Other Methodological Considerations

Certain other conventions should be noted in interpreting the findings of the study and how they are presented. Below we discuss the distinction between clients and respondents, and we describe the structure of reports available from the project.

Clients Versus Respondents. Clients are defined differently by program type. The kitchen and shelter programs are viewed as serving only those who are present at the program site. (Thus in general for these providers, the survey respondents are representative of all clients.) 1 However, pantry programs are regarded as serving all members of respondents' households.

At the kitchen and shelter sites, the sampling unit was the individual. That is, the interviewers were instructed to treat members of a single household as separate respondents if they were selected by our random sampling process and met other eligibility criteria (such as being at least 18 years of age). At the pantry programs, on the other hand, the sampling unit was the household, and only one interview was completed for each randomly selected household, even when two or more members of the household were present at the program.

Ideally, the survey would have obtained all relevant information about every member of the household, especially among pantry users. However, to minimize the burden on the respondents, the survey was designed to acquire information about at most six members of the household, including the respondent, on a limited set of variables of interest, such as sex, age, relationship to the respondent, citizenship, and employment status. We have only limited information on additional household members among the 4.6% of households with more than six members. Because households with more than six members are uncommon, we do not believe that this has significantly affected our estimates.

National Versus Local Reports. Hunger in America 2001 has produced a set of reports to serve both national- and local-level interests and to be useful to a wide range of audiences with varying needs. This national report consists of information gathered through 104 participating food banks. In addition, in most cases, a local report was generated containing information on clients and agencies served by a particular food bank. There are approximately 100 food-bank-level local reports. In addition, state-level reports were produced when all A2H-affiliated food banks in a particular state participated in this study. Approximately 20 states achieved full participation of their food banks.

In addition to the comprehensive national and local reports, A2H has disseminated 'Hunger in America 2001: Extended Executive Summary.' This executive summary contains key findings from the comprehensive national report. A technical appendix, which describes the methodologies of the current study in detail, is available under separate cover for distribution for audiences with technical interests.

Tables in the local and national reports are numbered comparably. This will facilitate comparisons between the local and national findings. Not all tables from the national report are reproduced in the local documents.

--------------------------------------------------------------------------------

1. One exception was children at the kitchens and shelters. They were clients, but they were not respondents, because only clients age 18 or older were interviewed for this study. However, the children were taken into account in estimating total clients.