201101-1205-002ssb20110728

B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS

1.  Describe (including a numerical estimate) the potential respondent universe and any sampling or other respondent selection methods to be used.  Data on the number of entities (e.g., establishments, State and local government units, households, or persons) in the universe covered by the collection and in the corresponding sample are to be provided in tabular form for the universe as a whole and for each of the strata in the proposed sample.  Indicate expected response rates for the collection as a whole.  If the collection had been conducted previously, include the actual response rate achieved during the last collection.

Administrative data will be collected from all 10 grantees on the estimated 6,850 participants in these 10 AWI grants. No sampling will be used and all data collection and subsequent analysis will cover 100 percent of the population included in the database. In all reports and other publications and statements resulting from this research, no attempt will be made to draw inferences to any population other than the set of units that responded to the data collection effort.

The response rate will be 100 percent, as the collection will cover only those individuals who are enrolled under the grant, for whom data must be submitted, in accordance with the HGCBJT reporting requirements (already approved by OMB). The data collection has not been conducted previously.

2. Describe the procedures for the collection of information including: 

        Statistical methodology for stratification and sample selection,

        Estimation procedure,

        Degree of accuracy needed for the purpose described in the justification,

        Unusual problems requiring specialized sampling procedures, and

        Any use of periodic (less frequent than annual) data collection cycles to reduce burden.

Statistical Methodology for Stratification and Sample Selection: There will be no stratification or sample selected, as the entire population will be used. Administrative data will be collected from all 10 grantees on the estimated 6,850 participants in these AWI grants. No sampling will be used and all data collection and subsequent analysis will cover 100 percent of the population included in the database. In all reports and other publications and statements resulting from this research, no attempt will be made to draw inferences to any population other than the set of units that responded to the data collection effort.

Estimation Procedure: The analysis of data will make use of frequency distributions, means and cross-tabulations, which will provide basic information about customers’ experiences in the program. The relationship between certain program features and participant outcomes will also be examined using regression analysis.

The size of the respondent universe will ensure there are a sufficient number of respondents to conduct subgroup analyses. Standard formulas for comparing differences in means and proportions will be used. In addition, standard regression techniques will be used to determine the presence of a relationship between specific program components and participant outcomes. Thus, for continuous dependent variables (e.g. wages) ordinary least squares regression analysis will be employed, as given by: y = α + β₁x₁ + β₂x₂+…β_nx_n + e, in which y is the outcomes, α is the constant (representing the intercept), β’s represent the regression coefficients for the corresponding x (independent) terms, and e represents the error term reflected in the residuals.

For analysis of binomial dependent variables (e.g. completed training, obtained employment), logistic regression techniques will be used, as given by: z = α + β₁x₁ + β₂x₂ +…β_kx_k,

in which z is the logit (log odds) of the dependent variable, α is the constant, and β’s represent the logistic regression coefficients for the corresponding x (independent) terms.

Degree of Accuracy Needed: Because sampling will not be employed, results should be an accurate reflection of the relevant universe of program participants. Results are generalizable only to the characteristics and outcomes of these participants.

Unusual Problems: There are no unusual problems for this data collection requiring specialized sampling procedures.

Requiring Specialized Sampling: As noted, there is no sampling being utilized.

Use of Periodic Data Collection Cycles: The data from participants and from grantees will be collected only once.

3.   Describe methods to maximize response rates and to deal with issues of non-response.  The accuracy and reliability of information collected must be shown to be adequate for intended uses.  For collections based on sampling, a special justification must be provided for any collection that will not yield reliable data that can be generalized to the universe studied.

Methods to Maximize Response Rates: Methods to maximize response rates include making the provision of data for the evaluation a requirement of the grant agreement and requiring key participant data to be collected and provided to DOL on a quarterly basis on items required as part of the HGCBJT Common Measures framework, already approved by OMB (and under which the AWI grants are being conducted). Thus there will information on 100 percent of participants since, by definition, they must be enrolled and, in order to be enrolled, the provider must collect data on a host of required items (i.e., age, employment or retirement status, veteran status, initial services, post- program disposition). However, there may be some items where the provider does not provide information for particular participants. For example, we expect to have missing responses for information on simultaneous enrollment in partner programs and on intermediate interventions. Collection of data on these other items will also be encouraged as part of technical assistance from DOL.

Methods to Deal with Issues of Non-Response: The response rate will be 100 percent overall, as noted above, on items in the HGJBT framework. However, there may be some non-responses on individual data elements, as grantees may misreport or fail to record information for some items. However, since there is no sampling, there will be no statistical adjustments for non-response on these items. The actual number of responses will be noted in all tables and appropriate caveats provided in the text as to the number of responses per element.

4. Describe any tests of procedures or methods to be undertaken.  Testing is encouraged as an effective means of refining collections of information to minimize burden and improve utility.  Tests must be approved if they call for answers to identical questions from 10 or more respondents.  A proposed test or set of test may be submitted for approval separately or in combination with the main collection of information.

Description of Tests of Procedures or Methods: The overall format for obtaining information from grantees and participants was tested as part of the initial administration of the Common Measures under the HGCBJT, already approved by OMB. The collection of specific additional data items and the voluntary data system has been tested by AWI grantees.

5. Provide the name and telephone number of individuals consulted on statistical aspects of the design and the name of the agency unit, contractor(s), grantee(s), or other person(s) who will actually collect and/or analyze the information for the agency.

This survey will employ no sampling. Individuals consulted on use of multivariate analysis include:

Name	Affiliation	Telephone Number
Dr. Ronald D’Amico	Social Policy Research Associates	(510) 763-1499
Jeffrey Salzman	Social Policy Research Associates	(510) 763-1499
Jill Leufgen	Social Policy Research Associates	(510) 763-1499
Michelle Derr	Mathematica Policy Research	(202) 484-4830

3