Reaching the Working Poor and Poor Elderly Study: What We Learned and211 Recommendations for Future Research

摘要

The Food Stamp Program (FSP) is the cornerstone of America's overall strategy for211u001eensuring that all Americans have enough to eat. It plays an important role as a 211u001esafety net for low-income elderly persons and working persons with low earnings, 211u001eincluding those who have successfully made the transition from cash welfare to 211u001ework. But it is estimated that in 1994 less than one-half of working households 211u001eand less than two-fifths of elderly households that were thought to be eligible 211u001efor food stamp benefits actually received them. Low participation rates are a 211u001econcern because, depending on their cause, they may be an indication that the FSP 211u001eis not fulfilling its mission of providing food assistance to everyone who needs 211u001eit. To increase its understanding of the reasons for nonparticipation, the Food 211u001eand Nutrition Service (FNS) of the U.S. Department of Agriculture (USDA) 211u001econtracted with Mathematica Policy Research, Inc. (MPR) to conduct a study of 211u001enonparticipation by low-income working and elderly households, entitled Reaching 211u001ethe Working Poor and Poor Elderly. This report summarizes what was learned and 211u001eoffers recommendations for how a national survey of the reasons for 211u001enonparticipation in the FSP should be designed and fielded. The Working and 211u001eElderly Poor Study had three phases. First, we assessed what was known about FSP 211u001enonparticipation from existing surveys and focus groups and analyzed data on the 211u001echaracteristics and food security of FSP-eligible households. Second, we 211u001econducted 12 focus groups with FSP participants and nonparticipants from working 211u001eand elderly households that were probably FSP eligible. Third, we developed 211u001equestionnaires about the reasons for nonparticipation and tested them using 211u001ecognitive techniques on 16 low-income elderly and working persons. We then 211u001econducted a large pretest of the survey in the winter of 1998, administering the 211u001equestionnaires to 451 respondents, mostly by telephone.