426390395559612655
1482510019336146058660351680976782426102464865637481260069716741547847642
1498361246623628768519707824913925831737726846284222110268206921569929167
1504134781666829616041371020836580241533322660369871246795163082841305417
1513812851178750961360916110735334256459870293996783491055899175109689011
1526095000455732549606750717138780321678274658633701191283339149130349326
1539052417320298329611875792253262294024904805233892891374555979756738914
1544989718278671603940897056435178442659650202471929302019147900285205819
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
1551849499209810601948865596597874793991225987684368800438055480944382897
1566537372984301713774170203940948726130056581247013318936021385937209075
1574065312843042137092595360557747643961768528178115152188319405427349032
1585163822238563444458783231503177454107719254724160277318151455751507633
1596974299303625788357530337074885194184316420674969228616291010301373449
1605801274072674887458047992694825862417106475381345222620242604015574716
1611834819855428870827943206470774263745606000112066214642499849450956380
1625961409193580642908644385457407075205663490812696057454992642414274648
1637568828630392105289762748726589805967202727890186913496146638764589713
1641394177802691017006135460345761541281304587573549894830755210060397701
1659665686420964758645854463964195541741375768861900866877359345980100497
1660336382042159421454938324870941906967590110876857022591652328591591499
1677036608390691552549613240574079140749160073793444494567660353891865708
1684787036605129271604353257937965272173120480257607495605674224164614557
1697950477606227613051828373738983055076684773663227604690616676479866276
1704696774841509231533937755989954016289561691994225711647476034877997907
1711455850769354445903087516481930294500922481890472421263208929295590251
1721244025057011323861128135680891095410097542430646050856654357937753890
1733229329938686531049798919465877770199119931656778817638230972146836992
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
1741064021875724627798156550559998731069643451240034925748008449683130651
1754761523169395715697220628217885173633133726963260541569761489154421121
1761694811128716247275449084963032783045817678671806287453172267290471474
1772125861092666347033592448173548343249608665200644259664204203920169549
1781507248853151783073047481484904143625015499322047453821510157984132405
1799915457412098586567170655714796352031357569680672934465706850418425250
1800875961089237063299435426366666398898844375330826927021458862283578451
1816732357839611146219247547580236463034886987777544295592061414509673117
1820925513986848342076472206893933454893438887306180578955189524643658740
1833630474712003741010785061692288196992216035683963081869528245093727954
1841588467429866124736710242448801206044309466295510566793931730048013311
1851874532031353030813433925030445992995418049175759624878617339283464454
1867293440086882926572838300423236406898373489710904959943365380597682118
1871762602944209105766280181385792458090529523035043629401578248603981062
1882711761399509674139981636166631563479717946965924025543979896330690946
1899399518678900126364585701852696226368331003897257115210207694468696176
1906739289421096238072562620841628736829560005198454508004245264125214521
1910491012261375668001621245693775042085658552636866778770045331451318099
1928145320283799295983923875132454680874124747033576995588210143707839170
1931948075790485392370315537488850286186587745396522790799587899625702708
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
1942145613162746088101155512074819355172189762619120689941151323773859284
1958940620912461897637625538872122074812831571663502616817791211892940628
1960986607414559771641901101693431330594302807035791036933577714254603003
1978654124681234211352128000949170742357523972346395142876468290978158160
1981041496941062057222257167839020746069507106000885807685444726422027040
1994994206683414795898256288428539219620632620457881235895518518353410689
2002399568882422912337424299270246746094783409371696126053787494670421983
- 121 -
Table of Random Sampling Numbers
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))Q
Line/Col. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
- 122 -
APPENDIX B
Definitions
Note: The definitions listed below pertain to this manual only.
20.Absolute Value - disregards the sign
of a number; considers all numbers
positive.
26.Adult - An individual who is not a
minor child (See Section 419 of Act.)
27.AFDC - Aid to Families with
Dependent Children.
21.ACF - the Administration for
Children and Families.
22.Act - Social Security Act
23.Adequate Sample - pertains most
commonly to the size of a sample; a
sample is adequate if its size is large
enough to give the degree of precision or
reliability required in a given sample
estimate.
24.Adjusted State Family Assistance
Grant, or Adjusted SFAG, - the SFAG
amount, minus any reductions for Tribal
Family Assistance Grants paid to Tribal
grantees on behalf of Indian families
residing in the State and any transfers to
the Social Services Block Grant or the
Child Care and Development Block
Grant.
25.Administrative Costs has the meaning
specified at §263.01(b) of the final rule.
28.Aid to Families with Dependent
Children - the welfare program in effect
under title IV-A of prior law.
29.Alpha - the allowable probability
associated with observed differences
attributed to chance. If the probability
associated with sample differences is less
than alpha, we can reasonably conclude
that a real difference between samples
exists (see Risk).
30.Annual Sample Period - (also called
Fiscal Year) - The 12-month period,
October 1 through September 30.
31.Annual Work Participation Rate - the
overall (or two-parent) work
participation rate for a fiscal year is the
average of the State's overall (or twoparent) work participation rates for each
month in the fiscal year.
- 123 -
32.Application - The action by which an
individual indicates in writing to the
agency administering the State TANF
program (or separate State program)
his/her desire to receive assistance.
33.Assistance - The term "assistance"
includes cash, payments, vouchers, and
other forms of benefits designed to meet
a family's ongoing basic needs (i.e., for
food, clothing, shelter, utilities,
household goods, personal care items,
and general incidental expenses). It
includes such benefits even when they
are provided in the form of payments by
a TANF agency, or other agency on its
behalf, to individual recipients and
conditioned on their participation in
work experience, community service, or
other work activities (See §261.30 of
final rule).
The term "assistance" excludes:
a.Nonrecurrent, short-term benefits (such
as payments for rent deposits or
appliance repairs) that:
1)Are designed to deal with a specific
crisis situation or episode of need;
2)Are not intended to meet recurrent or
ongoing needs; and
3)Will not extend beyond four months.
b.Work subsidies (i.e., payments to
employers or third parties to help cover
the costs of employee wages, benefits,
supervision, and training);
c.Supportive services such as child care
and transportation provided to families
who are employed;
d.Refundable earned income tax credits;
e.Contributions to, and distributions
from, Individual Development Accounts;
f.Services such as counseling, case
management, peer support, child care
information and referral, transitional
services, job retention, job advancement,
and other employment-related services
that do not provide basic income
support; and
g.Transportation benefits provided under
an Access to Jobs or Reverse Commute
project, pursuant to section 404(k) of the
Act, to an individual who is not
otherwise receiving assistance.
The exclusion of nonrecurrent, shortterm benefits under (1) of this paragraph
also covers supportive services for
recently employed families, for
temporary periods of unemployment, in
order to enable continuity in their service
arrangements.
34.Basic MOE means the expenditure of
State funds that must be made in order to
meet the MOE requirement at section
409(a)(7) of the Act
35.Bias - systematic error, leading to
distortion in one direction of a statistical
result; distinct from random error, where
distortion in both directions may be
largely self-canceling.
36.Caseload - is comprised of the
members of the "target" population. For
example, the State's TANF caseload is
the families (cases) receiving assistance
under the State TANF Program. The
caseload size is the number of such
families.
- 124 -
37.Cash assistance - when provided to
participants in the Welfare-to-Work
program (WtW), has the meaning
specified at §260.32.
38.CCDBG - means the Child Care and
Development Block Grant Act of 1990,
as amended, 42 U.S.C. 9858 et. seq.
39.CCDF - means the Child Care and
Development Fund, or those child care
programs and services funded either
under section 418(a) of the Act or
CCDBG.
40.Child - (also called Minor Child) An
individual who has not attained 18 years
of age; or has not attained 19 years of
age and is a full time student in a
secondary school (or in the equivalent
level of vocational or technical training).
41.Child Care/ Family Care Services Services that assist an individual in
meeting his/her family care needs during
participation. Family care ranges from
day care inside or outside the home to
after school programs inside or outside
the home. It usually includes
supervision and shelter. It may also
include meals and transportation.
42.Closed Case (TANF) - A case
(family) whose assistance under the State
TANF Program was terminated for the
reporting month (does not include
families whose assistance was
temporarily suspended), but received
assistance under the State's TANF
Program in the prior month. Thus,
TANF eligible families that are
transferred to a separate State program
for the reporting month are considered
closed cases for reporting purposes in the
State TANF Program.
43.Closed Case (State MOE) - A case
(family) whose assistance under the
separate State program was terminated
for the reporting month (does not include
families whose assistance was
temporarily suspended), but received
assistance under the separate State
program in the prior month. Thus,
TANF eligible families that are
transferred to a State TANF Program
from a separate State programs for the
reporting month are considered closed
cases for reporting purposes in the
separate State program.
44.Commingled State TANF
Expenditures - means expenditures of
State funds that are made within the
TANF program and commingled with
Federal TANF funds.
45.Completed Case - A case for which
the State (or Tribe) collects all required
disaggregated data and reports the
information to DHHS.
46.Complete and Accurate Report for
Disaggregated Data - a report that -a.The reported data accurately reflect
information available to the State in its
case records, financial records, and
automated data systems;
- 125 -
b.The data are free from computational
errors and are internally consistent (e.g.,
items that should add to totals do so);
c.The data are reported for all elements
(i.e., no data are missing);
d.1) The data are provided for all
families; or
2) If the State opts to use sampling, for
all families selected in a sample that
meets the specifications and procedures
in the TANF Sampling Manual for
minimum sample size requirements
(except for families listed-in-error); and
e.Where estimates are required (e.g.,
some types of assistance may require
cost estimates), the State uses reasonable
methods to develop these estimates.
47.Complete and Accurate Report for
Aggregated Data - a report that -a.The reported data accurately reflect
information available to the State in its
case records, financial records, and
automated data systems;
b.The data are free from computational
errors and are internally consistent (e.g.,
items that should add to totals do so);
c.The data are reported for all applicable
elements; and
d.Monthly totals are unduplicated counts
for all families (e.g., the number of
families and the number of out-ofwedlock births are unduplicated counts).
48.Complete and Accurate Report for the
TANF Financial Data - a report that -a.The reported data accurately reflect
information available to the State in its
case records, financial records, and
automated data systems;
b.The data are free from computational
errors and are internally consistent (e.g.,
items that should add to totals do so);
c.The data are reported for all applicable
elements; and;
d.All expenditures have been made in
accordance with §92.20(a) of the Code
of Federal Regulations.
49.Confidence Interval - the interval
between two sample values, known as
confidence limits, within which it may
be asserted with a specified degree of
confidence that the true population value
lies.
50.Confidence Limits - the values that
form the upper and lower limits of the
confidence interval.
51.Contingency Fund - means Federal
TANF funds available under section
403(b) of the Act, and Contingency
Funds means the Federal monies made
available to States under that section.
Neither term includes any State funds
expended pursuant to section 403(b).
52.Contingency Fund MOE - means the
MOE expenditures that a State must
make in order to meet the MOE
requirements at sections 403(b)(6) and
409(a)(10) of the Act and subpart B of
part 264 of the regulations and to retain
the contingency funds made available to
the State. The only expenditures that
qualify for Contingency Fund MOE are
State TANF expenditures.
- 126 -
53.Control group is a term relevant to
continuation of a "waiver" and has the
meaning specified at §260.71.
54.Countable State expenditures has the
meaning specified at §264.0.
55.DHHS - U.S. Department of Health
and Human Services
56.Discretionary Fund of the CCDF
refers to child care funds appropriated
under the CCDBG.
57.Disposed of Case - A case (family)
for which the data was collected and
reported to DHHS or the case was
reported as dropped, listed-in-error.
58.Disabled Individual - An individual
who has a physical or mental impairment
that substantially limits one or more of
the major life activities of such an
individual, who has a record of such
impairment, or who is regarded as
having such an impairment.
59.DV Waiver (or Federally recognized
good cause domestic violence waiver)
has the meaning specified in §260.51.
60.EA - Emergency Assistance.
61.Earned Income - Income in cash or
in-kind earned by an individual through
the receipt of wages, salary,
commissions or profit from activities in
which he/she is engaged as a selfemployed individual or as an employee.
62.Earned Income Credit (EIC) - A
refundable tax credit for families with
dependent children. EIC payments are
received either monthly (as advance
payment through the employer), annually
(as a refund from IRS), or both.
63.Eligible State - means a State that,
during the 27-month period ending with
the close of the first quarter of the fiscal
year, has submitted a TANF plan that we
have determined is complete.
64.Emergency Assistance - the program
option available to States under sections
403(a)(5) and 406(e) of prior law to
provide short-term assistance to needy
families with children.
65.Employed - An individual who is
currently a paid employee; works in
his/her own business, profession, or
farm; works 15 hours or more per week
as an unpaid worker in an enterprise
operated by a member of the family; or is
one who is not working, but has a job or
business from which he/she is
temporarily absent because of illness,
bad weather, vacation, labormanagement dispute, or personal
reasons, whether or not paid by the
employer for time off and whether or not
seeking another job. Employed also
includes active duty military.
66.Equal Probability of Selection selection of a sample where every case
has an independent and equal chance of
inclusion in the sample (also called selfweighted sample).
- 127 -
67.Expenditure means any amount of
Federal TANF or State MOE funds that a
State expends, spends, pays out, or
disburses consistent with the
requirements of parts 260 through 265 of
the regulations. It may include
expenditures on the refundable portions
of State or local tax credits, if they are
consistent with the provisions at
§260.33. It does not include any
amounts that merely represent avoided
costs or foregone revenue. Avoided
costs include such items as contractor
penalty payments for poor performance
and purchase price discounts, rebates,
and credits that a State receives.
Foregone revenue includes State tax
provisions -- such as waivers,
deductions, exemptions, or
nonrefundable tax credits -- that reduce a
State's tax revenue.
68.Experimental group is a term relevant
to continuation of a "waiver" and has the
meaning specified at §260.71.
69.Family Violence Option (or FVO) has
the meaning specified at §260.51.
70.FAMIS - Family Assistance
Management Information System - the
automated statewide management
information system under sections
402(a)(30), 402(e), and 403 of prior law.
71.Federal Expenditures - expenditures
by a State of Federal TANF funds.
72.Federal TANF Funds - means all
funds provided to the State under section
403 of the Act, including WtW funds
awarded under section 403(a)(5). The
term includes the SFAG, any bonuses,
supplemental grants, or contingency
funds.
73.Federally recognized good cause
domestic violence waiver has the
meaning specified at §260.51.
74.Fiscal Year - (also called Annual
Sample Period) - The 12-month period,
October 1 through September 30.
75.Frame - the list of cases from which
the sample is actually selected; also
known as the sample selection list.
76.FY - fiscal year.
77.Good cause domestic violence waiver
has the meaning specified at §260.51.
78.Governor - the Chief Executive
Officer of the State. It thus includes the
Governor of each of the 50 States and
the Territories and the Mayor of the
District of Columbia.
79.Housing Assistance - Services that
assist individuals in maintaining or
obtaining adequate shelter for themselves
and their families while they are
receiving employment, training or other
supportive services.
80.IEVS - the Income and Eligibility
Verification System operated pursuant to
the provisions in section 1137 of the Act.
- 128 -
81.Inconsistent is a term relevant to
continuation of a "waiver" and has the
meaning specified at §260.71.
82.Indian Tribe - has the meaning given
such terms by section 4 of the Indian
Self-Determination and Education
Assistance Act (25 U.S.C. 450b), except
that the term "Indian tribe" means, with
respect to the State of Alaska, only the
Metlakatla Indian Community of the
Annette Islands Reserve and the
following Alaska Native regional
nonprofit corporations:
a.Arctic Slope Native Association;
b.Kawerak, Inc.;
c.Maniilaq Association;
d.Association of Village Council
Presidents;
e.Tanana Chiefs Council;
f.Cook Inlet Tribal Council;
g.Bristol Bay Native Association;
h.Aleutian and Pribilof Island
Association;
i.Chugachmuit;
j.Tlingit Haida Central Council;
k.Kodiak Area Native Association; and
l.Copper River Native Association.
83.Individual Development Accounts
has the meaning specified at §263.20 of
the Act.
84.Job Opportunities and Basic Skills
Training Program - the program under
title IV-F of prior law to provide
education, training and employment
services to welfare recipients.
85.JOBS - the Job Opportunities and
Basic Skills Training Program.
86.Listed-in-error - cases included in the
sample selection list that are not included
in the population of interest.
87.Mean - a measure of the central
tendency of data. The sum of the values
divided by the number of values.
88.Medical Assistance - Medical
assistance services received by an
individual under the State plan approved
under title XIX of the Social Security
Act.
89.Minor Child - An individual who has
not attained 18 years of age; or has not
attained 19 years of age and is a full time
student in a secondary school or in the
equivalent level of vocational or
technical training.
90.MOE - maintenance-of-effort.
91.Needy State - is a term that pertains to
the provisions regarding the Contingency
Fund and the penalty for failure to meet
participation rates. It means, for a
month, a State where:
a.1) The average rate of total
unemployment (seasonally adjusted) for
the most recent 3-month period for
which data are published for all States
equals or exceeds 6.5 percent; and
2) The average rate of total
unemployment (seasonally adjusted) for
such 3-month period equals or exceeds
110 percent of the average rate for either
- 129 -
(or both) of the corresponding 3-month
periods in the two preceding calendar
years; or
b.The Secretary of Agriculture has
determined that the average number of
individuals participating in the Food
Stamp program in the State has grown at
least 10 percent in the most recent 3month period for which data are
available.
92.Noncustodial Parent - as used here,
means a parent of a minor child who: (1)
lives in the State and (2) does not live in
the same household as the minor child.
93.Non-Sampling Error - the error or
deviation from the true population value
in sample estimates that cannot be
attributed to chance sampling variations.
Examples are errors resulting from
imperfections in the selection of sample
units, bias in the estimating procedure
used, mistakes in arithmetical
calculations, inconsistent review
procedures, etc.
94.Normal Distribution - a symmetrical,
bell shaped curve that describes the
sampling distribution of many common
sample statistics. While the sampling
distributions of proportions and ratios as
used in TANF are more correctly
described by the binomial distribution,
they are often closely approximated by
the normal distribution, and it is common
practice to use the normal distribution for
this purpose. The normal distribution
provides the theoretical basis for the
determination of confidence limits, for
the specification of particular levels or
degrees of confidence involved in
making sample estimates, and in
evaluating sampling error.
95.Not in Labor Force - An individual
who is classified as neither employed nor
unemployed.
96.Oversampling - selecting more
sample cases than required.
97.Overall Monthly Work Participation
Rate - (also known as All Families Work
Participation Rate) - The State's overall
participation rate for a month is defined
as follows:
a.The number of families receiving
TANF assistance that include an adult or
a minor head-of-household who is
engaged in work for the month (the
numerator), divided by
b.The number of families receiving
TANF assistance during the month that
include an adult or a minor head-ofhousehold minus the number of families
that are subject to a penalty for refusing
to work in that month (the denominator).
However, if a family has been sanctioned
for more than three of the last 12 months,
we will not deduct it from the
denominator. A State has the option of
not requiring a single custodial parent
caring for a child under age one to
engage in work. If the State adopts this
option, it may disregard such a family in
the participation rate calculation for a
maximum of 12 months.
- 130 -
98.Parameter - a value, property, or
characteristic of a population, which can
normally be estimated from a sample.
Examples are a mean, proportion or
percentage, total, range, or standard
deviation of a population.
99.Population of Interest - those units
about which we wish to form
conclusions from which a sample is
selected and estimates made.
100.Precision - see definition for
Reliability. The degree to which a
sample estimate approximates the value
obtained from a complete count of all
units using the same methods.
101.Prior law - means the provisions of
title IV-A and IV-F of the Social
Security Act in effect as of August 21,
1996. They include provisions related to
Aid to Families with Dependent Children
(or AFDC), Emergency Assistance (or
EA), Job Opportunities and Basic Skills
Training (or JOBS), and Family
Assistance Management Information
System (FAMIS).
102.Probability - relative frequency of
occurrence; the probability of an event is
the relative frequency of occurrence of
the event in an indefinitely large number
of observations.
103.Probability Sampling - any method
of sample selection that is based on the
theory of probability. Probability
sampling, which requires that at any
stage of selection the probability of any
unit or set of units being selected must be
known, is the only general method of
sampling that makes it possible to obtain
a mathematical measure of the precision
of the sample estimate. The term
"random sampling" is used in the sense
of probability sampling.
104.PRWORA - the Personal
Responsibility and Work Opportunity
Reconciliation Act of 1996, or Public
Law 104-193.
105.Qualified Aliens has the meaning
prescribed under section 431 of
PRWORA, as amended, 8 U.S.C. 1641.
106.Qualified State Expenditures means the total amount of State funds
expended during the fiscal year that
count for basic MOE purposes. It
includes expenditures, under any State
program, for any of the following with
respect to eligible families:
a.Cash assistance;
b.Child care assistance;
c.Educational activities designed to
increase self-sufficiency, job training,
and work, excluding any expenditure for
public education in the State except
expenditures involving the provision of
services or assistance of an eligible
family that is not generally available to
persons who are not members of an
eligible family;
d.Any other use of funds allowable under
subpart A of part 263 of the regulations;
and
e.Administrative costs in connection
with the matters described in paragraphs
- 131 -
(1), (2), (3) and (4) of this definition, but
only to the extent that such costs do not
exceed 15 percent of the total amount of
qualified State expenditures for the fiscal
year.
112.Reporting Month - the specific
calendar or fiscal month for which data
is being collected. The reporting month
and the sample month are always the
same month.
107.Random Numbers - series of digits,
each occurring independently of each
other. Each digit tends to appear as
many times as any other, in any
progression, if the series selected is
large.
113.Risk - as used here, refers to the
degree of risk associated with given
degrees of confidence. For example, if a
statement is made "with 95 percent
confidence" that the true population
parameter lies within a specified interval,
there is a "5 percent risk" that the
parameter actually lies outside that
interval (also called alpha).
108.Random Sampling - the process of
selecting a sample from a population so
that every unit in the population has a
known chance of being included in the
sample.
109.Random Start - in selecting a
systematic sample at intervals of some
specified number of items in an ordered
frame, it is mandatory to select the first
item completely without bias. Such
selection is then said to have given the
sample "a random start."
110.Range - the largest minus the
smallest of a group of values.
111.Reliability See definition of
Precision - the uniformity of sample
results when obtained from repeated
samples of the same size and type from
the sample population; the degree to
which a sample estimate approximates
the value obtained from a complete
count of all units using the same
methods.
114.Sample - part of a population; a
limited or finite number of items selected
from a population, by a prescribed
procedure, with the objective of
estimating certain values (mean, total
proportion, etc.) of the parent population,
or of testing the validity of certain
assumptions or hypotheses with respect
to particular properties of the population.
115.Sample Interval - in systematic
sampling, the number of cases between
two consecutive selections on the
sampling frame.
116.Sample Month - the specific
calendar or fiscal month for which the
sample is selected. The sample month
and the reporting month are always the
same month.
117.Sample Period - the 12 month period
October 1 through September 30.
- 132 -
118.Sample Selection List - the list of
cases from which the sample is actually
selected; also known as the sample
frame.
119.Sample Size - the number of items in
the sample.
120.Sampling Distribution - the
distribution of a (sample) statistic, such
as a sample mean or a sample proportion
or percentage, that would be formed by
obtaining such statistics from all possible
samples of a given fixed size selected by
some specified sampling procedure; a
population of all possible sample values
of the statistic under consideration.
121.Sampling Error - that part of the
difference between a population value,
and an estimate of that value obtained
from a random sample, which is due
solely to the fact that only a sample of
values is observed; to be distinguished
from non-sampling error which is due to
biased or imperfect sample selection, or
real differences due to changes over
time, error of observation, recording,
calculation, etc.
122.Scientifically Acceptable Sampling
Method - a probability sampling method
in which every sampling unit from the
population has a known, non-zero
chance to be included in the sample, and
the sample size requirements are met.
123.Secretary - Secretary of the
Department of Health and Human
Services or any other Department official
duly authorized to act on the Secretary's
behalf.
124.Segregated State TANF
Expenditures - means expenditures of
State funds within the TANF program
that are not commingled with Federal
TANF funds.
125.Separate State Program - means a
program operated outside of TANF in
which the expenditures of State funds
may count for basic MOE purposes.
126.SFAG - State Family Assistance
Grant.
127.SFAG Payable - means the SFAG
amount, reduced, as appropriate, for any
Tribal Family Assistance Grants made
on behalf of Indian families residing in
the State and any penalties imposed on a
State.
128.Significant Difference - a difference
is statistically significant if it can be
concluded from a sample, with a given
degree of risk, that the difference
actually exists in the universe. A
difference observed in a sample is judged
not statistically significant if it could
easily have occurred purely as a result of
random sampling variations.
129.Simple Random Sample - a
probability sample selected in such a
way that each unit of the frame has an
equal and independent chance of being
included in the sample; for samples of
any given size, all possible combinations
- 133 -
of units that could form samples of that
size must have the same probability of
selection (usually uses random digits for
item selection).
130.Single audit - means an audit or
supplementary review conducted under
the authority of the Single Audit Act at
31 U.S.C. chapter 75.
131.Social Services Block Grant means
the social services program operated
under title XX of the Act, pursuant to 42
U.S.C. 1397.
132.SSBG means the Social Services
Block Grant.
133.Standard Deviation - the most
widely used measure of the dispersion
(scatter or variability) of frequency
distributions from their arithmetic
means. The standard deviation of the
sampling distribution of any given
statistic is also known as the "standard
error" of that statistic.
134.Standard Error - the standard
deviation of the sampling distribution of
a given statistic; used in measuring
precision of an estimate.
135.State - the 50 States of the United
States, the District of Columbia, the
Commonwealth of Puerto Rico, the
United States Virgin Islands, Guam, and
American Samoa, unless otherwise
specified.
136.State agency - means the agency that
the Governor certifies as the
administering and supervising agency for
the TANF program, pursuant to section
402(a)(4) of the Act.
137.State Family Assistance Grant means the amount of the basic block
grant allocated to each eligible State
under the formula at section 403(a)(1) of
the Act.
138.State MOE Expenditures - means
the expenditure of State funds that may
count for purposes of the basic MOE
requirements at section 409(a)(7) of the
Act and the Contingency Fund MOE
requirements at sections 403(b)(4) and
409(a)(10) of the Act.
139.State MOE Family - For reporting
purposes only, the State MOE family is
the eligible family receiving assistance
plus the following persons living in the
household if they are not already in the
eligible family receiving assistance:
a.a parent or caretaker relative of any
minor child in the eligible family
receiving assistance,
b.a minor sibling of any child in the
eligible family receiving assistance, and
c.any person whose income or resources
are counted in determining the eligibility
for or the amount of the assistance for
the eligible family.
140.State TANF Expenditures - means
the expenditure of State funds within the
TANF program.
- 134 -
141.Stratified Random Sampling random sampling of a population that
has been divided in a number of subpopulations according to some
predetermined criterion (geographic
location, characteristic, etc.). The
percentage size of each sample must be
equal or have individual weighting
factors taken into account before the subpopulation sample results can be
combined.
142.Stratum - a segment of the
population for which separate estimates
are computed for some special reason.
All strata must be combined if an
estimate of the total population is to be
made.
143.Subsidized Child Care - A benefit
provided by the government to a parent
to support, in part or whole, the cost of
child care services provided by an
eligible provider to an eligible child.
144.Subsidized Housing - Money paid
by the government or through a private
social service agency to the family or to
the owner of the housing to assist the
family in paying rent.
145.Supplemental Case - a case added to
the caseload for the review month after
the regular sample frame, (i.e., the
payroll listing or master file listing) has
been compiled for the monthly sample
selection.
assistance provided to the family for one
or more months without removing the
family from the eligible rolls.
147.Systematic Sample - a sample
attained by selecting from a file, list or
computer tape, individual items at
equally spaced intervals (as every 10th,
140th, 850th, etc. item, as required to
obtain a total sample of a given size),
with the starting point within the first
such interval being determined by
random selection.
148.TANF - Temporary Assistance for
Needy Families.
149.TANF Family - For reporting
purposes only, the TANF family is the
eligible family receiving assistance plus
the following persons living in the
household if they are not already in the
eligible family receiving assistance:
a.a parent or caretaker relative of any
minor child in the eligible family
receiving assistance,
b.a minor sibling of any child in the
eligible family receiving assistance, and
c.any person whose income or resources
are counted in determining the eligibility
for or the amount of assistance of the
eligible family.
150.TANF Funds - all funds provided to
the State under section 403 of the Act,
including the SFAG, any bonuses,
supplemental grants, or contingency
funds, except Welfare to Work funds.
146.Suspended Case - a formalized
agency action that results in no
- 135 -
151.TANF MOE - the expenditure of
State funds that must be made in order to
meet the MOE requirement at section
409(a)(7) of the Act.
152.TANF Program - a State program of
family assistance operated by an
"eligible State" under its State TANF
plan.
153.Teen Parent - A teen parent is a
person who is under 20 years of age and
whose child is also a member of the
TANF family.
154.Territories - the Commonwealth of
Puerto Rico, the United States Virgin
Islands, Guam, and American Samoa.
155.Title IV-A - refers to the title and
part of the Act that now includes TANF,
but previously included AFDC and EA.
For the purpose of the TANF program
regulations, this term does not include
child care programs authorized and
funded under section 418 of the Act, or
their predecessors, unless we specify
otherwise.
156.Tolerance - the proportion of error
that has been determined to be
acceptable.
157.Transportation - Services that ensure
mobility between home and the location
of employment, training, or other
supportive services.
158.Tribal Family Assistance Grant means a grant paid to a Tribe that has an
approved Tribal family assistance plan
under section 412(a)(1) of the Act.
159.Tribal grantee means a Tribe that
receives Federal TANF funds to operate
a Tribal TANF program under section
412(a) of the Act.
160.Tribal TANF program - means a
TANF program developed by an eligible
Tribe, Tribal organization, or consortium
and approved by us under section 412 of
the Act.
161.Tribe - means Indian Tribe or Tribal
organization, as defined elsewhere in this
section. The definition may include
Tribal consortia (i.e., groups of federally
recognized Tribes or Alaska Native
entities that have banded together in a
formal arrangement to develop and
administer a Tribal TANF program).
162.Two-Parent Monthly Work
Participation Rate - The two-parent
participation rate for a fiscal year is the
average of the State's two-parent
participation rate for each month in the
fiscal year. We determine a State's twoparent participation rate for a month as
follows:
a.The number of two-parent families
receiving TANF assistance in which the
adults or minor child head-of-household
and other parent meet the requirements
set forth in §261.32 for the month (the
numerator), divided by
b.The number of two-parent families
receiving TANF assistance during the
month minus the number of two-parent
- 136 -
families`that are subject to a penalty for
refusing to work in that month (the
denominator). However, if a family has
been sanctioned for more than three of
the last 12 months, we will not deduct it
from the denominator.
163.Unemployed - An individual who is
not employed, who is available for work,
and who has made specific efforts to find
a job within the prior four weeks.
Included as unemployed are those who
are not working, are available for work,
and are waiting to be called back to a job
from which they were laid off.
164.Unearned Income - Cash payment or
in-kind contributions or benefits from
government agencies, private
organizations or individuals.
165.Victim of domestic violence has the
meaning specified at §260.51.
166.Waiver - has the meaning specified
at §260.71. (It is distinguished from the
domestic violence waiver.)
167.Welfare-to-Work - means the
program for funding work activities at
section 403(a)(5) of the Act.
168.WtW - Welfare-to-Work.
169.WtW cash assistance has the
meaning specified at §260.32.
- 137 -
- 138 -
APPENDIX C
Standard Error of Percentages Based on Selected Sample Sizes
The following table provides a quick reference of the standard error of percentages of case
error rates obtained from a specified sample size. For example, if a simple random sample of
800 cases is found to have 80 errors, the case error rate would be 10 percent and the standard
error would be 1.1 percent.
The entries in the table are estimated by the following equation for a normal distribution:
p (1 & p)
n
Sp '
where:
p=estimated proportion of error cases; and
n=sample size.
The 95 percent confidence interval can be obtained by multiplying the standard error by 1.96.
In the above example, the 95 percent confidence interval would be 1.96 x 1.1% or
approximately ± 2.2%.
Note that the table provides only approximate standard errors. The approximation is good
when the sample fraction n / N is small. When the fraction is large, the standard error given
in the table is overestimated by a factor of
.
1
N & n
N & 1
- 139 -
- 140 -
STANDARD ERROR OF PERCENTAGES BASED ON SELECTED SAMPLE SIZES
))))))))))))))))))))))))))))))))))))))))))))))))))))))
Sample Size or Base of Estimated Percentage
)))))))))))))))))))))))))))))))))))))))))))))))))
Est.
Percent5080100150200250300350400500600700800900100012001500
))))))))))))))))))))))))))))))))))))))))))))))))))))))
11.41.11.00.80.70.60.60.50.50.40.40.40.40.30.30.30.3
22.01.61.41.11.00.90.80.70.70.60.60.50.50.50.40.4 0.4
32.41.91.71.41.21.11.00.90.90.80.70.60.60.60.50.5 0.4
42.82.22.01.61.41.21.11.01.00.90.80.70.70.70.60.60.5
53.12.42.21.81.51.41.31.21.11.00.90.80.80.70.70.60.6
63.42.72.41.91.71.51.41.31.21.11.00.90.80.80.80.70.6
73.62.92.62.11.81.61.51.41.31.11.01.00.90.90.8 0.70.7
83.83.02.72.21.91.71.61.51.41.21.11.01.00.90.90.80.7
94.03.22.92.32.01.81.71.51.41.31.21.11.01.00.90.80.7
104.23.43.02.42.11.91.71.61.51.31.21.11.11.00.90.90.8
114.43.53.12.62.22.01.81.71.61.41.31.21.11.01.00.90.8
124.63.63.22.72.32.11.91.71.61.51.31.21.11.11.00.90.8
134.83.83.42.72.42.11.91.81.71.51.41.31.21.11.11.00.9
144.93.93.52.82.52.22.01.91.71.61.41.31.21.21.11.00.9
155.04.03.62.92.52.32.11.91.81.61.51.31.31.21.11.00.9
205.74.54.03.32.82.52.32.12.01.81.61.51.41.31.31.21.0
256.14.84.33.53.12.72.52.32.21.91.81.61.51.41.41.31.1
306.55.14.63.73.22.92.62.42.32.01.91.71.61.51.41.31.2
356.75.34.83.93.43.02.82.52.42.11.91.81.71.61.51.41.2
406.95.54.94.03.53.12.82.62.42.22.01.91.71.61.51.41.3
457.05.65.04.13.53.12.92.72.52:22.01.91.81.71.61.41.3
507.15.65.04.13.53.22.92.72.52.22.01.91.81.71.61.41.3
)))))))))))))))))))))))))))))))))))))))))))))))))))))))
- 142 -
APPENDIX D
TANF SAMPLE PLAN Guidance
TANF sampling plan requirements are detailed in the TANF Manual, Sampling and Statistical Methods
(Sections 1300, 1400, and 1500). Sampling plans for the active cases (including newly approved
applicants) and the closed cases are required. The plans must conform to principles of probability
sampling, i.e., each case in the population must have a known, non-zero probability of selection and
computational methods of estimation must lead to a unique estimate. More specifically the plan must
describe the following:
CLXX. Sample Frame - Documentation of methods for constructing and maintaining the sample
frame(s)., including assessment of frame completeness and any potential problems associated with
using the sample frame(s). The plan must explicitly describe the following sample frame
characteristics:
CLXX. Date(s) when the sample cases (both regular and supplemental, if applicable) for the sample
month are selected, e.g., first workday of the month following the sample month).
Comments:
CLXX. Source, components, accuracy and completeness of the sample frame in relation to the total
caseload; if not accurate or complete, explanation of why not and how the State (Tribe) plans to correct
for the problems with the sample frame.
Comments:
- 143 -
CLXX. Procedures for ensuring that the sample frame contains complete coverage of the applicable
caseload (e.g., active TANF sample frame includes all families receiving assistance under the State's or
Tribe's TANF Program, including all newly approved applicants for the sample month and closed
TANF sample frame includes all families no longer receiving assistance under the State's TANF
Program, i.e., assistance terminated effective for the sample month).
Comments:
CLXX. Whether or not the frame is constructed by combining more than one list; if more than one list,
an explanation of how the lists are identified and how duplication of cases on lists are prevented.
Comments:
CLXX. How the frame is compiled, e.g., whether the frame is compiled entirely in the State office,
entirely in local offices, in the State office based on information supplied by local offices, etc.
Comments:
CLXX. Form of the frame, e.g., a computer file, microfilm, or hard copy; if parts of the frame are in
different forms, specifications for each part.
Comments:
CLXX. Frequency and length of delays and method used in updating the frame or its sources.
Comments:
- 144 -
CLXX. Procedures for estimating the proportion of sample cases for which the State (Tribe) will not be
able to collect and report case record information (e.g., dropped as listed-in-error because the family
did not receive TANF assistance for the reporting month).
Comments:
CLXX. Methods of locating and deleting "listed-in-error" cases from the frame.
Comments:
CLXX. Structure of the frame, i.e., the order of cases within each list and the data elements on the
frame, including definitions of coded values.
Comments:
CLXX. Treatment of special populations under TANF (e.g., individuals under a tribal family assistance
plan, a non-custodial parent who participates in work activities).
Comments:
CLXX. Criteria for stratifying sample (if applicable).
Comments:
CLXXI. Sample Selection Procedures - The sampling plan must describe in detail the procedures for
selecting the sample cases. The plan must explicitly describe the following characteristics:
- 145 -
CLXXI. Procedures for estimation of caseload size.
Comments:
CLXXI. Procedures for determination of an appropriate allowance for cases that might be dropped
from the sample for acceptable reasons.
Comments:
CLXXI. Procedures for determining the required sample size and indication of the sample size.
Comments:
CLXXI. If stratified sample design, procedures for sample allocation.
Comments:
CLXXI. Procedures for the computation of sample intervals and the determination of random starts
(systematic random sampling or stratified systematic random sampling), if applicable.
Comments:
CLXXI. Application of selection procedures to identify sample cases.
Comments:
CLXXI. Procedures to compensate for excessive oversampling or undersampling.
- 146 -
Comments:
CLXXI. Time schedule for each step in the sampling procedure.
Comments:
CLXXI. Relationship, if appropriate, to sampling frames for other programs (e.g., Welfare-to-Work).
Comments:
CLXXII. Additional Sampling Plan Information
CLXXII. Treatment of any special cases or circumstances unique to the State or Tribe.
Comments:
CLXXII. Documentation of methods for estimating proportions and their sampling errors, including the
computation of weights where appropriate.
Comments:
- 147 -
- 148 -
APPENDIX E
TRIBAL CODES FOR THE TANF PROGRAM
This list of codes for tribes is based on the Federal Register: November 13, 1996 (Volume 61, Number
220), Notices, Page 58211-58216. From the Federal Register Online via GPO Access
[wais.access.gpo.gov]
Alaska Non-Profit Associations:
NOTE: The "Special Rules for Indian Tribes in Alaska" specify that "...only the Metlakatla Indian
Community, Annette Island Reserve, and..." Alaska native regional corporations are eligible.
However, the 'corporations' listed are actually the non-profit arms of the corporations
CODE Alaska Non-Profit Association
801Metlakatla Indian Community, Annette Island Reserve, Alaska
802Artic Slope Native Association
803Kawerak, Inc.
804Maniilag Association
805Association of Village Council Presidents
806Tananana Chiefs Conference
807Cook Inlet Tribal Council
808Bristol Bay Native Association
809Aleutian and Pribolof Islan Association
810Chugachmuit
811Tlinget Haida Central Council
812Kodiak Area Native Association
813Copper River Native Association
Consortiums: Codes for consortiums of tribal entities listed below can be obtained by contacting
John Bushman
Director, Division of Tribal Services
Office of Community Services, Administration for Children and Families
370 L'Enfant Promenade, Washington, D.C. 20447-0001
Phone - Voice (202) 401-2418;
Fax (202) 401-5718;
E-mail - JBushman@acf.dhhs.gov
- 149 -
CODES FOR TRIBAL TANF PROGRAMS (Cont.)
All Others:
Below are the codes for Indian entities in the contiguous 48 states which are Federally recognized and
eligible to establish a tribal TANF porgram or participate in a consortium of Tribes for a Tribal TANF
program. All three digits are to be used (for example, ‘001’ not ‘1’).
CODETRIBAL ENTITY
001Absentee-Shawnee Tribe of Indians of Oklahoma
002Agua Caliente Band of Cahuilla Indians of the Agua Caliente Indian Reservation, California
003Ak Chin Indian Community of Papago Indians of the Maricopa, Ak Chin Reservation, Arizona
004Alabama and Coushatta Tribes of Texas
005Alabama-Quassarte Tribal Town of the Creek Nation of Oklahoma
006Alturas Indian Rancheria of Pit River Indians of California
007Apache Tribe of Oklahoma
008Arapahoe Tribe of the Wind River Reservation, Wyoming
009Aroostook Band of Micmac Indians of Maine
010Assiniboine and Sioux Tribes of the Fort Peck Indian Reservation, Montana
011Augustine Band of Cahuilla Mission Indians of the Augustine Reservation, California
012Bad River Band of the Lake Superior Tribe of Chippewa Indians of the Bad River Reservation,
Wisconsin
013Bay Mills Indian Community of the Sault Ste. Marie Band of Chippewa Indians, Bay Mills
Reservation, Michigan
014Bear River Band of the Rohnerville Rancheria of California
015Berry Creek Rancheria of Maidu Indians of California
016Big Lagoon Rancheria of Smith River Indians of California
017Big Pine Band of Owens Valley Paiute Shoshone Indians of the Big Pine Reservation, California
018Big Sandy Rancheria of Mono Indians of California
019Big Valley Rancheria of Pomo & Pit River Indians of California
020Blackfeet Tribe of the Blackfeet Indian Reservation of Montana
021Blue Lake Rancheria of California
022Bridgeport Paiute Indian Colony of California
023Buena Vista Rancheria of Me-Wuk Indians of California
024Burns Paiute Tribe of the Burns Paiute Indian Colony of Oregon
025Cabazon Band of Cahuilla Mission Indians of the Cabazon Reservation,California
026Cachil DeHe Band of Wintun Indians of the Colusa Indian Community of the Colusa Rancheria,
California
027Caddo Indian Tribe of Oklahoma
028Cahuilla Band of Mission Indians of the Cahuilla Reservation, California
029Cahto Indian Tribe of the Laytonville Rancheria, California
030Campo Band of Diegueno Mission Indians of the Campo Indian Reservation, California
- 150 -
CODE
TRIBAL ENTITY
031Barona Group of Capitan Grande Band of Diegueno Mission Indians of the Barona Reservation,
California
032Viejas (Baron Long) Group of Capitan Grande Band of Diegueno Mission Indians of the Viejas
Reservation, California
033Catawba Tribe of South Carolina
034Cayuga Nation of New York
035Cedarville Rancheria of Northern Paiute Indians of California
036Chemehuevi Indian Tribe of the Chemehuevi Reservation, California
037Cher-Ae Heights Indian Community of the Trinidad Rancheria, California
038Cherokee Nation of Oklahoma
039Cheyenne-Arapaho Tribes of Oklahoma
040Cheyenne River Sioux Tribe of the Cheyenne River Reservation, South Dakota
041Chickasaw Nation of Oklahoma
042Chicken Ranch Rancheria of Me-Wuk Indians of California
043Chippewa-Cree Indians of the Rocky Boy's Reservation, Montana
044Chitimacha Tribe of Louisiana
045Choctaw Nation of Oklahoma
046Citizen Potawatomi Nation, Oklahoma
047Cloverdale Rancheria of Pomo Indians of California
048Coast Indian Community of Yurok Indians of the Resighini Rancheria, California
049Cocopah Tribe of Arizona
050Coeur D'Alene Tribe of the Coeur D'Alene Reservation, Idaho
051Cold Springs Rancheria of Mono Indians of California
052Colorado River Indian Tribes of the Colorado River Indian Reservation, Arizona and California
053Comanche Indian Tribe, Oklahoma
054Confederated Salish & Kootenai Tribes of the Flathead Reservation, Montana
055Confederated Tribes of the Chehalis Reservation, Washington
056Confederated Tribes of the Colville Reservation, Washington
057Confederated Tribes of the Coos, Lower Umpqua and Siuslaw Indians of Oregon
058Confederated Tribes of the Goshute Reservation, Nevada and Utah
059Confederated Tribes of the Grand Ronde Community of Oregon
060Confederated Tribes of the Siletz Reservation, Oregon
061Confederated Tribes of the Umatilla Reservation, Oregon
062Confederated Tribes of the Warm Springs Reservation of Oregon
063Confederated Tribes and Bands of the Yakama Indian Nation of the Yakama Reservation,
Washington
064Coquille Tribe of Oregon
065Cortina Indian Rancheria of Wintun Indians of California
066Coushatta Tribe of Louisiana
067Cow Creek Band of Umpqua Indians of Oregon
068Coyote Valley Band of Pomo Indians of California
069Crow Tribe of Montana
070Crow Creek Sioux Tribe of the Crow Creek Reservation, South Dakota
- 151 -
CODE
TRIBAL ENTITY
071Cuyapaipe Community of Diegueno Mission Indians of the Cuyapaipe Reservation, California
072Death Valley Timbi-Sha Shoshone Band of California
073Delaware Tribe of Indians, Oklahoma
074Delaware Tribe of Western Oklahoma
075Devils Lake Sioux Tribe of the Devils Lake Sioux Reservation, North Dakota
076Dry Creek Rancheria of Pomo Indians of California
077Duckwater Shoshone Tribe of the Duckwater Reservation, Nevada
078Eastern Band of Cherokee Indians of North Carolina
079Eastern Shawnee Tribe of Oklahoma
080Elem Indian Colony of Pomo Indians of the Sulphur Bank Rancheria, California
081Elk Valley Rancheria of California
082Ely Shoshone Tribe of Nevada
083Enterprise Rancheria of Maidu Indians of California
084Flandreau Santee Sioux Tribe of South Dakota
085Forest County Potawatomi Community of Wisconsin Potawatomi Indians, Wisconsin
086Fort Belknap Indian Community of the Fort Belknap Reservation of Montana
087Fort Bidwell Indian Community of Paiute Indians of the Fort Bidwell Reservation, California
088Fort Independence Indian Community of Paiute Indians of the Fort Independence Reservation,
California
089Fort McDermitt Paiute and Shoshone Tribes of the Fort McDermitt Indian Reservation, Nevada
090Fort McDowell Mohave-Apache Indian Community of the Fort McDowell Indian Reservation,
Arizona
091Fort Mojave Indian Tribe of Arizona, California & Nevada
092Fort Sill Apache Tribe of Oklahoma
093Gila River Pima-Maricopa Indian Community of the Gila River Indian Reservation of Arizona
094Grand Traverse Band of Ottawa & Chippewa Indians of Michigan
095Greenville Rancheria of Maidu Indians of California
096Grindstone Indian Rancheria of Wintun-Wailaki Indians of California
097Guidiville Rancheria of California
098Hannahville Indian Community of Wisconsin Potawatomie Indians of Michigan
099Havasupai Tribe of the Havasupai Reservation, Arizona
100Ho-Chunk Nation of Wisconsin - (formerly known as the Wisconsin Winnebago Tribe)
101Hoh Indian Tribe of the Hoh Indian Reservation, Washington
102Hoopa Valley Tribe of the Hoopa Valley Reservation, California
103Hopi Tribe of Arizona
104Hopland Band of Pomo Indians of the Hopland Rancheria, California
105Houlton Band of Maliseet Indians of Maine
106Hualapai Indian Tribe of the Hualapai Indian Reservation, Arizona
107Huron Potawatomi, Inc., Michigan
108Inaja Band of Diegueno Mission Indians of the Inaja and Cosmit
109Reservation, California
110Ione Band of Miwok Indians of California
111Iowa Tribe of Kansas and Nebraska
- 152 -
CODE
TRIBAL ENTITY
112Iowa Tribe of Oklahoma
113Jackson Rancheria of Me-Wuk Indians of California
114Jamestown Klallam Tribe of Washington
115Jamul Indian Village of California
116Jena Band of Choctaw Indians, Louisiana
117Jicarilla Apache Tribe of the Jicarilla Apache Indian Reservation, New Mexico
118Kaibab Band of Paiute Indians of the Kaibab Indian Reservation, Arizona
119Kalispel Indian Community of the Kalispel Reservation, Washington
120Karuk Tribe of California
121Kashia Band of Pomo Indians of the Stewarts Point Rancheria, California
122Kaw Nation, Oklahoma
123Keweenaw Bay Indian Community of L'Anse and Ontonagon Bands of Chippewa Indians of the
L'Anse Reservation, Michigan
124Kialegee Tribal Town of the Creek Indian Nation of Oklahoma
125Kickapoo Tribe of Indians of the Kickapoo Reservation in Kansas
126Kickapoo Tribe of Oklahoma
127Kickapoo Traditional Tribe of Texas
128Kiowa Indian Tribe of Oklahoma
129Klamath Indian Tribe of Oregon
130Kootenai Tribe of Idaho
131La Jolla Band of Luiseno Mission Indians of the La Jolla Reservation, California
132La Posta Band of Diegueno Mission Indians of the La Posta Indian Reservation, California
133La Courte Oreilles Band of Lake Superior Chippewa Indians of the Lac Courte Oreilles
Reservation of Wisconsin
134Lac du Flambeau Band of Lake Superior Chippewa Indians of the Lac du Flambeau Reservation of
Wisconsin
135Lac Vieux Desert Band of Lake Superior Chippewa Indians of Michigan
136Las Vegas Tribe of Paiute Indians of the Las Vegas Indian Colony, Nevada
137Little River Band of Ottawa Indians of Michigan
138Little Traverse Bay Bands of Odawa Indians of Michigan
139Los Coyotes Band of Cahuilla Mission Indians of the Los Coyotes Reservation, California
140Lovelock Paiute Tribe of the Lovelock Indian Colony, Nevada
141Lower Brule Sioux Tribe of the Lower Brule Reservation, South Dakota
142Lower Elwha Tribe of the Lower Elwha Reservation, Washington
143Lower Sioux Indian Community of Minnesota Mdewakanton Sioux Indians of the Lower Sioux
Reservation in Minnesota
144Lummi Tribe of the Lummi Reservation, Washington
145Lytton Rancheria of California
146Makah Indian Tribe of the Makah Indian Reservation, Washington
147Manchester Band of Pomo Indians of the Manchester-Point Arena Rancheria, California
148Manzanita Band of Diegueno Mission Indians of the Manzanita Reservation, California
149Mashantucket Pequot Tribe of Connecticut
150Mechoopda Indian Tribe of Chico Rancheria, California
- 153 -
CODE
TRIBAL ENTITY
151Menominee Indian Tribe of Wisconsin
152Mesa Grande Band of Diegueno Mission Indians of the Mesa Grande Reservation, California
153Mescalero Apache Tribe of the Mescalero Reservation, New Mexico
154Miami Tribe of Oklahoma
155Miccosukee Tribe of Indians of Florida
156Middletown Rancheria of Pomo Indians of California
157Minnesota Chippewa Tribe, Minnesota
(All six component reservations: Bois Forte Band (Nett Lake); Fond du Lac Band; Grand Portage
Band; Leech Lake Band; Mille Lacs Band; White Earth Band)
158Mississippi Band of Choctaw Indians, Mississippi
159Moapa Band of Paiute Indians of the Moapa River Indian Reservation, Nevada
160Modoc Tribe of Oklahoma
161Mohegan Indian Tribe of Connecticut
162Mooretown Rancheria of Maidu Indians of California
163Morongo Band of Cahuilla Mission Indians of the Morongo Reservation, California
164Muckleshoot Indian Tribe of the Muckleshoot Reservation, Washington
165Muscogee (Creek) Nation of Oklahoma
166Narragansett Indian Tribe of Rhode Island
167Navajo Nation of Arizona, New Mexico & Utah
168Nez Perce Tribe of Idaho
169Nisqually Indian Tribe of the Nisqually Reservation, Washington
170Nooksack Indian Tribe of Washington
171Northern Cheyenne Tribe of the Northern Cheyenne Indian Reservation, Montana
172Northfork Rancheria of Mono Indians of California
173Northwestern Band of Shoshoni Nation of Utah (Washakie)
174Oglala Sioux Tribe of the Pine Ridge Reservation, South Dakota
175Omaha Tribe of Nebraska
176Oneida Nation of New York
177Oneida Tribe of Wisconsin
178Onondaga Nation of New York
179Osage Nation of Oklahoma
180Ottawa Tribe of Oklahoma
181Otoe-Missouria Tribe of Indians, Oklahoma
182Paiute Indian Tribe of Utah
183Paiute-Shoshone Indians of the Bishop Community of the Bishop Colony, California
184Paiute-Shoshone Tribe of the Fallon Reservation and Colony, Nevada
185Paiute-Shoshone Indians of the Lone Pine Community of the Lone Pine
186Reservation, California
187Pala Band of Luiseno Mission Indians of the Pala Reservation,California
188Pascua Yaqui Tribe of Arizona
189Paskenta Band of Nomlaki Indians of California
190Passamaquoddy Tribe of Maine
191Pauma Band of Luiseno Mission Indians of the Pauma & Yuima Reservation, California
- 154 -
CODE
TRIBAL ENTITY
192Pawnee Indian Tribe of Oklahoma
193Pechanga Band of Luiseno Mission Indians of the Pechanga Reservation, California
194Penobscot Tribe of Maine
195Peoria Tribe of Oklahoma
196Picayune Rancheria of Chukchansi Indians of California
197Pinoleville Rancheria of Pomo Indians of California
198Pit River Tribe of California (includes Big Bend, Lookout, Montgomery Creek & Roaring Creek
Rancherias & XL Ranch)
199Poarch Band of Creek Indians of Alabama
200Pokagon Band of Potawatomi Indians of Michigan
201Ponca Tribe of Indians of Oklahoma
202Ponca Tribe of Nebraska
203Port Gamble Indian Community of the Port Gamble Reservation, Washington
204Potter Valley Rancheria of Pomo Indians of California
205Prairie Band of Potawatomi Indians, Kansas
206Prairie Island Indian Community of Minnesota Mdewakanton Sioux Indians of the Prairie Island
Reservation, Minnesota
207Pueblo of Acoma, New Mexico
208Pueblo of Cochiti, New Mexico
209Pueblo of Jemez, New Mexico
210Pueblo of Isleta, New Mexico
211Pueblo of Laguna, New Mexico
212Pueblo of Nambe, New Mexico
213Pueblo of Picuris, New Mexico
214Pueblo of Pojoaque, New Mexico
215Pueblo of San Felipe, New Mexico
216Pueblo of San Juan, New Mexico
217Pueblo of San Ildefonso, New Mexico
218Pueblo of Sandia, New Mexico
219Pueblo of Santa Ana, New Mexico
220Pueblo of Santa Clara, New Mexico
221Pueblo of Santo Domingo, New Mexico
222Pueblo of Taos, New Mexico
223Pueblo of Tesuque, New Mexico
224Pueblo of Zia, New Mexico
225Puyallup Tribe of the Puyallup Reservation, Washington
226Pyramid Lake Paiute Tribe of the Pyramid Lake Reservation, Nevada
227Quapaw Tribe of Oklahoma
228Quartz Valley Indian Community of the Quartz Valley Reservation of California
229Quechan Tribe of the Fort Yuma Indian Reservation, California & Arizona
230Quileute Tribe of the Quileute Reservation, Washington
231Quinault Tribe of the Quinault Reservation, Washington
232Ramona Band or Village of Cahuilla Mission Indians of California
- 155 -
CODE
TRIBAL ENTITY
233Red Cliff Band of Lake Superior Chippewa Indians of Wisconsin
234Red Lake Band of Chippewa Indians of the Red Lake Reservation, Minnesota
235Redding Rancheria of California
236Redwood Valley Rancheria of Pomo Indians of California
237Reno-Sparks Indian Colony, Nevada
238Rincon Band of Luiseno Mission Indians of the Rincon Reservation, California
239Robinson Rancheria of Pomo Indians of California
240Rosebud Sioux Tribe of the Rosebud Indian Reservation, South Dakota
241Round Valley Indian Tribes of the Round Valley Reservation, California - (formerly known as the
Covelo Indian Community)
243Rumsey Indian Rancheria of Wintun Indians of California
244Sac & Fox Tribe of the Mississippi in Iowa
245Sac & Fox Nation of Missouri in Kansas and Nebraska
246Sac & Fox Nation, Oklahoma
247Saginaw Chippewa Indian Tribe of Michigan, Isabella Reservation
248Salt River Pima-Maricopa Indian Community of the Salt River Reservation, Arizona
249Samish Indian Tribe
250San Carlos Apache Tribe of the San Carlos Reservation, Arizona
251San Juan Southern Paiute Tribe of Arizona
252San Manual Band of Serrano Mission Indians of the San Manual Reservation, California
253San Pasqual Band of Diegueno Mission Indians of California
254Santa Rosa Indian Community of the Santa Rosa Rancheria, California
255Santa Rosa Band of Cahuilla Mission Indians of the Santa Rosa Reservation, California
256Santa Ynez Band of Chumash Mission Indians of the Santa Ynez
257Reservation, California
258Santa Ysabel Band of Diegueno Mission Indians of the Santa Ysabel Reservation, California
259Santee Sioux Tribe of the Santee Reservation of Nebraska
260Sauk-Suiattle Indian Tribe of Washington
261Sault Ste. Marie Tribe of Chippewa Indians of Michigan
262Scotts Valley Band of Pomo Indians of California
263Seminole Nation of Oklahoma
264Seminole Tribe of Florida, Dania, Big Cypress & Brighton Reservations
265Seneca Nation of New York
266Seneca-Cayuga Tribe of Oklahoma
267Shakopee Mdewakanton Sioux Community of Minnesota (Prior Lake)
268Sheep Ranch Rancheria of Me-Wuk Indians of California
269Sherwood Valley Rancheria of Pomo Indians of California
270Shingle Springs Band of Miwok Indians, Shingle Springs Rancheria (Verona Tract), California
271Shoalwater Bay Tribe of the Shoalwater Bay Indian Reservation, Washington
272Shoshone Tribe of the Wind River Reservation, Wyoming
273Shoshone-Bannock Tribes of the Fort Hall Reservation of Idaho
274Shoshone-Paiute Tribes of the Duck Valley Reservation, Nevada
275Sisseton-Wahpeton Sioux Tribe of the Lake Traverse Reservation, South Dakota
- 156 -
CODE
TRIBAL ENTITY
276Skokomish Indian Tribe of the Skokomish Reservation, Washington
277Skull Valley Band of Goshute Indians of Utah
278Smith River Rancheria of California
279Soboba Band of Luiseno Mission Indians of the Soboba Reservation, California
280Sokaogon Chippewa Community of the Mole Lake Band of Chippewa Indians, Wisconsin
281Southern Ute Indian Tribe of the Southern Ute Reservation, Colorado
282Spokane Tribe of the Spokane Reservation, Washington
283Squaxin Island Tribe of the Squaxin Island Reservation, Washington
284St. Croix Chippewa Indians of Wisconsin, St. Croix Reservation
285St. Regis Band of Mohawk Indians of New York
286Standing Rock Sioux Tribe of North & South Dakota
287Stockbridge-Munsee Community of Mohican Indians of Wisconsin
288Stillaguamish Tribe of Washington
289Summit Lake Paiute Tribe of Nevada
290Suquamish Indian Tribe of the Port Madison Reservation, Washington
291Susanville Indian Rancheria of Paiute, Maidu, Pit River & Washoe Indians of California
292Swinomish Indians of the Swinomish Reservation, Washington
293Sycuan Band of Diegueno Mission Indians of California
294Table Bluff Rancheria of Wiyot Indians of California
295Table Mountain Rancheria of California
296Te-Moak Tribes of Western Shoshone Indians of Nevada
297Thlopthlocco Tribal Town of the Creek Nation of Oklahoma
298Three Affiliated Tribes of the Fort Berthold Reservation, North Dakota
299Tohono O'odham Nation of Arizona - (formerly known as the Papago Tribe of the Sells, Gila Bend
& San Xavier Reservation, Arizona)
300Tonawanda Band of Seneca Indians of New York
301Tonkawa Tribe of Indians of Oklahoma
302Tonto Apache Tribe of Arizona
303Torres-Martinez Band of Cahuilla Mission Indians of California
304Tule River Indian Tribe of the Tule River Reservation, California
305Tulalip Tribes of the Tulalip Reservation, Washington
306Tunica-Biloxi Indian Tribe of Louisiana
307Tuolumne Band of Me-Wuk Indians of the Tuolumne Rancheria of California
308Turtle Mountain Band of Chippewa Indians of North Dakota
309Tuscarora Nation of New York
310Twenty-Nine Palms Band of Luiseno Mission Indians of California
311United Auburn Indian Community of the Auburn Rancheria of California
312United Keetoowah Band of Cherokee Indians of Oklahoma
313Upper Lake Band of Pomo Indians of Upper Lake Rancheria of California
314Upper Sioux Indian Community of the Upper Sioux Reservation, Minnesota
315Upper Skagit Indian Tribe of Washington
316Ute Indian Tribe of the Uintah & Ouray Reservation, Utah
317Ute Mountain Tribe of the Ute Mountain Reservation, Colorado, New Mexico & Utah
- 157 -
CODE
TRIBAL ENTITY
318Utu Utu Gwaitu Paiute Tribe of the Benton Paiute Reservation, California
319Walker River Paiute Tribe of the Walker River Reservation, Nevada
320Wampanoag Tribe of Gay Head (Aquinnah) of Massachusetts
321Washoe Tribe of Nevada & California (Carson Colony, Dresslerville & Washoe Ranches)
322White Mountain Apache Tribe of the Fort Apache Reservation, Arizona
323Wichita and Affiliated Tribes (Wichita, Keechi, Waco & Tawakonie), Oklahoma
324Winnebago Tribe of Nebraska
325Winnemucca Indian Colony of Nevada
326Wyandotte Tribe of Oklahoma
327Yankton Sioux Tribe of South Dakota
328Yavapai-Apache Nation of the Camp Verde Indian Reservation, Arizona
329Yavapai-Prescott Tribe of the Yavapai Reservation, Arizona
330Yerington Paiute Tribe of the Yerington Colony & Campbell Ranch, Nevada
331Yomba Shoshone Tribe of the Yomba Reservation, Nevada
332Ysleta Del Sur Pueblo of Texas
333Yurok Tribe of the Yurok Reservation, California
334Zuni Tribe of the Zuni Reservation, New Mexico
- 158 -
CODE
TRIBAL ENTITY
APPENDIX F
FIPS County Codes - Alphabetical List
Note that FIPS county codes are unique within state. You will usually need to pre-append the 2-digit
FIPS state code to form a complete FIPS county code.
ALABAMA - 01
001 =Autauga
003 =Baldwin
005 =Barbour
007 =Bibb
009 =Blount
011 =Bullock
013 =Butler
015 =Calhoun
017 =Chambers
019 =Cherokee
021 =Chilton
023 =Choctaw
025 =Clarke
027 =Clay
029 =Cleburne
031 =Coffee
033 =Colbert
035 =Conecuh
037 =Coosa
039 =Covington
041 =Crenshaw
043 =Cullman
045 =Dale
047 =Dalla
049 =DeKalb
051 =Elmore
053 =Escambia
055 =Etowah
057 =Fayette
059 =Franklin
061 =Geneva
063 =Greene
065 =Hale
067 =Henry
069 =Houston
071 =Jackson
073 =Jefferson
075 =Lamar
077 =Lauderdale
079 =Lawrence
081 =Lee
083 =Limestone
085 =Lowndes
087 =Macon
089 =Madison
091 =Marengo
093 =Marion
095 =Marshall
097 =Mobile
099 =Monroe
101 =Montgomery
103 =Morgan
105 =Perry
107 =Pickens
109 =Pike
111 =Randolph
113 =Russell
115 =St. Clair
117 =Shelby
119 =Sumter
121 =Talladega
123 =Tallapoosa
- 159 -
125 =Tuscaloosa
127 =Walker
129 =Washington
131 =Wilcox
133 =Winston
ALASKA - 02
013 =Aleutians East
016 =Aleutians West Census
020 =Anchorage
050 =Bethel Census
060 =Bristol Bay
070 =Dillingham Census
090 =Fairbanks North Star
100 =Haines
110 =Juneau
122 =Kenai Peninsula
130 =Ketchikan Gateway
150 =Kodiak Island
164 =Lake and Peninsula
170 =Matanuska-Susitna
180 =Nome Census
185 =North Slope
188 =Northwest Arctic
201 =Prince of Wales-Outer
Ketchikan Census
220 =Sitka
231 =
Skagway-Yakutat-Angoon
Census
240 =Southeast Fairbanks
Census
261 =Valdez-Cordova
Census
270 =Wade Hampton
Census
280 =Wrangell-Petersburg
Census
290 =Yukon-Koyukuk
Census
ARIZONA - 04
001 =Apache
003 =Cochise
005 =Coconino
007 =Gila
009 =Graham
011 =Greenlee
012 =La Paz
013 =Maricopa
015 =Mohave
017 =Navajo
019 =Pima
021 =Pinal
023 =Santa Cruz
025 =Yavapai
027 =Yuma
ARKANSAS - 05
001 =Arkansas
003 =Ashley
005 =Baxter
007 =Benton
009 =Boone
011 =Bradley
013 =Calhoun
015 =Carroll
017 =Chicot
019 =Clark
021 =Clay
023 =Cleburne
025 =Cleveland
027 =Columbia
029 =Conway
031 =Craighead
033 =Crawford
035 =Crittenden
037 =Cross
039 =Dallas
041 =Desha
043 =Drew
045 =Faulkner
047 =Franklin
049 =Fulton
051 =Garland
053 =Grant
055 =Greene
057 =Hempstead
059 =Hot Spring
061 =Howard
063 =Independence
065 =Izard
067 =Jackson
069 =Jefferson
071 =Johnson
073 =Lafayette
075 =Lawrence
077 =Lee
079 =Lincoln
081 =Little River
083 =Logan
085 =Lonoke
087 =Madison
089 =Marion
091 =Miller
093 =Mississippi
095 =Monroe
097 =Montgomery
099 =Nevada
101 =Newton
103 =Ouachita
105 =Perry
107 =Phillips
109 =Pike
- 160 -
111 =Poinsett
113 =Polk
115 =Pope
117 =Prairie
119 =Pulaski
121 =Randolph
123 =St. Francis
125 =Saline
127 =Scott
129 =Searcy
131 =Sebastian
133 =Sevier
135 =Sharp
137 =Stone
139 =Union
141 =Van Buren
143 =Washington
145 =White
147 =Woodruff
149 =Yell
CALIFORNIA - 06
001 =Alameda
003 =Alpine
005 =Amador
007 =Butte
009 =Calaveras
011 =Colusa
013 =Contra Costa
015 =Del Norte
017 =El Dorado
019 =Fresno
021 =Glenn
023 =Humboldt
025 =Imperial
027 =Inyo
029 =Kern
031 =Kings
033 =Lake
035 =Lassen
037 =Los Angeles
039 =Madera
041 =Marin
043 =Mariposa
045 =Mendocino
047 =Merced
049 =Modoc
051 =Mono
053 =Monterey
055 =Napa
057 =Nevada
059 =Orange
061 =Placer
063 =Plumas
065 =Riverside
067 =Sacramento
069 =San Benito
071 =San Bernardino
073 =San Diego
075 =San Francisco
077 =San Joaquin
079 =San Luis Obispo
081 =San Mateo
083 =Santa Barbara
085 =Santa Clara
087 =Santa Cruz
089 =Shasta
091 =Sierra
093 =Siskiyou
095 =Solano
097 =Sonoma
099 =Stanislaus
101 =Sutter
103 =Tehama
105 =Trinity
107 =Tular
109 =Tuolumne
111 =Ventura
113 =Yolo
115 =Yuba
COLORADO - 08
001 =Adams
003 =Alamosa
005 =Arapahoe
007 =Archuleta
009 =Baca
011 =Bent
013 =Boulder
015 =Chaffee
017 =Cheyenne
019 =Clear Creek
021 =Conejos
023 =Costilla
025 =Crowley
027 =Custer
029 =Delta
031 =Denver
033 =Dolores
035 =Douglas
037 =Eagle
039 =Elbert
041 =El Paso
043 =Fremont
045 =Garfield
047 =Gilpin
049 =Grand
051 =Gunnison
053 =Hinsdale
055 =Huerfano
057 =Jackson
059 =Jefferson
061 =Kiowa
063 =Kit Carson
065 =Lake
067 =La Plata
069 =Larimer
071 =Las Animas
073 =Lincoln
075 =Logan
077 =Mesa
079 =Mineral
081 =Moffat
083 =Montezuma
085 =Montrose
087 =Morgan
089 =Otero
091 =Ouray
- 161 -
093 =Park
095 =Phillips
097 =Pitkin
099 =Prowers
101 =Pueblo
103 =Rio Blanco
105 =Rio Grande
107 =Routt
109 =Saguache
111 =San Juan
113 =San Miguel
115 =Sedgwick
117 =Summit
119 =Teller
121 =Washington
123 =Weld
125 =Yuma
CONNECTICUT - 09
001 =Fairfield
003 =Hartford
005 =Litchfield
007 =Middlesex
009 =New Haven
011 =New London
013 =Tolland
015 =Windham
DELAWARE - 10
001 =Kent
003 =New Castle
005 =Sussex
DIST. OF COL. - 11
001 =District of Columbia
FLORIDA - 12
001 =Alachua
003 =Baker
005 =Bay
007 =Bradford
009 =Brevard
011 =Broward
013 =Calhoun
015 =Charlotte
017 =Citrus
019 =Clay
021 =Collier
023 =Columbia
025 =Dade
027 =DeSoto
029 =Dixie
031 =Duval
033 =Escambia
035 =Flagler
037 =Franklin
039 =Gadsden
041 =Gilchrist
043 =Glades
045 =Gulf
047 =Hamilton
049 =Hardee
051 =Hendry
053 =Hernando
055 =Highlands
057 =Hillsborough
059 =Holmes
061 =Indian River
063 =Jackson
065 =Jefferson
067 =Lafayette
069 =Lake
071 =Lee
073 =Leon
075 =Levy
077 =Liberty
079 =Madison
081 =Manatee
083 =Marion
085 =Martin
087 =Monroe
089 =Nassau
091 =Okaloosa
093 =Okeechobee
095 =Orange
097 =Osceola
099 =Palm Beach
101 =Pasco
103 =Pinellas
105 =Polk
107 =Putnam
109 =St. Johns
111 =St. Lucie
113 =Santa Rosa
115 =Sarasota
117 =Seminole
119 =Sumter
121 =Suwannee
123 =Taylor
125 =Union
127 =Volusia
129 =Wakulla
131 =Walton
133 =Washington
GEORGIA - 13
001 =Appling
003 =Atkinson
005 =Bacon
007 =Baker
009 =Baldwin
011 =Banks
013 =Barrow
015 =Bartow
017 =Ben Hill
019 =Berrien
021 =Bibb
023 =Bleckley
025 =Brantley
027 =Brooks
029 =Bryan
031 =Bulloch
033 =Burke
- 162 -
035 =Butts
037 =Calhoun
039 =Camden
043 =Candler
045 =Carroll
047 =Catoosa
049 =Charlton
051 =Chatham
053 =Chattahoochee
055 =Chattooga
057 =Cherokee
059 =Clarke
061 =Clay
063 =Clayton
065 =Clinch
067 =Cobb
069 =Coffee
071 =Colquitt
073 =Columbia
075 =Cook
077 =Coweta
079 =Crawford
081 =Crisp
083 =Dade
085 =Dawson
087 =Decatur
089 =DeKalb
091 =Dodge
093 =Dooly
095 =Dougherty
097 =Douglas
099 =Early
101 =Echols
103 =Effingham
105 =Elbert
107 =Emanuel
109 =Evans
111 =Fannin
113 =Fayette
115 =Floyd
117 =Forsyth
119 =Franklin
121 =Fulton
123 =Gilmer
125 =Glascock
127 =Glynn
129 =Gordon
131 =Grady
133 =Greene
135 =Gwinnett
137 =Habersham
139 =Hall
141 =Hancock
143 =Haralson
145 =Harris
147 =Hart
149 =Heard
151 =Henry
153 =Houston
155 =Irwin
157 =Jackson
159 =Jasper
161 =Jeff Davis
163 =Jefferson
165 =Jenkins
167 =Johnson
169 =Jones
171 =Lamar
173 =Lanier
175 =Laurens
177 =Lee
179 =Liberty
181 =Lincoln
183 =Long
185 =Lowndes
187 =Lumpkin
189 =McDuffie
191 =McIntosh
193 =Macon
195 =Madison
197 =Marion
199 =Meriwether
201 =Miller
205 =Mitchell
207 =Monroe
209 =Montgomery
211 =Morgan
213 =Murray
215 =Muscogee
217 =Newton
219 =Oconee
221 =Oglethorpe
223 =Paulding
225 =Peach
227 =Pickens
229 =Pierce
231 =Pike
233 =Polk
235 =Pulaski
237 =Putnam
239 =Quitman
241 =Rabun
243 =Randolph
245 =Richmond
247 =Rockdale
249 =Schley
251 =Screven
253 =Seminole
255 =Spalding
257 =Stephens
259 =Stewart
261 =Sumter
263 =Talbot
265 =Taliaferro
267 =Tattnall
269 =Taylor
271 =Telfair
273 =Terrell
275 =Thomas
277 =Tift
279 =Toombs
281 =Towns
283 =Treutlen
285 =Troup
287 =Turner
289 =Twiggs
291 =Union
293 =Upson
295 =Walker
297 =Walton
299 =Ware
301 =Warren
- 163 -
303 =Washington
305 =Wayne
307 =Webster
309 =Wheeler
311 =White
313 =Whitfield
315 =Wilcox
317 =Wilkes
319 =Wilkinson
321 =Worth
HAWAII - 15
001 =Hawaii
003 =Honolulu
005 =Kalawao
007 =Kauai
009 =Maui
IDAHO - 16
001 =Ada
003 =Adams
005 =Bannock
007 =Bear Lake
009 =Benewah
011 =Bingham
013 =Blaine
015 =Boise
017 =Bonner
019 =Bonneville
021 =Boundary
023 =Butte
025 =Camas
027 =Canyon
029 =Caribou
031 =Cassia
033 =Clark
035 =Clearwater
037 =Custer
039 =Elmore
041 =Franklin
043 =Fremont
045 =Gem
047 =Gooding
049 =Idaho
051 =Jefferson
053 =Jerome
055 =Kootenai
057 =Latah
059 =Lemhi
061 =Lewis
063 =Lincoln
065 =Madison
067 =Minidoka
069 =Nez Perce
071 =Oneida
073 =Owyhee
075 =Payette
077 =Power
079 =Shoshone
081 =Teton
083 =Twin Falls
085 =Valley
087 =Washington
ILLINOIS - 17
001 =Adams
003 =Alexander
005 =Bond
007 =Boone
009 =Brown
011 =Bureau
013 =Calhoun
015 =Carroll
017 =Cass
019 =Champaign
021 =Christian
023 =Clark
025 =Clay
027 =Clinton
029 =Coles
031 =Cook
033 =Crawford
035 =Cumberland
037 =DeKalb
039 =De Witt
041 =Douglas
043 =DuPage
045 =Edgar
047 =Edwards
049 =Effingham
051 =Fayette
053 =Ford
055 =Franklin
057 =Fulton
059 =Gallatin
061 =Greene
063 =Grundy
065 =Hamilton
067 =Hancock
069 =Hardin
071 =Henderson
073 =Henry
075 =Iroquois
077 =Jackson
079 =Jasper
081 =Jefferson
083 =Jersey
085 =Jo Daviess
087 =Johnson
089 =Kane
091 =Kankakee
093 =Kendall
095 =Knox
097 =Lake
099 =La Salle
101 =Lawrence
103 =Lee
105 =Livingston
107 =Logan
109 =McDonough
111 =McHenry
113 =McLean
115 =Macon
117 =Macoupin
119 =Madison
121 =Marion
- 164 -
123 =Marshall
125 =Mason
127 =Massac
129 =Menard
131 =Mercer
133 =Monroe
135 =Montgomery
137 =Morgan
139 =Moultrie
141 =Ogle
143 =Peoria
145 =Perry
147 =Piatt
149 =Pike
151 =Pope
153 =Pulaski
155 =Putnam
157 =Randolph
159 =Richland
161 =Rock Island
163 =St. Clair
165 =Saline
167 =Sangamon
169 =Schuyler
171 =Scott
173 =Shelby
175 =Stark
177 =Stephenson
179 =Tazewell
181 =Union
183 =Vermilion
185 =Wabash
187 =Warren
189 =Washington
191 =Wayne
193 =White
195 =Whiteside
197 =Will
199 =Williamson
201 =Winnebago
203 =Woodford
INDIANA - 18
001 =Adams
003 =Allen
005 =Bartholomew
007 =Benton
009 =Blackford
011 =Boone
013 =Brown
015 =Carroll
017 =Cass
019 =Clark
021 =Clay
023 =Clinton
025 =Crawford
027 =Daviess
029 =Dearborn
031 =Decatur
033 =De Kalb
035 =Delaware
037 =Dubois
039 =Elkhart
041 =Fayette
043 =Floyd
045 =Fountain
047 =Franklin
049 =Fulton
051 =Gibson
053 =Grant
055 =Greene
057 =Hamilton
059 =Hancock
061 =Harrison
063 =Hendricks
065 =Henry
067 =Howard
069 =Huntington
071 =Jackson
073 =Jasper
075 =Jay
077 =Jefferson
079 =Jennings
081 =Johnson
083 =Knox
085 =Kosciusko
087 =Lagrange
089 =Lake
091 =La Porte
093 =Lawrence
095 =Madison
097 =Marion
099 =Marshall
101 =Martin
103 =Miami
105 =Monroe
107 =Montgomery
109 =Morgan
111 =Newton
113 =Noble
115 =Ohio
117 =Orange
119 =Owen
121 =Parke
123 =Perry
125 =Pike
127 =Porter
129 =Posey
131 =Pulaski
133 =Putnam
135 =Randolph
137 =Ripley
139 =Rush
141 =St. Joseph
143 =Scott
145 =Shelby
147 =Spencer
149 =Starke
151 =Steuben
153 =Sullivan
155 =Switzerland
157 =Tippecanoe
159 =Tipton
161 =Union
163 =Vanderburgh
165 =Vermillion
167 =Vigo
169 =Wabash
171 =Warren
173 =Warrick
175 =Washington
- 165 -
177 =Wayne
179 =Wells
181 =White
183 =Whitley
IOWA - 19
001 =Adair
003 =Adams
005 =Allamakee
007 =Appanoose
009 =Audubon
011 =Benton
013 =Black Hawk
015 =Boone
017 =Bremer
019 =Buchanan
021 =Buena Vista
023 =Butler
025 =Calhoun
027 =Carroll
029 =Cass
031 =Cedar
033 =Cerro Gordo
035 =Cherokee
037 =Chickasaw
039 =Clarke
041 =Clay
043 =Clayton
045 =Clinton
047 =Crawford
049 =Dallas
051 =Davis
053 =Decatur
055 =Delaware
057 =Des Moines
059 =Dickinson
061 =Dubuque
063 =Emmet
065 =Fayette
067 =Floyd
069 =Franklin
071 =Fremont
073 =Greene
075 =Grundy
077 =Guthrie
079 =Hamilton
081 =Hancock
083 =Hardin
085 =Harrison
087 =Henry
089 =Howard
091 =Humboldt
093 =Ida
095 =Iowa
097 =Jackson
099 =Jasper
101 =Jefferson
103 =Johnson
105 =Jones
107 =Keokuk
109 =Kossuth
111 =Lee
113 =Linn
115 =Louisa
117 =Lucas
119 =Lyon
121 =Madison
123 =Mahaska
125 =Marion
127 =Marshall
129 =Mills
131 =Mitchell
133 =Monona
135 =Monroe
137 =Montgomery
139 =Muscatine
141 =O'Brien
143 =Osceola
145 =Page
147 =Palo Alto
149 =Plymouth
151 =Pocahontas
153 =Polk
155 =Pottawattamie
157 =Poweshiek
159 =Ringgold
161 =Sac
163 =Scott
165 =Shelby
167 =Sioux
169 =Story
171 =Tama
173 =Taylor
175 =Union
177 =Van Buren
179 =Wapello
181 =Warren
183 =Washington
185 =Wayne
187 =Webster
189 =Winnebago
191 =Winneshiek
193 =Woodbury
195 =Worth
197 =Wright
KANSAS - 20
001 =Allen
003 =Anderson
005 =Atchison
007 =Barber
009 =Barton
011 =Bourbon
013 =Brown
015 =Butler
017 =Chase
019 =Chautauqua
021 =Cherokee
023 =Cheyenne
025 =Clark
027 =Clay
029 =Cloud
031 =Coffey
033 =Comanche
035 =Cowley
037 =Crawford
039 =Decatur
041 =Dickinson
- 166 -
043 =Doniphan
045 =Douglas
047 =Edwards
049 =Elk
051 =Ellis
053 =Ellsworth
055 =Finney
057 =Ford
059 =Franklin
061 =Geary
063 =Gove
065 =Graham
067 =Grant
069 =Gray
071 =Greeley
073 =Greenwood
075 =Hamilton
077 =Harper
079 =Harvey
081 =Haskell
083 =Hodgeman
085 =Jackson
087 =Jefferson
089 =Jewell
091 =Johnson
093 =Kearny
095 =Kingman
097 =Kiowa
099 =Labette
101 =Lane
103 =Leavenworth
105 =Lincoln
107 =Linn
109 =Logan
111 =Lyon
113 =McPherson
115 =Marion
117 =Marshall
119 =Meade
121 =Miami
123 =Mitchell
125 =Montgomery
127 =Morris
129 =Morton
131 =Nemaha
133 =Neosho
135 =Ness
137 =Norton
139 =Osage
141 =Osborne
143 =Ottawa
145 =Pawnee
147 =Phillips
149 =Pottawatomie
151 =Pratt
153 =Rawlins
155 =Reno
157 =Republic
159 =Rice
161 =Riley
163 =Rooks
165 =Rush
167 =Russell
169 =Saline
171 =Scott
173 =Sedgwick
175 =Seward
177 =Shawnee
179 =Sheridan
181 =Sherman
183 =Smith
185 =Stafford
187 =Stanton
189 =Stevens
191 =Sumner
193 =Thomas
195 =Trego
197 =Wabaunsee
199 =Wallace
201 =Washington
203 =Wichita
205 =Wilson
207 =Woodson
209 =Wyandotte
001 =Adair
003 =Allen
005 =Anderson
007 =Ballard
009 =Barren
011 =Bath
013 =Bell
015 =Boone
017 =Bourbon
019 =Boyd
021 =Boyle
023 =Bracken
025 =Breathitt
027 =Breckinridge
029 =Bullitt
031 =Butler
033 =Caldwell
035 =Calloway
037 =Campbell
039 =Carlisle
041 =Carroll
043 =Carter
045 =Casey
047 =Christian
049 =Clark
051 =Clay
053 =Clinton
055 =Crittenden
057 =Cumberland
059 =Daviess
KENTUCKY - 21
- 167 -
061 =Edmonson
063 =Elliott
065 =Estill
067 =Fayette
069 =Fleming
071 =Floyd
073 =Franklin
075 =Fulton
077 =Gallatin
079 =Garrard
081 =Grant
083 =Graves
085 =Grayson
087 =Green
089 =Greenup
091 =Hancock
093 =Hardin
095 =Harlan
097 =Harrison
099 =Hart
101 =Henderson
103 =Henry
105 =Hickman
107 =Hopkins
109 =Jackson
111 =Jefferson
113 =Jessamine
115 =Johnson
117 =Kenton
119 =Knott
121 =Knox
123 =Larue
125 =Laurel
127 =Lawrence
129 =Lee
131 =Leslie
133 =Letcher
135 =Lewis
137 =Lincoln
139 =Livingston
141 =Logan
143 =Lyon
145 =McCracken
147 =McCreary
149 =McLean
151 =Madison
153 =Magoffin
155 =Marion
157 =Marshall
159 =Martin
161 =Mason
163 =Meade
165 =Menifee
167 =Mercer
169 =Metcalfe
171 =Monroe
173 =Montgomery
175 =Morgan
177 =Muhlenberg
179 =Nelson
181 =Nicholas
183 =Ohio
185 =Oldham
187 =Owen
189 =Owsley
191 =Pendleton
193 =Perry
195 =Pike
197 =Powell
199 =Pulaski
201 =Robertson
203 =Rockcastle
205 =Rowan
207 =Russell
209 =Scott
211 =Shelby
213 =Simpson
215 =Spencer
217 =Taylor
219 =Todd
221 =Trigg
223 =Trimble
225 =Union
227 =Warren
229 =Washington
231 =Wayne
233 =Webster
235 =Whitley
- 168 -
237 =Wolfe
239 =Woodford
LOUISIANA - 22
001 =Acadia
003 =Allen
005 =Ascension
007 =Assumption
009 =Avoyelles
011 =Beauregard
013 =Bienville
015 =Bossier
017 =Caddo
019 =Calcasieu
021 =Caldwell
023 =Cameron
025 =Catahoula
027 =Claiborne
029 =Concordia
031 =De Soto
033 =East Baton Rouge
035 =East Carroll
037 =East Feliciana
039 =Evangeline
041 =Franklin
043 =Grant
045 =Iberia
047 =Iberville
049 =Jackson
051 =Jefferson
053 =Jefferson Davis
055 =Lafayette
057 =Lafourche
059 =La Salle
061 =Lincoln
063 =Livingston
065 =Madison
067 =Morehouse
069 =Natchitoches
071 =Orleans
073 =Ouachita
075 =Plaquemines
077 =Pointe Coupee
079 =Rapides
081 =Red River
083 =Richland
085 =Sabine
087 =St. Bernard
089 =St. Charles
091 =St. Helena
093 =St. James
095 =St. John the Baptist
097 =St. Landry
099 =St. Martin
101 =St. Mary
103 =St. Tammany
105 =Tangipahoa
107 =Tensas
109 =Terrebonne
111 =Union
113 =Vermilion
115 =Vernon
117 =Washington
119 =Webster
121 =West Baton Rouge
123 =West Carroll
125 =West Feliciana
127 =Winn
MAINE - 23
001 =Androscoggin
003 =Aroostook
005 =Cumberland
007 =Franklin
009 =Hancock
011 =Kennebec
013 =Knox
015 =Lincoln
017 =Oxford
019 =Penobscot
021 =Piscataquis
023 =Sagadahoc
025 =Somerset
027 =Waldo
029 =Washington
031 =York
MARYLAND - 24
001 =Allegany
003 =Anne Arundel
005 =Baltimore
009 =Calvert
011 =Caroline
013 =Carroll
015 =Cecil
017 =Charles
019 =Dorchester
021 =Frederick
023 =Garrett
025 =Harford
027 =Howard
029 =Kent
031 =Montgomery
033 =Prince George's
035 =Queen Anne's
037 =St. Mary's
039 =Somerset
041 =Talbot
043 =Washington
045 =Wicomico
047 =Worcester
510 =Baltimore
MASSACHUSETTS - 25
001 =Barnstable
003 =Berkshire
005 =Bristol
007 =Dukes
009 =Essex
011 =Franklin
013 =Hampden
015 =Hampshire
017 =Middlesex
019 =Nantucket
- 169 -
021 =Norfolk
023 =Plymouth
025 =Suffolk
027 =Worcester
MICHIGAN - 26
001 =Alcona
003 =Alger
005 =Allegan
007 =Alpena
009 =Antrim
011 =Arenac
013 =Baraga
015 =Barry
017 =Bay
019 =Benzie
021 =Berrien
023 =Branch
025 =Calhoun
027 =Cass
029 =Charlevoix
031 =Cheboygan
033 =Chippewa
035 =Clare
037 =Clinton
039 =Crawford
041 =Delta
043 =Dickinson
045 =Eaton
047 =Emmet
049 =Genesee
051 =Gladwin
053 =Gogebic
055 =Grand Traverse
057 =Gratiot
059 =Hillsdale
061 =Houghton
063 =Huron
065 =Ingham
067 =Ionia
069 =Iosco
071 =Iron
073 =Isabella
075 =Jackson
077 =Kalamazoo
079 =Kalkaska
081 =Kent
083 =Keweenaw
085 =Lake
087 =Lapeer
089 =Leelanau
091 =Lenawee
093 =Livingston
095 =Luce
097 =Mackinac
099 =Macomb
101 =Manistee
103 =Marquette
105 =Mason
107 =Mecosta
109 =Menominee
111 =Midland
113 =Missaukee
115 =Monroe
117 =Montcalm
119 =Montmorency
121 =Muskegon
123 =Newaygo
125 =Oakland
127 =Oceana
129 =Ogemaw
131 =Ontonagon
133 =Osceola
135 =Oscoda
137 =Otsego
139 =Ottawa
141 =Presque Isle
143 =Roscommon
145 =Saginaw
147 =St. Clair
149 =St. Joseph
151 =Sanilac
153 =Schoolcraft
155 =Shiawassee
157 =Tuscola
159 =Van Buren
161 =Washtenaw
163 =Wayne
165 =Wexford
MINNESOTA - 27
001 =Aitkin
003 =Anoka
005 =Becker
007 =Beltrami
009 =Benton
011 =Big Stone
013 =Blue Earth
015 =Brown
017 =Carlton
019 =Carver
021 =Cass
023 =Chippewa
025 =Chisago
027 =Clay
029 =Clearwater
031 =Cook
033 =Cottonwood
035 =Crow Wing
037 =Dakota
039 =Dodge
041 =Douglas
043 =Faribault
045 =Fillmore
047 =Freeborn
049 =Goodhue
051 =Grant
053 =Hennepin
055 =Houston
057 =Hubbard
059 =Isanti
061 =Itasca
063 =Jackson
065 =Kanabec
067 =Kandiyohi
069 =Kittson
071 =Koochiching
073 =Lac qui Parle
- 170 -
075 =Lake
077 =Lake of the Woods
079 =Le Sueur
081 =Lincoln
083 =Lyon
085 =McLeod
087 =Mahnomen
089 =Marshall
091 =Martin
093 =Meeker
095 =Mille Lacs
097 =Morrison
099 =Mower
101 =Murray
103 =Nicollet
105 =Nobles
107 =Norman
109 =Olmsted
111 =Otter Tail
113 =Pennington
115 =Pine
117 =Pipestone
119 =Polk
121 =Pope
123 =Ramsey
125 =Red Lake
127 =Redwood
129 =Renville
131 =Rice
133 =Rock
135 =Roseau
137 =St. Louis
139 =Scott
141 =Sherburne
143 =Sibley
145 =Stearns
147 =Steele
149 =Stevens
151 =Swift
153 =Todd
155 =Traverse
157 =Wabasha
159 =Wadena
161 =Waseca
163 =Washington
165 =Watonwan
167 =Wilkin
169 =Winona
171 =Wright
173 =Yellow Medicine
MISSISSIPPI - 28
001 =Adams
003 =Alcorn
005 =Amite
007 =Attala
009 =Benton
011 =Bolivar
013 =Calhoun
015 =Carroll
017 =Chickasaw
019 =Choctaw
021 =Claiborne
023 =Clarke
025 =Clay
027 =Coahoma
029 =Copiah
031 =Covington
033 =DeSoto
035 =Forrest
037 =Franklin
039 =George
041 =Greene
043 =Grenada
045 =Hancock
047 =Harrison
049 =Hinds
051 =Holmes
053 =Humphreys
055 =Issaquena
057 =Itawamba
059 =Jackson
061 =Jasper
063 =Jefferson
065 =Jefferson Davis
067 =Jones
069 =Kemper
071 =Lafayette
073 =Lamar
075 =Lauderdale
077 =Lawrence
079 =Leake
081 =Lee
083 =Leflore
085 =Lincoln
087 =Lowndes
089 =Madison
091 =Marion
093 =Marshall
095 =Monroe
097 =Montgomery
099 =Neshoba
101 =Newton
103 =Noxubee
105 =Oktibbeha
107 =Panola
109 =Pearl River
111 =Perry
113 =Pike
115 =Pontotoc
117 =Prentiss
119 =Quitman
121 =Rankin
123 =Scott
125 =Sharkey
127 =Simpson
129 =Smith
131 =Stone
133 =Sunflower
135 =Tallahatchie
137 =Tate
139 =Tippah
141 =Tishomingo
143 =Tunica
145 =Union
147 =Walthall
149 =Warren
151 =Washington
153 =Wayne
155 =Webster
- 171 -
157 =Wilkinson
159 =Winston
161 =Yalobusha
163 =Yazoo
MISSOURI - 29
001 =Adair
003 =Andrew
005 =Atchison
007 =Audrain
009 =Barry
011 =Barton
013 =Bates
015 =Benton
017 =Bollinger
019 =Boone
021 =Buchanan
023 =Butler
025 =Caldwell
027 =Callaway
029 =Camden
031 =Cape Girardeau
033 =Carroll
035 =Carter
037 =Cass
039 =Cedar
041 =Chariton
043 =Christian
045 =Clark
047 =Clay
049 =Clinton
051 =Cole
053 =Cooper
055 =Crawford
057 =Dade
059 =Dallas
061 =Daviess
063 =DeKalb
065 =Dent
067 =Douglas
069 =Dunklin
071 =Franklin
073 =Gasconade
075 =Gentry
077 =Greene
079 =Grundy
081 =Harrison
083 =Henry
085 =Hickory
087 =Holt
089 =Howard
091 =Howell
093 =Iron
095 =Jackson
097 =Jasper
099 =Jefferson
101 =Johnson
103 =Knox
105 =Laclede
107 =Lafayette
109 =Lawrence
111 =Lewis
113 =Lincoln
115 =Linn
117 =Livingston
119 =McDonald
121 =Macon
123 =Madison
125 =Maries
127 =Marion
129 =Mercer
131 =Miller
133 =Mississippi
135 =Moniteau
137 =Monroe
139 =Montgomery
141 =Morgan
143 =New Madrid
145 =Newton
147 =Nodaway
149 =Oregon
151 =Osage
153 =Ozark
155 =Pemiscot
157 =Perry
159 =Pettis
161 =Phelps
163 =Pike
165 =Platte
167 =Polk
169 =Pulaski
171 =Putnam
173 =Ralls
175 =Randolph
177 =Ray
179 =Reynolds
181 =Ripley
183 =St. Charles
185 =St. Clair
186 =Ste. Genevieve
187 =St. Francois
189 =St. Louis
195 =Saline
197 =Schuyler
199 =Scotland
201 =Scott
203 =Shannon
205 =Shelby
207 =Stoddard
209 =Stone
211 =Sullivan
213 =Taney
215 =Texas
217 =Vernon
219 =Warren
221 =Washington
223 =Wayne
225 =Webster
227 =Worth
229 =Wright
510 =St. Louis
MONTANA - 30
001 =Beaverhead
003 =Big Horn
005 =Blaine
007 =Broadwater
009 =Carbon
- 172 -
011 =Carter
013 =Cascade
015 =Chouteau
017 =Custer
019 =Daniels
021 =Dawson
023 =Deer Lodge
025 =Fallon
027 =Fergus
029 =Flathead
031 =Gallatin
033 =Garfield
035 =Glacier
037 =Golden Valley
039 =Granite
041 =Hill
043 =Jefferson
045 =Judith Basin
047 =Lake
049 =Lewis and Clark
051 =Liberty
053 =Lincoln
055 =McCone
057 =Madison
059 =Meagher
061 =Mineral
063 =Missoula
065 =Musselshell
067 =Park
069 =Petroleum
071 =Phillips
073 =Pondera
075 =Powder River
077 =Powell
079 =Prairie
081 =Ravalli
083 =Richland
085 =Roosevelt
087 =Rosebud
089 =Sanders
091 =Sheridan
093 =Silver Bow
095 =Stillwater
097 =Sweet Grass
099 =Teton
101 =Toole
103 =Treasure
105 =Valley
107 =Wheatland
109 =Wibaux
111 =Yellowstone
113 =Yellowstone National
NEBRASKA - 31
001 =Adams
003 =Antelope
005 =Arthur
007 =Banner
009 =Blaine
011 =Boone
013 =Box Butte
015 =Boyd
017 =Brown
019 =Buffalo
021 =Burt
023 =Butler
025 =Cass
027 =Cedar
029 =Chase
031 =Cherry
033 =Cheyenne
035 =Clay
037 =Colfax
039 =Cuming
041 =Custer
043 =Dakota
045 =Dawes
047 =Dawson
049 =Deuel
051 =Dixon
053 =Dodge
055 =Douglas
057 =Dundy
059 =Fillmore
061 =Franklin
063 =Frontier
065 =Furnas
067 =Gage
069 =Garden
071 =Garfield
073 =Gosper
075 =Grant
077 =Greeley
079 =Hall
081 =Hamilton
083 =Harlan
085 =Hayes
087 =Hitchcock
089 =Holt
091 =Hooker
093 =Howard
095 =Jefferson
097 =Johnson
099 =Kearney
101 =Keith
103 =Keya Paha
105 =Kimball
107 =Knox
109 =Lancaster
111 =Lincoln
113 =Logan
115 =Loup
117 =McPherson
119 =Madison
121 =Merrick
123 =Morrill
125 =Nance
127 =Nemaha
129 =Nuckolls
131 =Otoe
133 =Pawnee
135 =Perkins
137 =Phelps
139 =Pierce
141 =Platte
143 =Polk
145 =Red Willow
147 =Richardson
149 =Rock
151 =Saline
- 173 -
153 =Sarpy
155 =Saunders
157 =Scotts Bluff
159 =Seward
161 =Sheridan
163 =Sherman
165 =Sioux
167 =Stanton
169 =Thayer
171 =Thomas
173 =Thurston
175 =Valley
177 =Washington
179 =Wayne
181 =Webster
183 =Wheeler
185 =York
NEVADA - 32
001 =Churchill
003 =Clark
005 =Douglas
007 =Elko
009 =Esmeralda
011 =Eureka
013 =Humboldt
015 =Lander
017 =Lincoln
019 =Lyon
021 =Mineral
023 =Nye
027 =Pershing
029 =Storey
031 =Washoe
033 =White Pine
510 =Carson
NEW HAMPSHIRE - 33
001 =Belknap
003 =Carroll
005 =Cheshire
007 =Coos
009 =Grafton
011 =Hillsborough
013 =Merrimack
015 =Rockingham
017 =Strafford
019 =Sullivan
NEW JERSEY - 34
001 =Atlantic
003 =Bergen
005 =Burlington
007 =Camden
009 =Cape May
011 =Cumberland
013 =Essex
015 =Gloucester
017 =Hudson
019 =Hunterdon
021 =Mercer
023 =Middlesex
025 =Monmouth
027 =Morris
029 =Ocean
031 =Passaic
033 =Salem
035 =Somerset
037 =Sussex
039 =Union
041 =Warren
NEW MEXICO - 35
001 =Bernalillo
003 =Catron
005 =Chaves
006 =Cibola
007 =Colfax
009 =Curry
011 =DeBaca
013 =Dona Ana
015 =Eddy
017 =Grant
019 =Guadalupe
021 =Harding
023 =Hidalgo
025 =Lea
027 =Lincoln
028 =Los Alamos
029 =Luna
031 =McKinley
033 =Mora
035 =Otero
037 =Quay
039 =Rio Arriba
041 =Roosevelt
043 =Sandoval
045 =San Juan
047 =San Miguel
049 =Santa Fe
051 =Sierra
053 =Socorro
055 =Taos
057 =Torrance
059 =Union
061 =Valencia
NEW YORK - 36
001 =Albany
003 =Allegany
005 =Bronx
007 =Broome
009 =Cattaraugus
011 =Cayuga
013 =Chautauqua
015 =Chemung
017 =Chenango
019 =Clinton
021 =Columbia
023 =Cortland
025 =Delaware
027 =Dutchess
- 174 -
029 =Erie
031 =Essex
033 =Franklin
035 =Fulton
037 =Genesee
039 =Greene
041 =Hamilton
043 =Herkimer
045 =Jefferson
047 =Kings
049 =Lewis
051 =Livingston
053 =Madison
055 =Monroe
057 =Montgomery
059 =Nassau
061 =New York
063 =Niagara
065 =Oneida
067 =Onondaga
069 =Ontario
071 =Orange
073 =Orleans
075 =Oswego
077 =Otsego
079 =Putnam
081 =Queens
083 =Rensselaer
085 =Richmond
087 =Rockland
089 =St. Lawrence
091 =Saratoga
093 =Schenectady
095 =Schoharie
097 =Schuyler
099 =Seneca
101 =Steuben
103 =Suffolk
105 =Sullivan
107 =Tioga
109 =Tompkins
111 =Ulster
113 =Warren
115 =Washington
117 =Wayne
119 =Westchester
121 =Wyoming
123 =Yates
NORTH CAROLINA - 37
001 =Alamance
003 =Alexander
005 =Alleghany
007 =Anson
009 =Ashe
011 =Avery
013 =Beaufort
015 =Bertie
017 =Bladen
019 =Brunswick
021 =Buncombe
023 =Burke
025 =Cabarrus
027 =Caldwell
029 =Camden
031 =Carteret
033 =Caswell
035 =Catawba
037 =Chatham
039 =Cherokee
041 =Chowan
043 =Clay
045 =Cleveland
047 =Columbus
049 =Craven
051 =Cumberland
053 =Currituck
055 =Dare
057 =Davidson
059 =Davie
061 =Duplin
063 =Durham
065 =Edgecombe
067 =Forsyth
069 =Franklin
071 =Gaston
073 =Gates
075 =Graham
077 =Granville
079 =Greene
081 =Guilford
083 =Halifax
085 =Harnett
087 =Haywood
089 =Henderson
091 =Hertford
093 =Hoke
095 =Hyde
097 =Iredell
099 =Jackson
101 =Johnston
103 =Jones
105 =Lee
107 =Lenoir
109 =Lincoln
111 =McDowell
113 =Macon
115 =Madison
117 =Martin
119 =Mecklenburg
121 =Mitchell
123 =Montgomery
125 =Moore
127 =Nash
129 =New Hanover
131 =Northampton
133 =Onslow
135 =Orange
137 =Pamlico
139 =Pasquotank
141 =Pender
143 =Perquimans
145 =Person
147 =Pitt
149 =Polk
151 =Randolph
153 =Richmond
155 =Robeson
157 =Rockingham
159 =Rowan
- 175 -
161 =Rutherford
163 =Sampson
165 =Scotland
167 =Stanly
169 =Stokes
171 =Surry
173 =Swain
175 =Transylvania
177 =Tyrrell
179 =Union
181 =Vance
183 =Wake
185 =Warren
187 =Washington
189 =Watauga
191 =Wayne
193 =Wilkes
195 =Wilson
197 =Yadkin
199 =Yancey
NORTH DAKOTA - 38
001 =Adams
003 =Barnes
005 =Benson
007 =Billings
009 =Bottineau
011 =Bowman
013 =Burke
015 =Burleigh
017 =Cass
019 =Cavalier
021 =Dickey
023 =Divide
025 =Dunn
027 =Eddy
029 =Emmons
031 =Foster
033 =Golden Valley
035 =Grand Forks
037 =Grant
039 =Griggs
041 =Hettinger
043 =Kidder
045 =LaMoure
047 =Logan
049 =McHenry
051 =McIntosh
053 =McKenzie
055 =McLean
057 =Mercer
059 =Morton
061 =Mountrail
063 =Nelson
065 =Oliver
067 =Pembina
069 =Pierce
071 =Ramsey
073 =Ransom
075 =Renville
077 =Richland
079 =Rolette
081 =Sargent
083 =Sheridan
085 =Sioux
087 =Slope
089 =Stark
091 =Steele
093 =Stutsman
095 =Towner
097 =Traill
099 =Walsh
101 =Ward
103 =Wells
105 =Williams
OHIO - 39
001 =Adams
003 =Allen
005 =Ashland
007 =Ashtabula
009 =Athens
011 =Auglaize
013 =Belmont
015 =Brown
017 =Butler
019 =Carroll
021 =Champaign
023 =Clark
025 =Clermont
027 =Clinton
029 =Columbiana
031 =Coshocton
033 =Crawford
035 =Cuyahoga
037 =Darke
039 =Defiance
041 =Delaware
043 =Erie
045 =Fairfield
047 =Fayette
049 =Franklin
051 =Fulton
053 =Gallia
055 =Geauga
057 =Greene
059 =Guernsey
061 =Hamilton
063 =Hancock
065 =Hardin
067 =Harrison
069 =Henry
071 =Highland
073 =Hocking
075 =Holmes
077 =Huron
079 =Jackson
081 =Jefferson
083 =Knox
085 =Lake
087 =Lawrence
089 =Licking
091 =Logan
093 =Lorain
095 =Lucas
097 =Madison
099 =Mahoning
101 =Marion
- 176 -
103 =Medina
105 =Meigs
107 =Mercer
109 =Miami
111 =Monroe
113 =Montgomery
115 =Morgan
117 =Morrow
119 =Muskingum
121 =Noble
123 =Ottawa
125 =Paulding
127 =Perry
129 =Pickaway
131 =Pike
133 =Portage
135 =Preble
137 =Putnam
139 =Richland
141 =Ross
143 =Sandusky
145 =Scioto
147 =Seneca
149 =Shelby
151 =Stark
153 =Summit
155 =Trumbull
157 =Tuscarawas
159 =Union
161 =Van Wert
163 =Vinton
165 =Warren
167 =Washington
169 =Wayne
171 =Williams
173 =Wood
175 =Wyandot
OKLAHOMA - 40
001 =Adair
003 =Alfalfa
005 =Atoka
007 =Beaver
009 =Beckham
011 =Blaine
013 =Bryan
015 =Caddo
017 =Canadian
019 =Carter
021 =Cherokee
023 =Choctaw
025 =Cimarron
027 =Cleveland
029 =Coal
031 =Comanche
033 =Cotton
035 =Craig
037 =Creek
039 =Custer
041 =Delaware
043 =Dewey
045 =Ellis
047 =Garfield
049 =Garvin
051 =Grady
053 =Grant
055 =Greer
057 =Harmon
059 =Harper
061 =Haskell
063 =Hughes
065 =Jackson
067 =Jefferson
069 =Johnston
071 =Kay
073 =Kingfisher
075 =Kiowa
077 =Latimer
079 =Le Flore
081 =Lincoln
083 =Logan
085 =Love
087 =McClain
089 =McCurtain
091 =McIntosh
093 =Major
095 =Marshall
097 =Mayes
099 =Murray
101 =Muskogee
103 =Noble
105 =Nowata
107 =Okfuskee
109 =Oklahoma
111 =Okmulgee
113 =Osage
115 =Ottawa
117 =Pawnee
119 =Payne
121 =Pittsburg
123 =Pontotoc
125 =Pottawatomie
127 =Pushmataha
129 =Roger Mills
131 =Rogers
133 =Seminole
135 =Sequoyah
137 =Stephens
139 =Texas
141 =Tillman
143 =Tulsa
145 =Wagoner
147 =Washington
149 =Washita
151 =Woods
153 =Woodward
OREGON - 41
001 =Baker
003 =Benton
005 =Clackamas
007 =Clatsop
009 =Columbia
011 =Coos
013 =Crook
015 =Curry
017 =Deschutes
019 =Douglas
- 177 -
021 =Gilliam
023 =Grant
025 =Harney
027 =Hood River
029 =Jackson
031 =Jefferson
033 =Josephine
035 =Klamath
037 =Lake
039 =Lane
041 =Lincoln
043 =Linn
045 =Malheur
047 =Marion
049 =Morrow
051 =Multnomah
053 =Polk
055 =Sherman
057 =Tillamook
059 =Umatilla
061 =Union
063 =Wallowa
065 =Wasco
067 =Washington
069 =Wheeler
071 =Yamhill
PENNSYLVANIA - 42
001 =Adams
003 =Allegheny
005 =Armstrong
007 =Beaver
009 =Bedford
011 =Berks
013 =Blair
015 =Bradford
017 =Bucks
019 =Butler
021 =Cambria
023 =Cameron
025 =Carbon
027 =Centre
029 =Chester
031 =Clarion
033 =Clearfield
035 =Clinton
037 =Columbia
039 =Crawford
041 =Cumberland
043 =Dauphin
045 =Delaware
047 =Elk
049 =Erie
051 =Fayette
053 =Forest
055 =Franklin
057 =Fulton
059 =Greene
061 =Huntingdon
063 =Indiana
065 =Jefferson
067 =Juniata
069 =Lackawanna
071 =Lancaster
073 =Lawrence
075 =Lebanon
077 =Lehigh
079 =Luzerne
081 =Lycoming
083 =Mc Kean
085 =Mercer
087 =Mifflin
089 =Monroe
091 =Montgomery
093 =Montour
095 =Northampton
097 =Northumberland
099 =Perry
101 =Philadelphia
103 =Pike
105 =Potter
107 =Schuylkill
109 =Snyder
111 =Somerset
113 =Sullivan
115 =Susquehanna
117 =Tioga
119 =Union
121 =Venango
123 =Warren
125 =Washington
127 =Wayne
129 =Westmoreland
131 =Wyoming
133 =York
RHODE ISLAND - 44
001 =Bristol
003 =Kent
005 =Newport
007 =Providence
009 =Washington
SOUTH CAROLINA - 45
001 =Abbeville
003 =Aiken
005 =Allendale
007 =Anderson
009 =Bamberg
011 =Barnwell
013 =Beaufort
015 =Berkeley
017 =Calhoun
019 =Charleston
021 =Cherokee
023 =Chester
025 =Chesterfield
027 =Clarendon
029 =Colleton
031 =Darlington
033 =Dillon
035 =Dorchester
037 =Edgefield
039 =Fairfield
041 =Florence
043 =Georgetown
- 178 -
045 =Greenville
047 =Greenwood
049 =Hampton
051 =Horry
053 =Jasper
055 =Kershaw
057 =Lancaster
059 =Laurens
061 =Lee
063 =Lexington
065 =McCormick
067 =Marion
069 =Marlboro
071 =Newberry
073 =Oconee
075 =Orangeburg
077 =Pickens
079 =Richland
081 =Saluda
083 =Spartanburg
085 =Sumter
087 =Union
089 =Williamsburg
091 =York
SOUTH DAKOTA - 46
003 =Aurora
005 =Beadle
007 =Bennett
009 =Bon Homme
011 =Brookings
013 =Brown
015 =Brule
017 =Buffalo
019 =Butte
021 =Campbell
023 =Charles Mix
025 =Clark
027 =Clay
029 =Codington
031 =Corson
033 =Custer
035 =Davison
037 =Day
039 =Deuel
041 =Dewey
043 =Douglas
045 =Edmunds
047 =Fall River
049 =Faulk
051 =Grant
053 =Gregory
055 =Haakon
057 =Hamlin
059 =Hand
061 =Hanson
063 =Harding
065 =Hughes
067 =Hutchinson
069 =Hyde
071 =Jackson
073 =Jerauld
075 =Jones
077 =Kingsbury
079 =Lake
081 =Lawrence
083 =Lincoln
085 =Lyman
087 =McCook
089 =McPherson
091 =Marshall
093 =Meade
095 =Mellette
097 =Miner
099 =Minnehaha
101 =Moody
103 =Pennington
105 =Perkins
107 =Potter
109 =Roberts
111 =Sanborn
113 =Shannon
115 =Spink
117 =Stanley
119 =Sully
121 =Todd
123 =Tripp
125 =Turner
127 =Union
129 =Walworth
135 =Yankton
137 =Ziebach
TENNESSEE - 47
001 =Anderson
003 =Bedford
005 =Benton
007 =Bledsoe
009 =Blount
011 =Bradley
013 =Campbell
015 =Cannon
017 =Carroll
019 =Carter
021 =Cheatham
023 =Chester
025 =Claiborne
027 =Clay
029 =Cocke
031 =Coffee
033 =Crockett
035 =Cumberland
037 =Davidson
039 =Decatur
041 =DeKalb
043 =Dickson
045 =Dyer
047 =Fayette
049 =Fentress
051 =Franklin
053 =Gibson
055 =Giles
057 =Grainger
059 =Greene
061 =Grundy
063 =Hamblen
065 =Hamilton
067 =Hancock
069 =Hardeman
- 179 -
071 =Hardin
073 =Hawkins
075 =Haywood
077 =Henderson
079 =Henry
081 =Hickman
083 =Houston
085 =Humphreys
087 =Jackson
089 =Jefferson
091 =Johnson
093 =Knox
095 =Lake
097 =Lauderdale
099 =Lawrence
101 =Lewis
103 =Lincoln
105 =Loudon
107 =McMinn
109 =McNairy
111 =Macon
113 =Madison
115 =Marion
117 =Marshall
119 =Maury
121 =Meigs
123 =Monroe
125 =Montgomery
127 =Moore
129 =Morgan
131 =Obion
133 =Overton
135 =Perry
137 =Pickett
139 =Polk
141 =Putnam
143 =Rhea
145 =Roane
147 =Robertson
149 =Rutherford
151 =Scott
153 =Sequatchie
155 =Sevier
157 =Shelby
159 =Smith
161 =Stewart
163 =Sullivan
165 =Sumner
167 =Tipton
169 =Trousdale
171 =Unicoi
173 =Union
175 =Van Buren
177 =Warren
179 =Washington
181 =Wayne
183 =Weakley
185 =White
187 =Williamson
189 =Wilson
TEXAS - 48
001 =Anderson
003 =Andrews
005 =Angelina
007 =Aransas
009 =Archer
011 =Armstrong
013 =Atascosa
015 =Austin
017 =Bailey
019 =Bandera
021 =Bastrop
023 =Baylor
025 =Bee
027 =Bell
029 =Bexar
031 =Blanco
033 =Borden
035 =Bosque
037 =Bowie
039 =Brazoria
041 =Brazos
043 =Brewster
045 =Briscoe
047 =Brooks
049 =Brown
051 =Burleson
053 =Burnet
055 =Caldwell
057 =Calhoun
059 =Callahan
061 =Cameron
063 =Camp
065 =Carson
067 =Cass
069 =Castro
071 =Chambers
073 =Cherokee
075 =Childress
077 =Clay
079 =Cochran
081 =Coke
083 =Coleman
085 =Collin
087 =Collingsworth
089 =Colorado
091 =Comal
093 =Comanche
095 =Concho
097 =Cooke
099 =Coryell
101 =Cottle
103 =Crane
105 =Crockett
107 =Crosby
109 =Culberson
111 =Dallam
113 =Dallas
115 =Dawson
117 =Deaf Smith
119 =Delta
121 =Denton
123 =DeWitt
125 =Dickens
127 =Dimmit
129 =Donley
131 =Duval
133 =Eastland
135 =Ector
- 180 -
137 =Edwards
139 =Ellis
141 =El Paso
143 =Erath
145 =Falls
147 =Fannin
149 =Fayette
151 =Fisher
153 =Floyd
155 =Foard
157 =Fort Bend
159 =Franklin
161 =Freestone
163 =Frio
165 =Gaines
167 =Galveston
169 =Garza
171 =Gillespie
173 =Glasscock
175 =Goliad
177 =Gonzales
179 =Gray
181 =Grayson
183 =Gregg
185 =Grimes
187 =Guadalupe
189 =Hale
191 =Hall
193 =Hamilton
195 =Hansford
197 =Hardeman
199 =Hardin
201 =Harris
203 =Harrison
205 =Hartley
207 =Haskell
209 =Hays
211 =Hemphill
213 =Henderson
215 =Hidalgo
217 =Hill
219 =Hockley
221 =Hood
223 =Hopkins
225 =Houston
227 =Howard
229 =Hudspeth
231 =Hunt
233 =Hutchinson
235 =Irion
237 =Jack
239 =Jackson
241 =Jasper
243 =Jeff Davis
245 =Jefferson
247 =Jim Hogg
249 =Jim Wells
251 =Johnson
253 =Jones
255 =Karnes
257 =Kaufman
259 =Kendall
261 =Kenedy
263 =Kent
265 =Kerr
267 =Kimble
269 =King
271 =Kinney
273 =Kleberg
275 =Knox
277 =Lamar
279 =Lamb
281 =Lampasas
283 =La Salle
285 =Lavaca
287 =Lee
289 =Leon
291 =Liberty
293 =Limestone
295 =Lipscomb
297 =Live Oak
299 =Llano
301 =Loving
303 =Lubbock
305 =Lynn
307 =McCulloch
309 =McLennan
311 =McMullen
313 =Madison
315 =Marion
317 =Martin
319 =Mason
321 =Matagorda
323 =Maverick
325 =Medina
327 =Menard
329 =Midland
331 =Milam
333 =Mills
335 =Mitchell
337 =Montague
339 =Montgomery
341 =Moore
343 =Morris
345 =Motley
347 =Nacogdoches
349 =Navarro
351 =Newton
353 =Nolan
355 =Nueces
357 =Ochiltree
359 =Oldham
361 =Orange
363 =Palo Pinto
365 =Panola
367 =Parker
369 =Parmer
371 =Pecos
373 =Polk
375 =Potter
377 =Presidio
379 =Rains
381 =Randall
383 =Reagan
385 =Real
387 =Red River
389 =Reeves
391 =Refugio
393 =Roberts
395 =Robertson
397 =Rockwall
399 =Runnels
- 181 -
401 =Rusk
403 =Sabine
405 =San Augustine
407 =San Jacinto
409 =San Patricio
411 =San Saba
413 =Schleicher
415 =Scurry
417 =Shackelford
419 =Shelby
421 =Sherman
423 =Smith
425 =Somervell
427 =Starr
429 =Stephens
431 =Sterling
433 =Stonewall
435 =Sutton
437 =Swisher
439 =Tarrant
441 =Taylor
443 =Terrell
445 =Terry
447 =Throckmorton
449 =Titus
451 =Tom Green
453 =Travis
455 =Trinity
457 =Tyler
459 =Upshur
461 =Upton
463 =Uvalde
465 =Val Verde
467 =Van Zandt
469 =Victoria
471 =Walker
473 =Waller
475 =Ward
477 =Washington
479 =Webb
481 =Wharton
483 =Wheeler
485 =Wichita
487 =Wilbarger
489 =Willacy
491 =Williamson
493 =Wilson
495 =Winkler
497 =Wise
499 =Wood
501 =Yoakum
503 =Young
505 =Zapata
507 =Zavala
UTAH - 49
001 =Beaver
003 =Box Elder
005 =Cache
007 =Carbon
009 =Daggett
011 =Davis
013 =Duchesne
015 =Emery
017 =Garfield
019 =Grand
021 =Iron
023 =Juab
025 =Kane
027 =Millard
029 =Morgan
031 =Piute
033 =Rich
035 =Salt Lake
037 =San Juan
039 =Sanpete
041 =Sevier
043 =Summit
045 =Tooele
047 =Uintah
049 =Utah
051 =Wasatch
053 =Washington
055 =Wayne
057 =Weber
VERMONT - 50
001 =Addison
003 =Bennington
005 =Caledonia
007 =Chittenden
009 =Essex
011 =Franklin
013 =Grand Isle
015 =Lamoille
017 =Orange
019 =Orleans
021 =Rutland
023 =Washington
025 =Windham
027 =Windsor
VIRGINIA - 51
001 =Accomack
003 =Albemarle
005 =Alleghany
007 =Amelia
009 =Amherst
011 =Appomattox
013 =Arlington
015 =Augusta
017 =Bath
019 =Bedford
021 =Bland
023 =Botetourt
025 =Brunswick
027 =Buchanan
029 =Buckingham
031 =Campbell
033 =Caroline
035 =Carroll
036 =Charles City
037 =Charlotte
041 =Chesterfield
043 =Clarke
045 =Craig
047 =Culpeper
- 182 -
049 =Cumberland
051 =Dickenson
053 =Dinwiddie
057 =Essex
059 =Fairfax
061 =Fauquier
063 =Floyd
065 =Fluvanna
067 =Franklin
069 =Frederick
071 =Giles
073 =Gloucester
075 =Goochland
077 =Grayson
079 =Greene
081 =Greensville
083 =Halifax
085 =Hanover
087 =Henrico
089 =Henry
091 =Highland
093 =Isle of Wight
095 =James City
097 =King and Queen
099 =King George
101 =King William
103 =Lancaster
105 =Lee
107 =Loudoun
109 =Louisa
111 =Lunenburg
113 =Madison
115 =Mathews
117 =Mecklenburg
119 =Middlesex
121 =Montgomery
125 =Nelson
127 =New Kent
131 =Northampton
133 =Northumberland
135 =Nottoway
137 =Orange
139 =Page
141 =Patrick
143 =Pittsylvania
145 =Powhatan
147 =Prince Edward
149 =Prince George
153 =Prince William
155 =Pulaski
157 =Rappahannock
159 =Richmond
161 =Roanoke
163 =Rockbridge
165 =Rockingham
167 =Russell
169 =Scott
171 =Shenandoah
173 =Smyth
175 =Southampton
177 =Spotsylvania
179 =Stafford
181 =Surry
183 =Sussex
185 =Tazewell
187 =Warren
191 =Washington
193 =Westmoreland
195 =Wise
197 =Wythe
199 =York
510 =Alexandria
515 =Bedford
520 =Bristol
530 =Buena Vista
540 =Charlottesville
550 =Chesapeake
560 =Clifton Forge
570 =Colonial Heights
580 =Covington
590 =Danville
595 =Emporia
600 =Fairfax
610 =Falls Church
620 =Franklin
630 =Fredericksburg
640 =Galax
650 =Hampton
660 =Harrisonburg
670 =Hopewell
678 =Lexington
680 =Lynchburg
683 =Manassas
685 =Manassas Park
690 =Martinsville
700 =Newport News
710 =Norfolk
720 =Norton
730 =Petersburg
735 =Poquoson
740 =Portsmouth
750 =Radford
760 =Richmond
770 =Roanoke
775 =Salem
780 =South Boston
790 =Staunton
800 =Suffolk
810 =Virginia Beach
820 =Waynesboro
830 =Williamsburg
840 =Winchester
WASHINGTON - 53
001 =Adams
003 =Asotin
005 =Benton
007 =Chelan
009 =Clallam
011 =Clark
013 =Columbia
015 =Cowlitz
017 =Douglas
019 =Ferry
021 =Franklin
023 =Garfield
025 =Grant
027 =Grays Harbor
029 =Island
031 =Jefferson
- 183 -
033 =King
035 =Kitsap
037 =Kittitas
039 =Klickitat
041 =Lewis
043 =Lincoln
045 =Mason
047 =Okanogan
049 =Pacific
051 =Pend Oreille
053 =Pierce
055 =San Juan
057 =Skagit
059 =Skamania
061 =Snohomish
063 =Spokane
065 =Stevens
067 =Thurston
069 =Wahkiakum
071 =Walla Walla
073 =Whatcom
075 =Whitman
077 =Yakima
WEST VIRGINIA - 54
001 =Barbour
003 =Berkeley
005 =Boone
007 =Braxton
009 =Brooke
011 =Cabell
013 =Calhoun
015 =Clay
017 =Doddridge
019 =Fayette
021 =Gilmer
023 =Grant
025 =Greenbrier
027 =Hampshire
029 =Hancock
031 =Hardy
033 =Harrison
035 =Jackson
037 =Jefferson
039 =Kanawha
041 =Lewis
043 =Lincoln
045 =Logan
047 =McDowell
049 =Marion
051 =Marshall
053 =Mason
055 =Mercer
057 =Mineral
059 =Mingo
061 =Monongalia
063 =Monroe
065 =Morgan
067 =Nicholas
069 =Ohio
071 =Pendleton
073 =Pleasants
075 =Pocahontas
077 =Preston
079 =Putnam
081 =Raleigh
083 =Randolph
085 =Ritchie
087 =Roane
089 =Summers
091 =Taylor
093 =Tucker
095 =Tyler
097 =Upshur
099 =Wayne
101 =Webster
103 =Wetzel
105 =Wirt
107 =Wood
109 =Wyoming
WISCONSIN - 55
001 =Adams
003 =Ashland
005 =Barron
007 =Bayfield
009 =Brown
011 =Buffalo
013 =Burnett
015 =Calumet
017 =Chippewa
019 =Clark
021 =Columbia
023 =Crawford
025 =Dane
027 =Dodge
029 =Door
031 =Douglas
033 =Dunn
035 =Eau Claire
037 =Florence
039 =Fond du Lac
041 =Forest
043 =Grant
045 =Green
047 =Green Lake
049 =Iowa
051 =Iron
053 =Jackson
055 =Jefferson
057 =Juneau
059 =Kenosha
061 =Kewaunee
063 =La Crosse
065 =Lafayette
067 =Langlade
069 =Lincoln
071 =Manitowoc
073 =Marathon
075 =Marinette
077 =Marquette
078 =Menominee
079 =Milwaukee
081 =Monroe
083 =Oconto
085 =Oneida
087 =Outagamie
089 =Ozaukee
- 184 -
091 =Pepin
093 =Pierce
095 =Polk
097 =Portage
099 =Price
101 =Racine
103 =Richland
105 =Rock
107 =Rusk
109 =St. Croix
111 =Sauk
113 =Sawyer
115 =Shawano
117 =Sheboygan
119 =Taylor
121 =Trempealeau
123 =Vernon
125 =Vilas
127 =Walworth
129 =Washburn
131 =Washington
133 =Waukesha
135 =Waupaca
137 =Waushara
139 =Winnebago
141 =Wood
WYOMING - 56
001 =Albany
003 =Big Horn
005 =Campbell
007 =Carbon
009 =Converse
011 =Crook
013 =Fremont
015 =Goshen
017 =Hot Springs
019 =Johnson
021 =Laramie
023 =Lincoln
025 =Natrona
027 =Niobrara
029 =Park
031 =Platte
033 =Sheridan
035 =Sublette
037 =Sweetwater
039 =Teton
041 =Uinta
043 =Washakie
045 =Weston
AMER. SAMOA - 60
010 =Eastern
020 =Manu'a
030 =Rose
040 =Swains
050 =Western
GUAN - 66
010 =Guam
PUERTO RICO - 72
001 =Adjuntas
003 =Aguada
005 =Aguadilla
007 =Aguas Buenas
009 =Aibonito
011 =Añasco
013 =Arecibo
015 =Arroyo
017 =Barceloneta
019 =Barranquitas
021 =Bayamon
023 =Cabo Rojo
025 =Caguas
027 =Camuy
029 =Canovanas
031 =Carolina
033 =Cataño
035 =Cayey
037 =Ceiba
039 =Ciales
041 =Cidra
043 =Coamo
045 =Comerio
047 =Corozal
049 =Culebra
051 =Dorado
053 =Fajardo
054 =Florida
055 =Guanica
057 =Guayama
059 =Guayanilla
061 =Guaynabo
063 =Gurabo
065 =Hatillo
067 =Hormigueros
069 =Humacao
071 =Isabela
073 =Jayuya
075 =Juana Diaz
077 =Juncos
079 =Lajas
081 =Lares
083 =Las Marias
085 =Las Piedras
087 =Loiza
089 =Luquillo
091 =Manati
093 =Maricao
095 =Maunabo
097 =Mayagüez
099 =Moca
101 =Morovis
103 =Naguabo
105 =Naranjito
107 =Orocovis
109 =Patillas
111 =Peñuelas
113 =Ponce
115 =Quebradillas
117 =Rincon
119 =Rio Grande
- 185 -
121 =Sabana Grande
123 =Salinas
125 =San German
127 =San Juan
129 =San Lorenzo
131 =San Sebastian
133 =Santa Isabel
135 =Toa Alta
137 =Toa Baja
139 =Trujillo Alto
141 =Utuado
143 =Vega Alta
145 =Vega Baja
147 =Vieques
149 =Villalba
151 =Yabucoa
153 =Yauco
VIRGIN ISLANDS - 78
010 =St. Croix
020 =St. John
030 =St. Thomas
- 186 -
INDEX
Error-prone Profile 103
Expected Value 84, 90
Expenditures 31, 122-125, 129, 130, 132, 133
Fiscal Year 50, 53-57, 59-61, 119, 120, 124,
126, 129, 130, 136
Foster Care 31
Frame 2, 6, 17-19, 26-38, 42-44, 46, 47, 50, 51,
126, 130-132, 134, 141-143
Guam 133, 135, 183
Hours of Participation 59-61, 63, 77
Ineligible 77
Initial Payment 29
Intercept 95, 97
Least Squares 94
Listed-in-Error 19, 20, 26, 28, 33, 37, 42, 43,
50, 51, 66, 68, 72, 122, 124, 127, 142
Lower Limit 13
Lower Limits 13, 123
Mean 11, 79, 101, 127, 129, 131
Moving Average 90, 91
Multiple Regression 102, 103
Non-Sampling Error 128, 131
Normal Distribution 12, 128, 137
Null Hypothesis 77, 99
Official 132
Optimal Allocation 8, 9
Oversampling 10, 20, 33, 41, 42, 44, 49, 51,
128, 144
Parameter 10, 129, 131
Payroll Listing 134
Plan Approval 17
Plan Requirements 2, 3, 17, 141
Population 2, 5-7, 10, 12-15, 18, 22, 26, 41, 64,
77-79, 85, 107, 121, 123, 127-134, 141
Population of Interest 127, 129
Precision 6-11, 13, 14, 21-24, 41, 119, 129,
130, 133
Probability 5-7, 11-13, 18, 26, 41, 84, 85, 119,
125, 129, 131, 132, 141
30% Limit 61, 66, 69, 70, 73, 74
Absolute Value 84, 119
ACF 3, 17, 28, 36, 41, 48, 51, 119, 147
Adequate Sample 119
Agency 17, 25, 27, 41, 50, 104, 120, 133, 134
Alpha 84, 85, 119, 131
Alternate method 44
Annual Sample 17, 21-25, 32-34, 36-38, 41, 45,
47, 49, 50, 94, 95, 119, 126
Annual Sample Period 17, 24, 32, 34, 36, 38,
45, 47, 94, 95, 119, 126
Annual Samples 49
Audit Trail 39
Bias 5, 9, 10, 80, 84, 121, 128, 130
Caseload 2, 5, 6, 8-11, 14, 17-19, 21, 22, 24, 25,
28, 29, 31-34, 36, 38, 41-48, 50, 53-56, 65, 66,
68-70, 72, 73, 77-81, 90, 104, 121, 134, 141,
143
Caseload Reduction Credit 21, 53-56
Caseload Size 6, 8, 19, 32, 34, 36, 38, 42, 43,
121, 143
Chi-square 78, 80, 83, 84, 86, 89, 90, 99, 100
Coefficient of Correlation 100-102
Coefficient of Determination 101, 102
Computation of Sample Size 13
Confidence Interval 10-13, 16, 77-79, 123, 137
Confidence Limits 22, 79, 123, 128
Countable Work Activities 53, 58, 60
Custodial Parent 19, 29, 30, 56, 61, 64, 66, 68,
72, 75, 128, 143
Data Analysis 77
Degrees of Freedom 83, 84, 89
Discriminant Analysis 102, 103
Disposed of Case 124
Disposition 75
Dropped Cases 47
Emergency Assistance 124, 129
Equal Probability of Selection 125
Error-prone Cases 102
- 187 -
Sampling Error 5, 11, 77, 128, 131
Sampling Errors 5, 17, 18, 145
Sampling Plan 2, 3, 8, 17-19, 26, 28, 36, 38, 39,
46, 48, 49, 141, 143, 145
Sampling Plans 2, 3, 141
Sampling Techniques 5
Significant Difference 77, 85, 132
Significant Differences 77, 78, 80
Simple Random Sample 11, 14, 36-38, 132, 137
Simple Random Sampling 2, 3, 5-7, 16, 28, 32,
36, 37, 41, 49, 50, 78
Standard Deviation 79, 129, 132, 133
Standard Error 8, 9, 11-13, 15, 16, 78-80, 133,
137, 139
Standard Sample Size 47
State Agencies 5, 21, 26, 28, 41, 42
State Agency 17, 25, 41, 50, 133
Strata 7-9, 15, 41, 42, 45, 49, 71, 72, 134
Stratified Random Sample 16
Stratified Random Sampling 6, 133
Stratified Sample 7, 8, 14, 15, 20, 22, 36, 38,
49, 68, 70, 72, 74, 144
Stratified Samples 7, 14, 65
Stratum 7-9, 14, 15, 22-24, 32, 36-38, 45, 46,
49, 72, 74, 75, 134
Supplemental Case 134
Supplemental Cases 28
Suspended Case 134
Systematic Random Sampling 2, 3, 144
Territories 126, 135
Thirty (30) Percent Limit 61
Tolerance 11, 135
Trend 90, 91, 94, 95, 97-100
Trend Line 94, 95, 97-99
Trends 32, 77, 90
Tribal 1, 2, 14, 17, 19, 21, 25-28, 30-32, 36, 41,
42, 44, 46, 50, 56, 57, 63, 64, 66, 68, 72, 77, 78,
119, 126, 132, 135, 143, 147, 148, 151, 156
Tribes 2, 3, 21-24, 26, 28-30, 33, 38, 41, 45, 49,
50, 53, 62, 135, 147-150, 154-156
Undersampling 20, 33, 41-44, 46, 47, 49-51,
144
Universe 2, 13, 28, 41, 65, 80, 107, 132
Upper Limit 13
Probability Sampling 5, 18, 26, 41, 129, 131,
141
Proportional Allocation 8
Puerto Rico 133, 135, 183
Random Number 6, 38, 46
Random Numbers 6, 38, 43, 48, 49, 107, 130
Random Sampling 2, 3, 5-7, 16, 22, 28, 32, 36,
37, 41, 49, 50, 78, 129, 130, 132, 133, 144
Random Start 17, 34-36, 38, 42, 43, 47, 48, 130
Random Starts 20, 34, 144
Range 10, 11, 28, 99, 129, 130
Regional Administrator 3, 17, 36, 48
Regional Office 17, 41, 51
Reliability 9-11, 21, 23, 119, 129, 130
Reserve Pool 36, 38, 39, 46-49
Reserve Sample Pool 46, 47, 49
resources 14, 26, 102, 103, 133, 134
Retention of Sampling Records 38
Review Month 134
Risk 11, 102, 119, 131, 132
Sample 2, 3, 5-39, 41-51, 67-70, 72-74, 77-81,
84, 86, 87, 91-95, 100, 102, 103, 107, 119, 122,
123, 125-134, 137, 139, 141-144
Sample Frame 2, 6, 18, 26-32, 35-38, 42-44, 47,
50, 51, 131, 134, 141
Sample Frames 17, 38, 44, 50
Sample Interval 6, 32-36, 38, 42-49, 131
Sample Intervals 17, 20, 38, 44, 48, 144
Sample Month 18, 20, 25, 27-30, 33, 37, 50,
130, 131, 141, 142
Sample Period 6, 17, 24, 32-34, 36-38, 43-45,
47, 48, 78, 86, 94, 95, 100, 119, 126, 131
Sample Selection 2, 3, 17, 19, 36, 37, 46, 48,
50, 51, 78, 126, 127, 129, 131, 134, 143
Sample Selection List 51, 126, 127, 131
Sample Selection Process 78
Sample Size 3, 6-9, 11, 13-15, 20-26, 33, 34,
37-39, 41, 44-47, 50, 68, 72, 102, 107, 122,
131, 132, 137, 139, 144
Sample Sizes 15, 21-23, 25, 26, 33, 37, 41, 42,
78, 137, 139
Samples 2, 5-8, 10-12, 14, 17, 18, 21, 25, 26,
41, 42, 49, 50, 65, 67, 77, 80, 85, 119, 130-132
Sampling Distribution 128, 131, 133
- 188 -
Validity 9, 10, 84, 131
Variance 8, 11, 14-16, 67, 71
Virgin Islands 133, 135, 183
Waiver 53, 57, 59-66, 68, 73, 75, 76, 124-126,
136
Weight 15, 38
Weighted Sample 125
Work Activities 19, 26, 27, 29, 53, 57-63, 69,
70, 76, 120, 136, 143
Work Activity 59, 60, 63
Work Participation Rate 22, 29, 53-58, 60, 61,
63-67, 69, 70, 73, 74, 120, 128, 136
Work Participation Rates 3, 21-23, 53, 56,
61-63, 74, 120
Yates 84, 173
- 189 -
| File Type | application/pdf |
| File Title | H:\qcfiles\TANFMAN\Tanfmanual.wpd |
| Author | asaulnier@opre@acf.wdc |
| File Modified | 2001-09-13 |
| File Created | 2001-09-13 |