Imputation

CJRP 2007 imputation documentation FOR PUBLIC USE Revised Ja.pdf

Census of Juveniles in Residential Placement (CJRP)

Imputation

OMB: 1121-0218

⚠️ Notice: This form may be outdated. More recent filings and information on OMB 1121-0218 can be found here:

Document [pdf]

Download: pdf | pdf

January 22, 2009

MEMORANDUM FOR THE RECORD
From:

Suzanne M. Dorinski
Statistical Support and Consulting Staff
Office of Statistical Methods and Research for Economic Programs

Subject:

Public Use Documentation of the Imputation Methodology for the 2007 Census of
Juveniles in Residential Placement

This memorandum presents a streamlined version of the imputation methodology for the 2007
Census of Juveniles in Residential Placement (CJRP). There is a longer version of this
memorandum which is for internal use only because it contains individual juvenile
offender records, which are confidential. The individual juvenile offender records have been
removed from this version.
This document uses intentional white space to improve readability. The document is available as
a PDF file, because the original formatting might not be reproduced on the reader’s computer.
The imputation methodology for the 2003 CJRP and succeeding years is completely different
from the methodology used for the 1997, 1999, and 2001 censuses. Since the methodology has
completely changed, the references section does not refer to any imputation methodology
documentation for censuses before 2003.
Imputation programs written for the 2001 CJRP and earlier censuses used the section and
question number as variable names. It’s very easy to make a typing mistake while using that
convention. That naming convention also makes it more difficult to read the program code and
debug it. For the 2007 CJRP, we assigned variable names that are more descriptive. The naming
convention is shown in Table 1 on page 2. The section and question number for each item are
shown in parentheses.

Page 2 of 30
Table 1.

Naming conventions used in 2007 CJRP Imputation System

Concept

2007 CJRP

2006 JRFC

2006 CJRP

Persons assigned to
beds

total_2007
(S1Q10b)

total_2006_jrfc
(S1Q5b)

total_2006
(S1Q10b)

Persons assigned to
beds age 21 or older

adults_2007
(S1Q11)

Persons under age 21
assigned to beds

kids_2007
(S1Q12b)

Persons under age 21
assigned to beds due to
offenses

kid_offenders_2007
(S1Q13b)

Persons assigned to
beds for reasons other
than offenses

kid_nonoffenders_2007
(S1Q14b)

Juvenile offender ID

kid_id
(S2Q1)

Juvenile offender’s sex

kid_sex
(S2Q2)

Juvenile offender’s
birth date

kid_birth_month
(S2Q3)
kid_birth_day
(S2Q3)
kid_birth_year
(S2Q3)

Juvenile offender’s race

kid_race
(S2Q4)

Agency that placed the
juvenile offender in
facility

kid_placed_by
(S2Q5)

Juvenile offender’s
most serious offense

kid_offense
(S2Q7)

Juvenile offender’s
adjudication status

kid_adjudication_status
(S2Q9)

Juvenile offender’s date
of admission

kid_admitted_month
(S2Q10)
kid_admitted_day
(S2Q10)
kid_admitted_year
(S2Q10)

adults_2006_jrfc
(S1Q6)
kids_2006_jrfc
(S1Q7b)
kid_offenders_2006_jrfc
(S1Q8b)

kid_nonoffenders_2006_jrfc
(S1Q9b)

adults_2006
(S1Q11)
kids_2006
(S1Q12b)
kid_offenders_2006
(S1Q13b)

kid_nonoffenders_2006
(S1Q14b)

Page 3 of 30

Table of Contents
I.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

II.

Summary of the Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

III.

Critical Item Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

IV.

Questionnaire Items Eligible for Imputation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Imputation Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

VI.

Collapsed Facility Type Codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

VII.

Imputation Methodology for Section I Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

VIII. Imputation Methodology for Item Nonresponse in Section II Data . . . . . . . . . . 16
IX.

Imputation Methodology for Section II Data for Critical Item Facilities . . . . . . 18

Imputing Missing Data for Juvenile Offenders in Tribal Facilities . . . . . . . . . . . 20

XI.

Special Handling for Critical Item Units in the District of Columbia (DC) . . . . . 22

XII.

Caution When Comparing State Data over Time . . . . . . . . . . . . . . . . . . . . . . . . . . 23

XIII. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Attachment A:

Percentage of Juvenile Offenders Held in Critical Item Facilities by
State of Facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Attachment B:

Section II Item Imputation Rates by State of Facility . . . . . . . . . . 27

Attachment C:

Data Quality Issues in CJRP Over Time . . . . . . . . . . . . . . . . . . . . . . 29

Page 4 of 30

Introduction

The CJRP is a mail canvass census that was first conducted in 1997. The CJRP asks juvenile
residential custody facilities in the U.S. to describe each youth assigned to a bed in the facility on
the last Wednesday in October. The census is not sent to adult facilities, or facilities exclusively
for drug or mental health treatment, or abused or neglected children. The census is normally
conducted in odd-numbered years; the collection scheduled for 2005 was delayed several months
until early 2006. The reference date for the 2007 CJRP was Wednesday, October 24, 2007.
CJRP replaced the Census of Public and Private Juvenile Detention, Correctional, and Shelter
Facilities, also known as the Children in Custody census, which had been conducted since the
early 1970s. Previous censuses collected data on the facilities and the juvenile offenders that
were held in the facilities.
CJRP collects an individual record on each juvenile held in the residential facility, with
information on the juvenile’s sex, date of birth, race, who placed the offender there, most serious
offense, court adjudication status, and date of admission to the facility. These data are requested
for all offenders under 21 years of age in the facility.
The National Center for Juvenile Justice, the research division of the National Council of
Juvenile and Family Court Judges, maintains the CJRP databook online. The databook contains
a set of pre-defined tables detailing the characteristics of juvenile offenders in residential
placement facilities. Tables are currently available for 1997, 1999, 2001, 2003, and 2006.
The project sponsor is the Office of Juvenile Justice and Delinquency Prevention (OJJDP). The
Office of Statistical Methods and Research for Economic Programs (OSMREP) is responsible for
imputation in the CJRP, while the Governments Division (GOVS) is responsible for data
collection and editing.
For the 2007 CJRP, we mailed out questionnaires to 3,116 facilities. Of the 3,116 facilities, there
were 136 that had closed, 14 that were added after mail out, 21 temporarily out-of-scope (i.e.,
they did not hold offenders on the reference date), and 62 that were permanently out-of-scope
(i.e., they never hold juvenile offenders). Thus, there were 2,911 in-scope facilities on reference
day. 2,902 of the 2,911 facilities responded to the 2007 CJRP, for a 99.7 percent unit response
rate. 9 facilities refused to participate in the 2007 CJRP, but we imputed records for those
facilities.
The 2007 questionnaire is divided into two sections: Section I, which collects general
information about the facility, and Section II, which collects individual person data for juvenile
offenders held at the facility.
This document includes the response rates and describes the imputation methodology that we
used to make complete data sets for analytical purposes. We summarize the numbers of facilities

Page 5 of 30
and records reporting on our 2007 edited and imputed files in Chapter II. In Chapter III, we
discuss facilities, referred to as critical item facilities, that were only able to respond to a critical
subset of the requested data. We did not impute for every item in the questionnaire, and Chapter
IV covers the items that were eligible for imputation.
The discussion of imputation rates begins in Chapter V and continues in Chapters X through XII.
As described above, the unit response rate is very high and leads to very low imputation rates for
Total Persons, Total Adults, Total Juveniles, Total Juvenile Offenders, and Total Juvenile
Nonoffenders, as seen in Chapter V.
Special caution should be taken when using the State data over time. There is a marked increase
in high imputation rates for 2007. The exact imputation rates by State for 2007 can be found in
Attachment B. Items that are highlighted have imputation rates that exceed 30 percent, and
caution should be taken when using these data. Attachment C shows the items within each State
that are more than 30 percent imputed for the period from 1997 through 2007.
A description of the imputation methodology is provided in Chapters VI through IX. References
are included in Chapter XIII.

Page 6 of 30

II. Summary of the Files
Table 2.
628
59
307

Summary of records on the 2007 edited file
facilities that hold offenders and reported only critical items
facilities that hold nonoffenders only and reported only critical items
facilities that hold nonoffenders only and reported more than critical items

69,995

juvenile offender records from 1,908 facilities that reported more than critical items

70,989

records on the 2007 edited file

Table 3.
628
59
307

Summary of facilities on the 2007 edited file
facilities that hold offenders and reported only critical items
facilities that hold nonoffenders only and reported only critical items
facilities that hold nonoffenders only and reported more than critical items

1,908

facilities that hold offenders and reported more than critical items

2,902

facilities in the 2007 edited file

Table 4.
16,932
59
307

Summary of records on the 2007 imputed file
juvenile offender records from facilities that hold offenders and reported only critical items
facilities that hold nonoffenders only and reported only critical items
facilities that hold nonoffenders only and reported more than critical items

69,995

records for juvenile offenders from 1,908 facilities that reported more than critical items

87,293

records on the 2007 imputed file

Table 5.
637
59
307

Summary of facilities on the 2007 imputed file
facilities that hold offenders and reported only critical items
facilities that hold nonoffenders only and reported only critical items
facilities that hold nonoffenders only and reported more than critical items

1,908

facilities that hold offenders and reported more than critical items

2,911

facilities in the 2007 imputed file

Page 7 of 30
Table 6.

2007 CJRP head counts

103,900

people in residential placement

696
103,204

adults
juveniles

86,927

juvenile offenders

16,277

juvenile nonoffenders

III. Critical Item Facilities
In follow-up interviewing, GOVS attempted to collect as much data as possible to fill in both
sections of the questionnaire. The following data items were deemed critical:
Section I:
•
•
•
•
•
•

Question 5 (type of facility)
Question 10a and 10b (total persons assigned to beds in the facility)
Question 11 (number of persons age 21 or over assigned to beds in the facility)
Question 12a and 12b (number of persons under age 21 assigned to beds in the facility)
Question 13a and 13b (number of offenders under age 21 assigned to beds in the facility)
Question 14a and 14b (number of nonoffenders under age 21 assigned to beds in the
facility)

Section II:
•
•
•
•
•

Question 2 (whether the facility is all-male, all-female, or holds both sexes)
Question 5 (placement agency)
Question 6 (placement agency’s government level)
Question 7 (offense code)
Question 9 (adjudication status)

In previous CJRP data collections, the critical items field was set to either 0 or 1, with 1
indicating that the facility responded only to the critical items. In the 2007 CJRP, we have
expanded the possible values for the critical items field to include 4 statuses. (See Table 7 on the
next page.)

Page 8 of 30
Table 7.
Critical
item
field
value

Values for the critical items field
Meaning

Facility is neither a critical item facility nor a refusal.

Facility responded only to the critical items.

Facility responded to Section I of the questionnaire, but date of birth, offense, and
date of admission are missing for all records in Section II of the questionnaire.

Facility is a refusal; all data on the file for that facility have been imputed.

Facilities with the critical item field set to 1 have only one record per facility on the edited file,
and the information in Section II for those facilities is used to generate the juvenile offender
roster for each facility.
Facilities with the critical item field set to 2 are imputed in the same manner as facilities with the
critical item field set to 1. When date of birth, offense, and date of admission are missing for
every juvenile in the facility, we have to impute for every juvenile in the facility, and we want to
avoid using the same donor over and over within the facility.
Refusal facilities are also imputed in the same manner as facilities with the critical item field set
to 1, to minimize the number of records within a facility imputed using the same donor. The
critical item field value of 3 is the way to readily identify refusals on the file.

Page 9 of 30

IV. Questionnaire Items Eligible for Imputation
The following items were eligible for imputation in the 2007 CJRP.
Section I:
•
•
•
•
•

Question 10b (total persons assigned to beds in the facility)
Question 11 (number of persons age 21 or over assigned to beds in the facility)
Question 12b (number of persons under age 21 assigned to beds in the facility)
Question 13b (number of offenders under age 21 assigned to beds in the facility)
Question 14b (number of nonoffenders under age 21 assigned to beds in the facility)

Section II:
•
•
•
•
•
•
•

Question 2 (juvenile offender’s sex)
Question 3 (juvenile offender’s birth date)
Question 4 (juvenile offender’s race)
Question 5 (placement agency)
Question 7 (juvenile offender’s most serious offense)
Question 9 (juvenile offender’s adjudication status)
Question 10 (juvenile offender’s date of admission to the facility)

Earlier versions of CJRP have included questions about the numbers of locked doors in a facility.
The 2007 CJRP included a set of questions about locked doors which did not ask for numbers of
locked doors. The 2007 questions asked under what circumstances doors were locked. There
was no imputation performed for these questions.

Page 10 of 30

V. Imputation Rates
The facility imputation rates for Section I are shown in Table 8 below. The facility imputation
rate is:

The only missing data in Section I was for the 9 facilities that refused to participate in the 2007
CJRP, so the imputation rates for each item in Section I are (9/2911) x 100.

Table 8.

Section I facility imputation rates in the 2007 CJRP
Item
Percent Imputed
Total persons
0.3
Adults
0.3
Juveniles
0.3
Juvenile offenders
0.3
Juvenile nonoffenders
0.3

The item imputation rates for Section II are shown in Table 9 on the next page. The juvenile
offender item imputation rate is:

Page 11 of 30
Table 9.

Section II item imputation rates in the 2007 CJRP
Item
Percent Imputed
Sex
7.5
Birth month
20.3
Birth day
20.3
Birth year
20.3
Race
20.6
Placed by
2.8
Offense
23.2
Adjudication status
8.5
Admitted month
20.3
Admitted day
20.3
Admitted year
20.3

There are 86,927 juvenile offender records on the imputed file. 21,889 (25.2 percent) juvenile
offender records have at least one item imputed.

Page 12 of 30

VI. Collapsed Facility Type Codes
For imputation purposes, we need to assign a collapsed facility type code (Cat) to every facility.
Cat is the variable on the 2007 file that contains the collapsed facility type code (column 621).
The 2007 Cat code is assigned with the following procedure:
1.

If the agency checks only one facility type box on the 2007 form, the checked box is
mapped to the appropriate collapsed facility type and Cat is set. See Table 10 below.
Note that if the agency only checks “other,” we don’t assign a collapsed facility type here.
Table 10. Cat codes for collapsed facility types
Cat

Collapsed Facility Type

Check box on 2007 form

Detention Center

S1Q501

Shelter

Reception / Diagnostic Center

S1Q503

Training School

S1Q502

Ranch, Camp, or Farm

S1Q506, S1Q507

Halfway House / Group Home

S1Q504, S1Q505

S1Q508, S1Q509

Note: In the table above, the check box naming convention is S1Q5___, where
S1 means section 1, Q5 means question 5, and the last two digits match the
checkbox on the questionnaire.

Page 13 of 30
2.

If the agency checks more than one facility type box on the 2007 form, we determine the
collapsed facility type based on a hierarchy. The Cat code is determined by the highest
checked box on the list. If an agency checks boxes that indicate it is both a reception
center and a training school, it will be coded as a training school, since the training school
is higher up in the hierarchy than the reception center is. The hierarchy is shown in Table
11 below.
Table 11. Hierarchy to be used when more than one
collapsed facility type indicated on 2006 CJRP
Cat

Collapsed Facility Type

Check box on 2007 form

Training School

S1Q502

Detention Center

S1Q501

Reception / Diagnostic Center

S1Q503

Ranch, Camp, or Farm

S1Q506, S1Q507

Shelter

S1Q508, S1Q509

Halfway House / Group Home

S1Q504, S1Q505

Note: In the table above, the check box naming convention is S1Q5___, where
S1 means section 1, Q5 means question 5, and the last two digits match the
checkbox on the questionnaire.

Page 14 of 30
3.

If the Cat code is still missing, we look at the information provided on the 2006 Juvenile
Residential Facility Census. We use a hierarchy here, because an agency could have
checked more than one box on the 2006 JRFC. The Cat code is determined by the
highest checked box on the list. If an agency checks boxes that indicate it is both a
detention center and a shelter, it will be coded as a detention center, since the detention
center is higher up in the hierarchy than the shelter is. The hierarchy is shown in Table 12
below.
Table 12. Hierarchy to be used when more than one
collapsed facility type indicated on 2006 JRFC
Cat

Collapsed Facility Type

Check box on 2006 JRFC

Training School

S1Q13_02

Detention Center

S1Q13_01

Reception / Diagnostic Center

S1Q13_03

Ranch, Camp, or Farm

S1Q13_05, S1Q13_07

Shelter

S1Q13_08, S1Q13_09

Halfway House / Group Home

S1Q13_04, S1Q13_06

Note: In the table above, the check box naming convention is S1Q13___, where
S1 means section 1, Q13 means question 13, and the last two digits match the
checkbox on the questionnaire.

If the Cat code is still missing, we hold the Cat code constant from the 2006 CJRP.

For the refusals in the 2007 CJRP, we hold the facility type answers constant from the
2006 CJRP. This was a change from previous approaches implemented due to data
quality concerns from one facility on the 2006 JRFC.

If a facility is still missing a Cat code, OJJDP will assign the code. [This situation
occurred for 1 facility in the 2007 CJRP. The facility is new, and marked the “other” box
and specified that it’s an independent living facility. It was classified as group home /
halfway house.]

2,493 Cat codes were assigned because the facility checked only one facility type box on the
2007 CJRP form, 377 Cat codes were assigned because the facility checked multiple facility type
boxes on the 2007 CJRP form, 15 Cat codes were assigned based on the facility responses on the
2006 JRFC form, 25 Cat codes were assigned based on the 2006 CJRP Cat codes, and 1 Cat code
was assigned based on guidance from OJJDP.

Page 15 of 30

VII. Imputation Methodology for Section I Data
The only missing data in Section I was for the refusal facilities.
Handling refusals
The 9 facilities that are refusals for the 2007 CJRP did respond to the 2006 JRFC, so we have
prior year data for all 9 facilities. We calculated average 1-year growth rates by imputation cell
and then applied those rates to the prior year data. When we apply the growth rate to prior year
data, we round the result to the nearest whole number.
The 1-year growth rate is calculated for every facility that reported data in both the 2006 JRFC
and the 2007 CJRP. The 1-year growth rate is the 2007 data item divided by the 2006 data item.
The imputation cell is all the facilities within a given state and Cat (collapsed facility type) code.
If there are fewer than 15 respondents or less than 70 percent response in the imputation cell, we
collapse the imputation cell to the national level. Collapsing was not required for the 2007
CJRP.
For all 9 refusals, we applied the average 1-year growth rate to the 2006 JRFC number of
juvenile offenders to impute the juvenile offenders for the 2007 CJRP.
We applied the average 1-year growth rate to the 2006 JRFC number of juvenile nonoffenders to
impute the juvenile nonoffenders for the 2007 CJRP.
We then added the imputed number of juvenile offenders to the imputed number of juvenile
nonoffenders to impute the number of juveniles for the 2007 CJRP.
We applied the average 1-year growth rate to the 2006 JRFC number of adults to impute the
adults for the 2007 CJRP.
We then added the imputed number of juveniles to the imputed number of adults to impute the
total number of persons for the 2007 CJRP.

Page 16 of 30

VIII.

Imputation Methodology for Item
Nonresponse in Section II Data

Change for 2007
The basic methodology for dealing with item nonresponse is still the hierarchical hot deck, which
was used in the 2003 and 2006 CJRP collections. However, for the 2007 CJRP, we impute
juvenile offenders in tribal facilities separately from the juvenile offenders in all other facilities.
Juvenile offenders in tribal facilities are not eligible to be donors for juvenile offenders in nontribal facilities. For more details, please see Chapter X, “Imputing Missing Data for Juvenile
Offenders in Tribal Facilities.”
Randomly imputing day of birth, month of birth, or day of admission
The first missing items that are imputed are:
•

kid_admitted_day if both kid_admitted_month and kid_admitted_year are reported,

•

kid_birth_month if kid_birth_day is missing but kid_birth_year is reported, and

•

kid_birth_day if both kid_birth_month and kid_birth_year are not missing.

When kid_admitted_month and kid_admitted_year are both reported, but kid_admitted_day is
missing, we impute kid_admitted_day by randomly selecting a day based on kid_birth_month.
This prevents the imputation of days that do not exist, such as February 30th.
Some of the records will only have reported kid_birth_year, while other records will only have
reported kid_birth_month and kid_birth_year for the birth date. We randomly assign a birth
month if kid_birth_month is not reported, then randomly assign a day of the birth month based
on kid_birth_month. This prevents the imputation of days that do not exist, such as February
30th.
Age and stay calculations
The reference date of the questionnaire is October 24, 2007. Some facilities may report based on
an alternative reference date. If an alternative reference date has been used, the date is shown in
the alternative reference date field of the data file (columns 612 through 619).
We calculate an age for all records where it is possible to do so. If the facility is reporting based
on an alternative reference date, the age of the juvenile offender is calculated as of the alternative
reference date; otherwise the age of the juvenile offender is calculated as of October 24, 2007.

Page 17 of 30
End users calculate a length of stay variable, which is based on the date that the juvenile offender
was admitted to the facility. If a facility is reporting based on an alternative reference date, the
length of stay is calculated as of the alternative reference date; otherwise the stay is calculated as
of October 24, 2007. We calculate length of stay for all records where it is possible to do so.
Hierarchical hot-deck for item nonresponse
The imputation methodology for item nonresponse in Section II data is a hierarchical hot-deck.
The record requiring imputation is matched to a pool of records where all the information is not
missing, and then a donor record is selected. The missing values in the record requiring
imputation are copied from the donor record. The match is first tried on all available
information. If a match is not found, the match is made less restrictive until a donor record is
found.
The definition of records where all the information is not missing includes those records for
which we only imputed kid_birth_month, kid_birth_day, or kid_admitted_day. These records
are considered eligible donors because if kid_birth_year is not imputed, we have a good idea how
old the offender is, and if kid_admitted_month and kid_admitted_year are not imputed, we have
a good idea how long the offender has been in placement.
The available information for matching is the Cat code, the state where the facility is located, and
any reported data for kid_sex, age, kid_race, kid_placed_by, kid_offense,
kid_adjudication_status, and length of stay.
When imputing kid_adjudication_status, those records with kid_adjudication_status = 08
(convicted in adult criminal court) are never part of the pool of potential donors. We confirm
with the sponsor that there should not be imputed values of 08 (convicted in adult criminal court)
on the final data file.
The advantage of the hierarchical hot-deck method is that imputed values should be consistent
with the rest of the juvenile offender record, because the donor record is a juvenile offender
record that has passed the edits.

Page 18 of 30

IX. Imputation Methodology for Section II Data for
Critical Item Facilities
Changes for 2007
The basic methodology for dealing with nonresponse in critical item facilities is the same as it
was in the 2003 and 2006 CJRP collections. However, for the 2007 CJRP, we impute juvenile
offenders in tribal facilities separately from the juvenile offenders in all other facilities. Juvenile
offenders in tribal facilities are not eligible to be donors for juvenile offenders in non-tribal
facilities. For more details, please see Chapter X, “Imputing Missing Data for Juvenile
Offenders in Tribal Facilities.”
We introduced a new classification of critical item facility in 2007. We noticed in the 2006
CJRP that some facilities will provide a roster of juvenile offenders, but not much information
about the individual offenders. If date of birth, offense, and date of admission are missing for all
the juvenile offenders in the facility, we really don’t have much information to work with.
If we try to impute those records as merely having item nonresponse, we run the risk of using the
same donor over and over within the facility, creating what looks like duplicate records in the
facility. To minimize that risk, we now handle such facilities like critical item facilities, and
have assigned them a code of 2 (Facility responded to Section I of the questionnaire, but date of
birth, offense, and date of admission are missing for all records in Section II of the questionnaire)
in the critical item field.
Background
The edited file has one record per critical item facility if the critical item field is set to 1 (Facility
responded to only the critical items) or 3 (Facility is a refusal; all data on the file for that facility
have been imputed).
If the critical item facility holds juvenile offenders, the Section II data on the record refers to all
the juveniles held by that facility. GOVS tried to find out as much as possible about the types of
juveniles held in critical item facilities.
If kid_sex = 1 in Section II of the critical item facility record, that means the facility only holds
males, while kid_sex = 2 means that the facility only holds females, and kid_sex = 3 means that
the facility holds both males and females.
Some critical item facilities were unable to indicate for which types of offenses they held
offenders (so kid_offense = 88 or 99 for those critical item facilities), some critical item facilities
were able to indicate that they held offenders for offense codes applicable to both underage

Page 19 of 30
persons and adults (so kid_offense = 97 for those critical item facilities), and some critical item
facilities were able to indicate that they held offenders for those offense codes applicable to
underage persons only (so kid_offense = 98 for those critical item facilities).
We generate the required number of juvenile offender records for each critical item facility and
assign kid_id to each juvenile offender record for the critical item facility. Kid_id is a 15
character juvenile identifier. We number the records sequentially within each critical item
facility, starting at 000000000000001. We also replicate the available reported information for
each juvenile offender record within the critical item facility.
If we know that the facility only holds males or only holds females, we do not consider kid_sex
to be imputed.
The edited file may have multiple records if the critical item field is set to 2 (Facility responded
to Section I of the questionnaire, but date of birth, offense, and date of admission are missing for
all records in Section II of the questionnaire). For example, the facility may have two sets of
offenders who have been placed in the facility by two different types of authorities. If date of
birth, offense, and date of admission are missing for all the offenders in that facility, we don’t
have much information to work with. We handle these facilities as critical item facilities rather
than item nonresponse facilities to minimize the amount of duplication in the imputed data.
Hierarchical hot-deck for critical item facilities
The hierarchical hot-deck methodology used for item nonresponse was modified for critical item
facilities. Instead of finding a matching donor pool for an individual juvenile offender record, we
find a donor pool for the critical item facility and then randomly select donors from the pool
without replacement. This modified version of the hierarchical hot-deck requires that the donor
pool have at least as many juvenile offenders as the critical item facility. This requirement
ensures that the imputed juvenile offender records for the critical item facility are not duplicated
within the facility.
The available information for matching is the Cat code, the state where the facility is located, and
any reported data for kid_sex, age, kid_race, kid_placed_by, kid_offense,
kid_adjudication_status, and stay.
When imputing kid_adjudication_status, those records with kid_adjudication_status = 08
(convicted in adult criminal court) are never part of the pool of potential donors. OJJDP does
not want any imputed values of convicted in adult criminal court on the imputed file.
The advantage of the hierarchical hot-deck method is that imputed values should be consistent
with the rest of the juvenile offender record, since the donor record is a juvenile offender record
that has passed the edits.

Page 20 of 30
In the 2003 and 2006 CJRP files, we only used the top 2 levels of the hierarchical hot-deck for
critical item facilities. For the 2007 CJRP, we used as many as 4 levels of the hierarchical hotdeck for critical item facilities.
We had to use 4 levels in states with large numbers of juvenile offenders held in critical item
facilities where the facility reported a relatively uncommon value for who placed the juvenile in
the facility. Specifically, this happened for:
C

a critical item facility in Arizona with juveniles placed by school official, parent or
guardian, or the young person himself or herself,

a critical item facility in the District of Columbia with juveniles placed by a social
services agency, and

a critical item facility in New York with juveniles placed by a social services agency who
were awaiting an adjudication hearing in juvenile court.

X. Imputing Missing Data for Juvenile Offenders in
Tribal Facilities
For the 2007 CJRP, we handled the juvenile offenders in tribal facilities separately from the
juvenile offenders in all other facilities. While juvenile offenders in tribal facilities were not
used as donors for juvenile offenders in all other facilities, we did use juvenile offenders in all
other facilities as donors for juvenile offenders in tribal facilities.
If race was missing for a juvenile offender in a tribal facility, we set race to 4 (American Indian /
Alaska Native, not of Hispanic origin) and set the race imputation flag to 1.
Item nonresponse for juvenile offenders in tribal facilities was handled in the same way as item
nonresponse for juvenile offenders in all other facilities. See Chapter VIII, “Imputation
Methodology for Item Nonresponse in Section II Data,” earlier in this document for more
information.
Missing data for juvenile offenders in tribal critical item facilities was handled in the same way
as missing data for juvenile offenders in all other critical item facilities. See Chapter IX,
“Imputation Methodology for Section II Data for Critical Item Facilities,” earlier in this
document for more information.
There are 113 tribal juvenile offenders in the 2007 CJRP. 77 of those offenders (68 percent)
were held in critical item facilities. Almost all the missing data for tribal offenders is due to the

Page 21 of 30
critical item facilities.
The tribal facility item imputation rates for Section II are shown in Table 13 below.
Table 13.

Tribal facility section II item imputation rates in the 2007 CJRP
Item
Percent Imputed
Sex
68.1
Birth month
68.1
Birth day
68.1
Birth year
68.1
Race
68.1
Placed by
68.1
Offense
68.1
Adjudication status
69.0
Admitted month
68.1
Admitted day
68.1
Admitted year
68.1

Due to the high rates of missing data on the juvenile roster for tribal offenders, non-tribal
offenders were used as donors for 53 (68 percent) of the 78 tribal juvenile offender records
requiring imputation.

Page 22 of 30

XI. Special Handling for Critical Item Units in the
District of Columbia (DC)
During our first attempt at imputing the 2007 CJRP, we calculated the percentage of juveniles
held in critical item facilities by the state of facility. The percentages are shown in Attachment
A. We noticed that 95.7 of all juvenile offenders in DC are held in critical item facilities. This
means that an unusually large percentage of the data in DC are imputed.
Further investigation revealed that there are two large facilities in DC that account for roughly 80
percent of the juvenile offenders held in DC, both of which are critical item facilities for the 2007
CJRP. GOVS called all the critical item facilities in DC to ask if they could provide the racial
distribution of the offenders. 7 of the critical item facilities in DC were able to give us the racial
distribution of their offenders.
Because there were so few fully-reported juvenile offender records in DC facilities, the
imputation system defaulted to using juvenile offender records from across the country. We
wrote additional SAS programs to reimpute the DC facilities after all other imputation processing
was completed. The special handling is last so that no other imputed values on the file would
change.
For the facilities that were able to provide the racial distribution, the kid_race_flag is set to 0
during the special handling. We then imputed the facilities as we would any other critical item
facility where race had been reported.
After we were finished with the special handling for the 7 facilities in DC, we calculated Section
II item imputation rates by state. Those rates are shown in Attachment B.

Page 23 of 30

XII. Caution When Comparing State Data over Time
In the 2003 CJRP documentation, we noted that 84.6 percent of all juvenile offenders in DC were
held in critical item facilities, which meant that an unusually large percentage of the data in DC
was imputed in 2003. For 2007, 95.7 percent of all juvenile offenders in DC are held in critical
item facilities. Comparing juvenile offenders held in DC facilities across the 2003, 2006, and
2007 data collections is not recommended, due to the extreme levels of missing data for the
juvenile offenders.
Attachments A and B show some high levels of juvenile roster item imputation for 2007.
Illinois, Mississippi, and Wyoming had more than half of their juvenile offenders in facilities that
only reported critical items. States with 30 percent or more imputation by item included Arizona,
Arkansas, Colorado, Florida (kid_race), Illinois, Iowa (kid_offense), Maine, Montana, Nevada,
New Mexico, New York, South Carolina, Utah, and West Virginia.
In the 2006 CJRP, both Illinois and Rhode Island had more than half of their juvenile offenders
in facilities that only reported critical items. States with 30 percent or more imputation by item
included Colorado (kid_offense), Illinois (kid_birth_date, kid_race, kid_offense,
kid_admitted_month and kid_admitted_day), Rhode Island (kid_sex, kid_birth_date, kid_race,
kid_offense, kid_adjudication_status, and kid_admitted_date), and Wyoming (kid_sex,
kid_birth_date, kid_race, kid_placed_by, kid_offense, and kid_admitted_date).
Users should be aware the differences in DC data from 2003 to 2006 may be due in part to the
high levels of imputation for DC in 2003, and from 2006 to 2007 due to high levels of imputation
for DC in 2007. Similarly, the differences in Colorado, Illinois, Rhode Island, and Wyoming
data from 2003 to 2006 may be due in part to the high levels of imputation for those states in
2006. The differences in Arizona, Arkansas, Colorado, Florida, Illinois, Iowa, Maine, Montana,
Nevada, New Mexico, New York, South Carolina, Utah, and West Virginia from 2006 to 2007
may be due in part to the high levels of imputation for those states in 2007.
Attachment C shows items by state for the 1997 through 2007 CJRP data collections. If the item
imputation rate was 30 percent or more for a given year, the year is shown in the cell of the table.

Page 24 of 30

XIII.

References

Census of Juveniles in Residential Placement Databook, currently online at
http://ojjdp.ncjrs.org/ojstatbb/cjrp/
Dorinski, Suzanne M. “How to collapse 2003 CJRP facility types for use in imputation,” memo
dated October 21, 2003. (Most of the memo has been reproduced in this document.)
Dorinski, Suzanne M. “How to impute Section I data,” memo dated January 8, 2004.
Dorinski, Suzanne M. “How to calculate age on individual juvenile records in 2003 CJRP,”
memo dated February 25, 2004. (Most of the memo has been reproduced in this document.)
Dorinski, Suzanne M. “How to impute for item nonresponse on juvenile offender records in
2003 CJRP,” memo dated April 1, 2004.
Dorinski, Suzanne M. “Documentation of the Imputation Methodology for the 2003 Census of
Juveniles in Residential Placement FOR INTERNAL USE ONLY,” memo dated December 1,
2004.
Dorinski, Suzanne M. “Documentation of the Imputation Methodology for the 2006 Census of
Juveniles in Residential Placement FOR INTERNAL USE ONLY,” memo dated March 22,
2007.
Dorinski, Suzanne M. “Documentation of the Imputation Methodology for the 2007 Census of
Juveniles in Residential Placement FOR INTERNAL USE ONLY,” memo dated January 15,
2009.

Page 25 of 30

Attachment A:

Percentage of Juvenile Offenders Held in Critical Item
Facilities by State of Facility

(Percentages of 30 or more have been highlighted)

Facility
State
Alabama

Percentage in
critical item
facilities

Juvenile
Offenders
1,667

11.5

321

19.6

1,580

39.1

819

31.7

California

14,033

7.5

Colorado

1,810

32.7

Connecticut

359

7.2

Delaware

301

0.0

D.C.

277

95.7

Florida

5,747

26.3

Georgia

2,733

9.0

Hawaii

130

2.3

Idaho

525

8.6

Illinois

2,552

73.1

Indiana

2,802

8.5

Iowa

1,257

22.1

Kansas

1,120

15.8

Kentucky

1,117

0.4

Louisiana

1,348

17.3

Maine

203

36.9

Maryland

862

4.4

Massachusetts

1,069

22.1

Michigan

2,616

8.4

Minnesota

1,439

21.5

449

52.1

Missouri

1,243

14.2

Montana

208

34.1

Nebraska

734

4.8

1,057

43.0

153

14.4

1,664

26.0

443

35.7

3,471

46.6

Alaska
Arizona
Arkansas

Mississippi

Nevada
New Hampshire
New Jersey
New Mexico
New York

Page 26 of 30

Attachment A:

Percentage of Juvenile Offenders Held in Critical Item
Facilities by State of Facility

(Percentages of 30 or more have been highlighted)

Facility
State
North Carolina

Percentage in
critical item
facilities

Juvenile
Offenders
1,039

9.8

203

16.7

4,316

7.5

897

17.4

Oregon

1,291

16.2

Pennsylvania

5,511

21.4

Rhode Island

317

0.0

1,201

46.8

509

29.7

Tennessee

1,320

17.0

Texas

7,034

10.8

Utah

1,050

38.7

Vermont

2.3

Virginia

2,163

6.0

W ashington

1,526

22.9

575

31.3

W isconsin

1,493

15.7

W yoming

329

55.0

86,927

19.5

North Dakota
Ohio
Oklahoma

South Carolina
South Dakota

W est Virginia

TOTAL

Attachment B:

Page 27 of 30

Section II Item Imputation Rates by State of Facility

(Imputation rates of 30 percent or more have been highlighted)
Kid_birth_
Facility
State
Alabama

Kid_
Offenders sex

month

day

year

Kid_
race

Kid_
placed_
by

Kid_
offense

Kid_
adjudication_
status

month

Kid_admitted_
day

year

1,667

2.0

11.5

11.8

0.8

11.8

3.6

11.5

321

18.4

19.6

4.0

24.6

21.2

19.6

1,580

32.5

39.1

39.2

0.8

39.1

33.1

39.1

819

28.8

33.2

33.1

4.2

32.7

1.7

32.4

California

14,033

2.6

7.7

8.3

0.6

14.0

6.3

8.1

Colorado

1,810

3.0

32.7

32.9

4.4

47.5

3.1

32.7

Connecticut

359

0.3

7.2

3.9

35.1

0.3

7.2

Delaware

301

0.0

7.6

0.0

D.C.

277

40.1

95.7

2.9

0.0

95.7

43.7

95.7

Florida

5,747

0.1

27.3

30.3

0.0

29.4

2.5

26.3

Georgia

2,733

6.3

9.0

9.1

3.7

12.0

6.1

9.0

Hawaii

130

0.0

2.3

0.0

3.1

2.3

Idaho

525

2.7

8.6

9.1

0.0

9.1

6.1

8.6

Illinois

2,552

15.9

73.1

1.7

73.1

17.3

73.1

Indiana

2,802

7.2

8.6

12.2

1.8

8.6

8.5

Iowa

1,257

3.1

22.4

22.1

1.4

30.6

1.8

22.2

Kansas

1,120

7.4

15.8

16.9

0.0

21.5

9.9

15.8

Kentucky

1,117

0.4

3.9

1.5

0.4

Louisiana

1,348

6.8

25.5

0.0

25.6

11.6

25.5

Maine

203

33.5

36.9

33.5

36.9

33.5

36.9

Maryland

862

0.0

4.8

4.5

4.4

0.0

12.2

1.2

4.4

Massachusetts

1,069

4.1

22.2

22.1

4.4

22.5

12.3

22.8

Michigan

2,616

7.0

8.4

9.2

5.2

9.8

8.4

Minnesota

1,439

14.4

21.5

21.8

12.2

23.7

16.1

22.8

449

9.6

52.1

0.0

52.3

14.7

52.1

Missouri

1,243

13.1

14.4

14.2

0.1

15.4

11.7

14.3

Montana

208

20.2

34.1

9.1

34.1

1.9

34.1

Nebraska

734

1.5

4.8

4.5

5.3

0.1

4.8

1,057

7.6

43.0

0.0

43.0

7.7

43.0

153

0.0

22.9

14.4

0.0

14.4

0.0

14.4

1,664

17.4

26.0

15.9

26.0

19.2

26.0

443

35.2

35.7

35.9

0.0

42.0

35.4

35.7

New York

3,471

4.9

46.7

47.9

1.7

47.9

9.8

46.8

North Carolina

1,039

5.8

9.8

4.0

22.7

3.3

9.8

203

12.3

16.7

17.2

16.7

19.2

0.0

16.7

Alaska
Arizona
Arkansas

Mississippi

Nevada
New Hampshire
New Jersey
New Mexico

North Dakota

Attachment B:

Page 28 of 30

Section II Item Imputation Rates by State of Facility

(Imputation rates of 30 percent or more have been highlighted)
Kid_birth_
Facility
State
Ohio

Kid_
Offenders sex

month

day

year

Kid_
race

Kid_
placed_
by

Kid_
offense

Kid_
adjudication_
status

month

Kid_admitted_
day

year

4,316

5.1

7.5

7.8

0.4

7.9

5.6

7.5

897

7.0

17.4

2.7

20.3

13.9

17.4

Oregon

1,291

9.3

16.4

16.2

0.0

17.0

11.2

16.2

Pennsylvania

5,511

5.6

21.9

22.0

12.1

24.7

4.4

22.2

Rhode Island

317

0.0

0.6

0.0

0.3

0.0

1,201

4.2

78.3

78.0

0.7

78.5

12.5

78.3

509

20.8

29.7

29.9

4.9

29.9

24.2

29.7

Tennessee

1,320

3.9

17.0

1.7

23.5

11.1

17.0

Texas

7,034

8.8

10.8

11.1

0.2

11.7

8.9

10.8

Utah

1,050

29.7

38.7

40.6

14.6

38.9

15.7

38.7

Vermont

0.0

2.3

0.0

2.3

0.0

9.1

Virginia

2,163

3.2

8.2

6.0

2.4

9.1

2.7

6.0

W ashington

1,526

20.6

22.9

26.1

2.3

23.6

12.2

22.9

575

17.7

31.3

11.0

31.8

12.0

31.3

W isconsin

1,493

7.8

18.6

1.7

20.6

12.7

18.6

W yoming

329

36.2

55.0

0.9

55.0

10.0

55.0

86,927

7.5

20.3

20.6

2.8

23.2

8.5

20.3

Oklahoma

South Carolina
South Dakota

W est Virginia

TOTAL

Attachment C:

Page 29 of 30

Data Quality Issues in CJRP Over Time

1997 through 2007 data collections
(Year is shown in table cell when imputation rate is 30 percent or more)
State

Sex
month

Birth
day

year

Race

2007
2007

Alabama
Alaska
Arizona
2007
Arkansas
California
Colorado

Connecticut

2007
2007

2007
2001

2003
2007

Placed Offense Adjudication
Admitted
by
status
month day
year
2001
2001 2001
1999
2007
2007

2007
2007

2007
2001

Delaware
D.C.
2003
2007

2007
2007

2007
2001

2003
2007

1999

2006
2007

2007

2003
2007

2007

Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
M aine
M aryland

2007
2007

1999
2001
2006
2007
2001
2007
1999
1999
2003
2007

2007

2006
2007

2007

2007
2001

2007

2001

1997
1999
2007

2007

1997
2007

2007

1997
1999
1999
2007
1999
2007

1999
2007

2007

2003
M assachusetts
M ichigan
M innesota
M ississippi

1997
1999

M issouri
M ontana
Nebraska
Nevada
New Hampshire
New Jersey
New M exico
2007

1999
2007

2007

New York

1997

2007

Attachment C:

Page 30 of 30

Data Quality Issues in CJRP Over Time

1997 through 2007 data collections
(Year is shown in table cell when imputation rate is 30 percent or more)
State

Sex
month

North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota

Birth
day

Race
year

Placed Offense Adjudication
Admitted
by
status
month day
year
1999

2006

1999

2006
2007
1997
1999

2007

2006
2007

1999
2006
2007
1997
1999
1999
2001

Tennessee
Texas
Utah
Vermont
Virginia
W ashington
W est Virginia
W isconsin
W yoming

2006
2007

1999
2007

2006

2007
1999
2006
2007

2006
1997
1999

2006
2007
1997
1999

2007

2006
2007

File Type	application/pdf
File Modified	2009-01-26
File Created	2009-01-26