G National Agricultural Statistics Service Comments

Appendix G:

National Agricultural Statistics Service Comments

FNS Information Collection Package for Scanner Capability Assessment of SNAP – Authorized Small Retailers (SCANR) Study.

OMB control number 0584-NEW

Reviewer:

Josephine Akato, Mathematical Statistician

Summary, Estimation, and Disclosure Methodology Branch

National Agricultural Statistics Service

United States Department of Agriculture

General Comments:

This docket is well described and documented. Data collection procedure is well thought-out and designed to achieve the goals of the study.

Appendices

Nothing to note.

Part A: Final

Nothing to note.

Part B: Final

Method to maximize response rate and to deal with non-response is well detailed, and the sampling population well defined. A national level response rate of 80% is reasonable and can be achieved with the survey design. In the event of poor response, there is already a plan in place for a nonresponse bias study. Pre-testing the instrument will reduce non-sampling error.

A response rate of 80% is targeted for SCANR survey (page B12) yet 70% response rate is assumed for the follow-up Interviews for Retails. Is there a reason not to assume 80% response rate for follow-up interviews?

On page B-12, 50 interviews out of 72 does not achieve 70% response rate. Might be 71 retails as shown in Table 1 of AppendixG1.

Please include more details about why a convenience sample for the industry interviews mentioned on page B-12 is sufficient since it is not representative.

Please include more details about the reserve sample will be released if needed since the initial sample was drawn using a systematic stratified design.