National Food Study Pilot
Response to NASS Comments on the ICR Package
On July 27, 2016 NASS reviewed the ICR package for the National Food Study (NFS) Pilot. Their thorough, careful, and insightful review included questions and comments on the following 12 documents. There were no changes requested to the documents not listed below.
We also made some minimal changes to all documents since NASS reviewed them that were not in response to NASS’s request. All these have been documented via track changes so that NASS can see all key changes since their review. This document focuses only on the changes made as a result of NASS’s review.
Supporting Statement Part A.
In Section A.1, Circumstances Making the Collection of Information Necessary, NASS requested: “Probably note use of incentives in this paragraph.”
We added a sentence about incentives to the paragraph. Details of the incentive approach are discussed in Section, A.9, Explanation of Any Payment or Gift to Respondents.
In Section A.2, Purpose and Use of the Information, NASS asked: “Do you intend to provide all forms (advance letter, info sheet, etc.) in English and Spanish or just the screener, interview, and food log? Do you need to provide Spanish versions of these documents?”
We intend to translate into Spanish all material that will be used in interactions with Spanish speaking respondents. These include the Advance Letter (A1), Information Sheet (A2), Disclaimer (A3), Consent Form (A4), Parental Consent Form (A5a), Individual Assent-Consent Form (A5b), Household Screener (B1), Initial Interview (B2), Food Log (B3), Income Worksheet (B4), Meals and Snacks Form (B5), Nonresponse Followup (B6), Final Interview (B7), and Respondent Feedback (B8). We had not planned on including these translations with the ICR submission but now plan to do so.
When discussing training in Section A.2, we stated that we would only train respondents age 16 and over. NASS asked: “Why not age 11 and over? In the previous bullet point, you indicate an interest in having minors 11-18 respond. App B5 mentions household members as young as 11.”
This was an error. We have edited the training statement to include children between the ages of 11 and 15 assuming we receive parental permission to have them participate for themselves.
We have now spelled these out.
Supporting Statement Part B.
In Section B.1, Respondent Universe and Sampling Methods, NASS asked: “What does ABS stand for?”
Address-based sampling is now spelled out.
The word non-SNAP was not used consistently in the document. It was sometime hyphenated and sometimes the hyphen was missing. NASS requested that we be consistent.
It now reads non-SNAP throughout all documents.
Information Sheet (Appendix A2)
NASS commented on a particular Q and A on the sheet as follows: “This is inconsistent with #6 on A4 which mentions tracking all meals and snacks. This sentence (response) does not make sense as a response to this question.”
In A4, Consent Form, item 6 refers to the “Meals and Snacks Form.” Respondents must indicate on that form whether they had meals and snacks but not the details of the meals and snacks eaten. We have modified the Q and A. We did not make any changes to the consent form as a result of this comment as we felt the problem lay with the Q and A..
OLD VERSION:
What if all the food I ate on a particular day came from my refrigerator or cupboard?
It’s not about what you eat, but what you got. You do not need to record details about this food. In the Food Log, check the day at the top of the Daily List and click “Day Complete”.”
NEW VERSION:
If I only ate food from my refrigerator or cupboard on a particular day, do I report the details of what I ate on the Food Log?
No. In the Food Log, we want to know only what you purchased or obtained for free during the study week. If some or all of the food you eat today came from a purchase or acquisition made earlier in the study week, it should have been included in the Food Log when it was obtained. Listing it again when it is eaten would cause double counting. Just check the day at the top of the Daily List and click “Day Complete”.
Disclaimer (A3)
NASS highlighted the phrase “. . . whether or not meals and snacks were taken. . .” and asked what it meant.
We agree that the phrase was not clear and have modified it. It now reads “. . . which meals (e.g., breakfast) and snacks were eaten during this period.”
Consent Form (A4)
The consent form indicates that participants age 11 and over can participate in training. NASS indicated: “The cut off age mentioned in Supporting Statement Part A is 16.”
The consent form was not edited as a result of this comment. As indicated earlier, we edited the Supporting Statement A.
Household Screener (B1)
NASS had 3 comments: (a) There were three instances where the Screener referred to the study as National Food Study and dropped the word “Pilot.” (b) Q2 was skipping to “R1.” NASS pointed out there was no R1. (c) Questions 4,5,13 and 14 were missing.
(a) The revised version now adds the word “Pilot” to the study name. (b) We replaced the word “R1” with “INTERVIEWER: ATTEMPT TO CONVERT RESPONDENT. IF NOT SUCCESSFUL, TERMINATE INTERVIEW AND CODE CASE A REFUSAL IN IMS.” (c) These questions, from the 2012 FoodAPS study, are not being asked during this Pilot. We did not renumber the instrument because it is interviewer-administered and the respondent will not see the numbers. Keeping the original numbering also allows easier references and comparisons between the FoodAPS and Pilot screeners when communicating findings with ERS.
Initial Interview (B2)
NASS commented on 3 questions: (a) B3. NASS pointed out that the question was missing answer categories and should be skipped if the answer to an earlier question was “yes.” (b) NASS asked if there is a difference in B5’s answer categories ‘high school and junior high’. (c) NASS pointed out an error in the skip pattern in question C1a1 noting that respondents whose store is on the list should not skip C1a1.
(a) We agree with NASS. We added response categories to question B3 and skips to questions B1 and B2. (b) There is either no difference or little difference in these categories, depending on school district, so both have been included. “Junior high” usually refers to grades 7-9, while “middle school” usually refers to grades 6-8. (c) We agree with NASS. We modified the skips in C1 and moved C1a1 before C2.
Food Log (B3)
NASS requested that the screen shots be larger and recommended one per page.
We have implemented the request.
Income Worksheet (B4)
The Study requires household members 16 years and older to complete this worksheet. NASS suggested: “Children as young as 14 are allowed to work in some states with parental permission. Do you want to drop the age to 14 or older?”
We agree with the comment, and ERS may want to ask all household members who receive an income to fill out the worksheet for the next FoodAPS. However, one of the objectives of this Pilot is to compare a revised set of procedures for collecting income with the procedures used before, and changing the criterion for who gets asked about income could threaten interpretation of the cause of any differences in the income data that are collected using the two approaches. ERS has asked that we therefore not change the age requirements for this Pilot.
NASS requested that the screen shots be larger and recommended one per page.
We have implemented the request.
Final Interview (B7)
Question B11 referred to a nutrition panel showcard. NASS requested that we include it in the docket.
We will not be using a hard copy show card during the administration of B11. Instead, the image will appear on the interviewer’s computer screen. The screen will be turned over to the respondent. We have modified question B11 and added the nutrition panel image to the question.
In question F12_4, NASS recommending changing the answer category “lottery winnings” to “lottery or other gambling winnings.”
We have implemented this request.
Several questions asked for income “before taxes” for different income sources. NASS pointed out that most of the income sources are not taxable income. Income sources identified were: worker’s compensation (F15_0B), TANF (F16_0B), child support (F18_0B), and SSI and SSDI (F20_0B).
We modified these questions and a few others removing the words “before taxes.”
NASS caught a skip error in question F27.
We have modified F27, changing the skip from F9a to F27a.
Question G3a asked about payment for real estate or property tax. NASS asked: “In VA (and possibly other states), residents pay property tax on their vehicles. Do you want respondents to include that amount here?”
In response to this comment we modified the question to read: How much (do you/does your household) pay for real estate or property taxes on your home?
Several questions in Section G asked about the household’s “monthly” expenses for different expense items. However, the following question then asked “How often?” and provided several duration options in addition to monthly such as weekly, twice per month, etc. NASS pointed out that this follow-up question did not make sense as the root question had already specified the unit “monthly.”
As a result of this comment, for all questions in Section G that mentioned a unit in the root of the question, we modified the followup question to say: “Let me confirm. Is that monthly, twice per month, every other week, weekly, or annually?” We decided to allow additional durations as our review of the 2012 FoodAPS data showed that some households did provide expenses in units other than monthly.
In Section H, a few questions asked “whether your household does ______ never, rarely . . .” NASS asked: “How should respondents answer if different HH members handle things differently.”
We agree that the respondent cannot always speak for the entire household. We therefore modified the stem of this series of questions as follows: “Next I'll read a list of financial practices. Please tell me whether your household does them never, rarely, sometimes, usually, or always. [IF THE RESPONDENT IS NOT ABLE TO RESPOND FOR THE HOUSEHOLD, REQUEST THAT THEY ANSWER FOR THEMSELVES.]”
Respondent Feedback (B8)
NASS asked how the respondent would be able to answer question Q1 “On a scale from 1 to 5, where 1 is “very easy” and a 5 is “very difficult”, please tell us how easy or difficult was it for you to get other household members to take part in the study?” if there are different answers for different household members.
We have modified the question. It now says: “On a scale from 1 to 5, where 1 is “very easy” and a 5 is “very difficult”, overall how easy or difficult was it for you to get other household members to take part in the study?”
Recall Interview (B9)
NASS caught a skip error in the “Food at Home Recall” Section. The introduction at the start of the module incorrectly sent the interviewer to question Q6 instead of Q3 for respondents who reported food.
The skip has been modified.
General Questions
NASS asked the following general questions.
For SNAP households, does it matter if not all HH members participate in SNAP? Is 1 SNAP participant sufficient for it to be considered a SNAP HH, or is there a stricter threshold? Would you only interview SNAP recipients in these HHs?
There is no requirement that all SNAP household members within a family participate in SNAP. We anticipate that there will be households where multiple families live at the same address and some are not SNAP participants.
The screener instrument considers the household eligible to participate in the NFS Pilot if the screener respondent or any member in the household participates in SNAP.
We would never limit our data collection to just the SNAP recipients in a household. We want all household members at the sampled address who are enumerated during the initial interview to complete the food log and meals and snack components, and those 16 and above to complete the income worksheet. Respondents 11 and above can participate for themselves; those under 11 participate via the primary respondent who serves as their proxy.
There is no indication on the Income Worksheet that a respondent’s answers will be shared with the primary respondent completing the Final Interview which is in contradiction to the privacy and confidentiality statements made elsewhere in the survey materials. How do you intend to deal with respondents who do not want this information shared with the primary respondent? Will you allow them to indicate their income information is not to be shared with the primary respondent on the Income Worksheet? Will you adjust the final interview instructions, so the interviewer does not ask the primary respondent to verify these respondents’ income information?
There are two options to address the problem you have raised. The first, as you have suggested, is to prevent the primary respondent from seeing the income amounts provided by other members. The second is to allow the primary respondent to see all income amounts after alerting all members of this ability and gaining their consent.
The first option is a technological fix, and ERS is concerned that there is insufficient time to program and test the necessary changes. We have therefore modified the Assent/Consent form by adding the sentence: “The primary adult will be able to view the information you provide.” Every household member will be asked to read this form and agree to the conditions before they are able to access the Food Log, Meals and Snacks Form, or Income Worksheet.
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Westat |
| File Modified | 0000-00-00 |
| File Created | 2021-01-23 |