NASS Comments

APPENDIX Z

AGE 3 EXTENSION INFANT AND TODDLER FEEDING PRACTICES STUDY – II

NATIONAL AGRICULTURE STATISTICS SERVICE COMMENTS

NASS Review of OMB No. 0584-0580

WIC Infant and Toddler Feeding Practices Study-2 (WIC ITFPS-2):

Age 3 Extension

November 12, 2014

Review Conducted by Tom Pordugal at NASS / MD / SEDMB / CS

General Comments: Part B, Sections 1 and 2

This study seems to be well designed. The estimation methodology takes into account the objectives of the project and the complexities of the data. The methods used should provide a good understanding of WIC program and policy related to the nutrition and wellbeing of children up to their 3rd birthday.

Response: Thank you.

B.1 Respondent Universe and Sampling Methods

The Age 3 Extension is a follow-on study from participants that were recruited in the base cohort study, but the sample design from the base study is not provided.

Response: We can provide information on the sample design for the base study, if requested.

B.1 Response Rates and Non-Response Bias Analysis

The projected response rates for the 30- and 36-month interviews are 64 percent and 60 percent, respectively. The respondents in the Age 3 Extension may be different from the population with respect to characteristics used in the analysis. The approach used from the base study to examine bivariate cross tabulations of data available for the eligible, enrolled cohort by response status at 36-month interview to check for evidence of non-response bias at 36-month interview is a good one. Also, in developing the groups for imputation of missing data on the non respondents or partial respondents, the good respondents should be cross-checked to those respondents from the base study for data quality purposes.

Response: Thank you. Regarding imputation, very little imputation of 30- and 36-month interview data is planned. However, we will examine differences in characteristics (using characteristics collected in the base study) between the nonrespondents or partial respondents and the respondents, and will account for such differences by including the relevant variables in the imputation procedure.

B.2 Degree of Accuracy Needed for the Purpose Described in the Justification

There is a typo in the second sentence of the second paragraph. Additionally, for subgroup estimates (for key subgroups), a 90 percent confidence interval should have a half-width of no more than 5 10 percentage points.

Response: That sentence is actually correct as stated. It may seem odd that the precision requirement (in terms of half-width of the confidence interval) for the 90 percent confidence interval is the same as the precision requirement for the 95 percent confidence interval. However, these requirements pertain to different types of estimates and different subgroups.