SUPPORTING STATEMENT PART B
OMB No. 0584-0609
WIC
Participant and Program Characteristics Study
2020 and 2022
August 29, 2019
Project Officer: Anna Potter Clifford
Office of Policy Support
Food and Nutrition Service
U.S. Department of Agriculture
3101 Park Center Drive
Alexandria, VA 22303
703.305.2719
Part B: Collection of Information Employing Statistical Methods 1
B.1 Respondent Universe and Selection Methods 1
B.2 Procedures for the Collection of Information 3
B.3 Methods To Maximize Response Rates and the Issue of Non-Response 6
Tables
Appendices
Appendix A. Regulations Authorizing Biennial Reports
Appendix B. Regulations Defining Participant Characteristics
Appendix C. Notification Letter for Guidance to WIC State Agencies
Appendix D. Guidance for WIC State Agencies Providing Participant Data
Appendix E.1 Comments from Consultation Outside FNS
Appendix E.2 Response to Comments from Consultation Outside FNS
Appendix F. Insight Policy Research Confidentiality Pledge
Appendix G. Excel Table of Total Public Burden Hours and Respondent Costs
Appendix H. Data Transmittal Worksheet
Appendix I. Nutritional Risk Worksheet
Appendix J. Contact Request Form
Appendix K. Sample Reminder Email to WIC State Agencies
Appendix L. Sample Follow-up Email to WIC State Agencies
Appendix M. Telephone Script for Follow-Up Calls After Data Submission
Appendix N.1 NASS Comments on OMB Package
Appendix N.2 FNS Response to NASS Comments on OMB Package
Appendix O. Excerpt from WIC PC2018 Final Data Cleaning Plan
Part
B: Collection of Information Employing
Statistical Methods
Describe (including a numerical estimate) the potential respondent universe and any sampling or other respondent selection method 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.
The sampling plan described in part B is a contingency plan for WIC State agencies (SAs) that are not able to provide a complete census of participants from their eligibility systems. The purpose of this section is to document any statistical procedures used for the WIC PC2020 and PC2022 studies. We estimate the collection will have a 100-percent response rate in PC2020 and PC2022 as it has had in previous years.
The data collection normally involves a census of participants enrolled in WIC in the study month of April for each of the 90 SAs that administer WIC; these SAs serve the 50 States, the District of Columbia, 5 U.S. territories, and 34 Indian Tribal Organizations (ITOs). To submit a census of participants, WIC SA personnel download information on all participants certified to receive WIC during the reference month (April) and submit that file to the contractor conducting the data collection via a secure FTP server. WIC SA personnel abstract the data needed for the study from the certification system, so no actual field contact with participants is necessary. For PC2018, data on 7,836,501 active participants were collected; for PC2016, data on 8,815,742 were collected. The PC2020 and PC2022 studies are expected to collect data on a similar number of participants.
These data may not be available due to unforeseen circumstances—for example, if a management information system upgrade results in loss of data, or if staff- or technology-related problems cause widespread data integrity issues. As in past WIC PC studies, the first attempted solution would be to have the WIC SA submit data from another month close to the study month, such as March or May. However, in the event a WIC SA was not able to provide a census of participants for any month, the contractor would advise the WIC SA on how to employ the statistical procedure outlined in section B.2 of this document to sample a subset of records. Prior to implementing this approach, FNS will consult with OMB, as stated in the terms of clearance (see Part A, Terms of Clearance).
For a WIC SA unable to provide a census of participants for any month, the universe would include all WIC participants enrolled to receive benefits through the WIC SA during a month the SA is able to sample. This could vary greatly depending on the size of the SA; those in some large States serve more than a million participants, whereas those in some smaller ITOs serve several dozen. Because the process outlined here would serve as an alternative means of data collection, the size of the universe and the size of the sample would differ based on the details of the situation. The sampling plan is not the primary means of data collection, so many normally required details cannot be reported here because these details would vary depending on the specifics of the situation that necessitated the use of sampling.
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
Any use of periodic (less frequent than annual) data collection cycles to reduce burden
This sampling plan is an alternate means of data collection for WIC SAs that are unable to provide a census of participants. We do not expect to use sampling during data collection. However, in the interest of covering all contingencies, the following sampling plan was developed for use in the event that a WIC SA cannot provide a census of participants for any month.
B.2.1 Statistical Methodology for Stratification and Sample Selection. The target population for the data abstraction would include all WIC participants enrolled to receive benefits through the WIC SA that was unable to provide a full census of participants. The timeframe would be the 1 to 3 months surrounding the April reference month of the study year.
The sampling frame for the affected WIC SA would be constructed from the administrative records of WIC participants known to be enrolled through the WIC SA and eligible for WIC during the target timeframe of the 1 to 3 months surrounding the April reference month of the study year.
The WIC PC studies aim to represent the entire population of WIC SA participants; therefore, we would develop a procedure for the affected WIC SA to select a stratified, systematic sample of 380 of its WIC participants. Prior to sample selection, we would first ask the WIC SA to stratify the data based on seven certification categories (pregnant women younger than 18 and those aged 18 or older, breastfeeding women, postpartum nonbreastfeeding women, infants 0–4 months of age, infants 5–11 months of age, and children aged 1–4).
The sample size of 380 WIC participants would be allocated proportionally to each stratum based on the size of that stratum related to the size of the WIC population for that WIC SA. One benefit to this approach is that all weights would be equal; as such, there would be no “oversampling” of certain strata causing variation in the weights.
The affected WIC SA would then be asked to perform systematic sampling within each of the stratum (the seven certification categories). This method would involve sorting the participants by local agency and numbering the WIC participants in the population from 1 to N (N = total records in the population). To select a sample of n participants, a participant would be selected at random from the first k participants (k = N/n), and every kth participant thereafter until the appropriate number of participants was achieved in the stratum. In this way, each participant in the sampling frame would be given a known, nonzero probability of selection so that weighted inferences could be made about the entire population of participants.
The WIC SA would then extract data from its eligibility system or case files and submit the data to the contractor for WIC PC.
B.2.2 Estimation Procedure. The purpose of the analyses is to estimate various characteristics about the WIC population enrolled through the affected WIC SA. Following receipt of the sample from the WIC SA, sample weights (or base weights) for WIC participants in the sample would be prepared based on initial probability of selection. The resulting weighting scheme would inflate participant records to represent the entire universe of WIC participants in the WIC SA.
B.2.3 Degree of Accuracy Needed for the Purpose Described in the Justification. The overall sample would be designed to achieve (within WIC SA estimates) 95-percent, two-tailed confidence intervals of between 3.0 and 5.0 percentage points for the affected WIC SA. This design would provide reliable estimates of the characteristics of the WIC SA that was unable to provide a census of participants.
We also have processes for validating the data submitted by WIC SAs to achieve the accuracy needs for the purpose described in the justification. After data are submitted, Insight runs a diagnostic protocol and contacts the WIC SA (see appendices L and M) to address concerns or document the reasons for any unanticipated findings in the data.
B.2.4. Unusual Problems Requiring Specialized Sampling Procedures. The sampling strategy described in this section is an alternative strategy to be used in the event that a WIC SA cannot provide a census data file. We do not anticipate using sampling or any additional specialized procedures.
B.2.5. Any Use of Periodic Data Collection Cycles to Reduce Burden. WIC PC data are collected every 2 years.
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.
We anticipate receiving data on 100 percent of WIC participants from each WIC SA. Since 2012 only one WIC SA could not provide data for 100 percent of participants. In this instance, data were weighted up to the full population by participant category.
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 tests may be submitted for approval separately or in combination with the main collection of information.
Each WIC SA already collects these data in the course of administering WIC; as such, no instrument or procedure would need to be tested before data collection could occur.
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.
Betsy Thorn and Nicole Kline of Insight Policy Research, Inc., (Insight) will provide consultation on the statistical aspects of the design. Insight is also responsible for collecting and analyzing all data for this study. Mathematical Statistician, Josephine Akato of the National Agricultural Statistics Service (NASS) reviewed both Part A and Part B of this OMB package and provided comments (see appendix N.1). FNS’s response to those comments are in Appendix N.2. Table B.5.1 identifies all individuals providing statistical consultation, and/or responsible for data collection and analysis.
Table B.5.1. Individuals Responsible for Statistical Aspects and Data Collection and Analysis
Name |
Title (Project Role) |
Organizational
Affiliation |
Phone Number |
Betsy Thorn |
Executive Project Director |
Insight Policy Research, Inc. 1901 N. Moore Street, Suite 1100 Arlington, VA 22209 |
703.504.9498 |
Nicole Kline |
Deputy Project Director |
Insight Policy Research, Inc. 1901 N. Moore Street, Suite 1100 Arlington, VA 22209 |
703.504.9488 |
Josephine Akato |
Mathematical Statistician |
National Agricultural Statistics Service 1400 Independence Ave SW, Washington, DC 20250 |
202.720.2641 |
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Elaine Ayo |
| File Modified | 0000-00-00 |
| File Created | 2021-01-14 |