• Component Description
• Quality Assurance & Quality Control
• Data Processing and Editing
• Analytic Notes
• References
• Codebook
  • SEQN - Respondent sequence number
  • FSD032A - HH Worried run out of food
  • FSD032B - HH Food didn't last
  • FSD032C - HH Couldn't afford balanced meals
  • FSQBOX1 - CHECK ITEM
  • FSD032D - HH Relied on low-cost food for child
  • FSD032E - HH Couldn't feed child balanced meal
  • FSQBOX2 - CHECK ITEM
  • FSQBOX3 - CHECK ITEM
  • FSD032F - HH Child not eating enough
  • FSD041 - HH Adults cut size or skip meals
  • FSD052 - HH How often adults cut size/skip meals
  • FSD061 - HH Eat less than should
  • FSD071 - HH Hungry, but didn't eat
  • FSD081 - HH Lost weight, no money for food
  • FSDBOX4 - CHECK ITEM
  • FSD092 - HH Adults not eat whole day
  • FSD102 - HH How often adults not eat for day
  • FSQBOX5 - CHECK ITEM
  • FSD111 - HH Cut size of child meals
  • FSD122 - HH Child skip meals
  • FSD132 - HH How often child skip meals
  • FSD141 - HH Child hungry in last 12 months
  • FSD146 - HH Child not eat whole day
  • FSDHH - Household food security category
  • FSDAD - Adult food security category
  • FSDCH - Child food security category
  • FSD151 - HH Emergency food received
  • FSDBOX6 - CHECK ITEM
  • FSD401 - Meal size cut
  • FSD411 - Skipped meals
  • FSD421 - Ate less than should
  • FSD431 - Hungry
  • FSQ440 - Lost weight
  • FSD451 - Not eat whole day
  • FSQ170 - Fd Stmp HH - authorized last 12 mos
  • FSD170N - Fd Stmp HH - No. people authorized
  • FSD180 - Fd Stmp SP authorized in last 12 mos
  • FSDBOX7 - CHECK ITEM
  • FSD190 - Fd Stmp SP mos authorized in last 12 mos
  • FSD200 - Fd Stmp SP currentlyauthorized
  • FSQ210 - Fd Stmp HH amount last month
  • FSQ220 - Fd Stmp HH amount last time
  • FSQ162 - WIC HH got past 12 mos
  • FSD650 - WIC SP got past 12 mos
  • FSD660 - WIC SP currently get
  • FSD670 - WIC SP No. mos received
• Appendix: Components of the Food Security Section (FSQ) 2005-2006

National Health and Nutrition Examination Survey

2005-2006 Data Documentation, Codebook, and Frequencies

Food Security (FSQ_D)

Data File: FSQ_D.xpt

First Published: January 2008

Last Revised: NA

Component Description

The existence of food security and hunger in the U.S. has been noted for several decades. In the 1980-90’s, qualitative and survey research were undertaken to characterize food security, and survey modules were developed to assess it ^1-4.

Since 1999, NHANES has assessed household food security with the US Food Security Survey Module (US FSSM)5. Individual-level food security items for adults and children were added to NHANES in 2000, and added for adolescents in 2005. The NHANES Food Security section (FSQ) also provides data on food assistance.

The four sub-components of the Food Security component are:

1. Household food security
2. Individual food security
3. Household and individual Food Stamp program benefits and
4. Household and individual WIC benefits.

A chart summarizing the various FSQ components and individual items is appended to this documentation.

I. Household food security

In the household interview, an adult responded to the US FSSM questions. There are 18 items for households with children under age 18 and 10 items for households without children. Questions refer to all household members, not just NHANES participants. All households were administered these questions in 2005-2006; no income screen was used. Three categorical household-level variables were created: household, adult, and child food security, as described by USDA5. Responses to the 10 or 18 US FSSM items and the categorical data derived from these responses are released on each household participant’s record.

II. Individual Food Security

Individual-level food security items were administered to all survey participants in the households that affirmed any FSSM item during the household interview. Adult (>15 years) and child (< 12 years) individual-level food security questions were administered after the 24 hour dietary recall. Generally, adults responded for themselves. An adult or older sibling responded for a child, although the child may have assisted. Adolescents (12-15 years) responded for themselves during the Audio Computer-Assisted Interview (ACASI).

Individual-level questions are derived from a subset of the household-referenced questions. Release of adolescent individual-level food security data is new for 2005-2006. The same variable labels are used for all age groups. These data are released only on the individual participant’s record.

III. Food program participation

1. Food stamp authorization: Household and individual-level food stamp data were collected for all household members during the household interview. One adult usually responded for the entire household. NCHS created a variable for the number of people in the HH who received food stamp benefits. New in 2005-06, the respondent was asked the dollar amount of the most recent food stamp benefit. The household level variables for “benefit amount” and “number of all people in the household who were authorized for food stamps” are released on all household participants’ records.

2. WIC: Household and Individual-level data on WIC benefits were collected during the household interview and in the reproductive health interview administered in the MEC. Data are released at the household and individual-level.

Quality Assurance & Quality Control

Incoming FSQ data were reviewed by the project officer. NHANES staff reviewed taped interviews, traveled to field sites to observe interviews, read interviewer comments, and attended training and re-training sessions to maintain the high quality of NHANES interviews. Data were reviewed for outliers, out of range responses, missing responses, and non-response.

Data Processing and Editing

Because responses other than “Never” were very rare for FSQ451 (Not eat for whole day), responses of “Often and Sometimes” were collapsed into one category, “Ever.”

WIC data for women that appeared unlikely (e.g. a mother receiving WIC for longer than WIC regulations stipulate) were not edited, as there are plausible explanations for longer WIC intake.

Dataset and Variable Label Structure

Variable labels were constructed to help distinguish between similar variables, as described below.

FSQ_FSSM: Household-level data:

18 questions were asked to households with children under ages of 18. 10 questions were asked to households without children. Three categorical variables, each with four response levels, were created. Categorical variables are released on all household members’ records.

1. Household food security (FSDHH): uses all 18 FSSM variables.
2. Adult food security category (FSDAD): uses the10 adult/household variables.
3. Child food security category (FSDCH): uses the 8 child-level items.

FSQ_FSSM: Individual-level data

Participants 12 years and older were asked 7 questions about their own food security. Participants under 12 years old were asked 6 questions; the question about having lost weight was not asked.

Food Stamp Program data

There are 2 household-level and 3 individual-level questions were asked. HH level data are released on all HH members’ records; individual level data are released only on the individual participant’s record.

WIC participation

One household-level and 3 individual-level questions for pregnant or breast-feeding women and for young children were administered.

Tabular display of all FSQ variables

There are many similar FSQ variables, often differing only by location of administration, referent time frame, and/or reference person. A multi-colored table at the end of this document displays the entirety of the FSQ variables.

Analytic Notes

DIFFERENCES BETWEEN THE ‘05-06 AND ‘99-04 FSQ DATA COLLECTION

1. All households were asked FSQ questions; no income screen was used.
2. Adolescents (12-15 years) were asked individual level food security questions as part of their Audio Computer Assisted Sample person Interview (ACASI).
3. Households that reported getting food stamps were asked the amount of their food stamp benefits last month, or the last month that they month they received them.

DIFFERENCES BETWEEN THE ‘05-06 AND ’99-04 FSQ DATA RELEASE

Most NHANES survey questions that assess a specific aspect of food security are identical or very similar for adults, teens, and children. In this 2005-2006 FSQ release, data collected separately for different age groups are combined and released as one variable for each topic, although the actual survey instruments may be slightly different by participants’ ages. The exact wording of the FSQ questionnaires used in the survey can be found through the “Questionnaire” link accompanying this release.

FOOD STAMP DATA

The prevalence of individuals reporting authorization to receive food stamps estimated from the NHANES data is lower than the prevalence indicated by USDA administrative data. Misinterpretation of the word “authorized” in the food stamp questions may have contributed to this discrepancy. Thus, variable FSD170N (number of people in the household authorized to receive food stamps) may be more reliable than variable FSD180 (individual was authorized to receive food stamps in the last 12 months) to identify NHANES participants who live in households that were authorized to receive food stamps in the last 12 months.

USEFUL DOCUMENTS FOR ANALYZING NHANES FOOD SECURITY DATA

The most recent NHANES Analytic Guidelines (September 2006) are available at the NHANES website⁶.

More information on collecting and analyzing US FSSM data is available at USDA’s website⁵.

Food security data are also collected in the Current Population Survey and are reported annually by the USDA’s Economic Research Service.

THE NATIONAL RESEARCH COUNCIL ASSESSMENT OF THE USFSSM

A recent assessment of US FSSM was conducted by the National Research Council⁷. The committee determined that “hunger” should be measured only at the individual level and that it is a physical condition that is distinct from food insecurity. It can result from food insecurity and should not be used as a food security category label.

THE USE OF SAMPLE WEIGHTS

FSQ data were collected in the household and in the MEC. For data items collected in the household home interview, the interview sample weights may be used in their analysis. However, if a data analysis requires merging of data collected in the household with data collected in the MEC, examination sample weights should be used for analyses. Please refer to the Analytic Guidelines for further details on the use of sample weights and other analytic issues.

References

• Development of indicators to assess hunger. Radimer KL, Olson CM, Campbell CC. J Nutr. 1990; 120: 15448.
• Food Insecurity and Hunger in the United States: An assessment of the measure. Wunderlich G, Norwood J, Editors. NRC, Committee on National Statistics. Washington DC: The National Academies Press. 2006. http://books.nap.edu/openbook.php?record_id=11578&page=5
• Guide to Measuring Household Food Security Revised 2000. Bickel G, Nord M, Price C, Hamilton W, Cook J. U.S. Department of Agriculture, Food and Nutrition Service. March, 2000. ttp://www.fns.usda.gov/fsec/FILES/FSGuide.pdf
• The Community Childhood Hunger Identification Project: a model of domestic hunger – demonstration project in Seattle, Washington. Wheler CA, Scott RI, Anderson JJ. J Nutr Educ 1992; 24(1): 29S-35S.
• The National Health and Nutrition Examination Survey (NHANES) Analytic and Reporting Guidelines. September 2006. http://www.cdc.gov/nchs/nhanes/survey_methods.htm
• Understanding hunger and developing indicators to assess it in women and children. Radimer KL, Olson CM, Green JC, Campbell CC, Habicht J-P. J Nutr Educ. 1992; 24: 36s-45s.
• Validation of the Radimer/Cornell measures of hunger and food insecurity. Kendall A, Olson CM, Frongillo EA Jr. J Nutr. 1995; 125: 2793-801.

Codebook and Frequencies