Food Stamp Program Participation Research Articles

Approximated Perfect Values in Logistic Regression for Prediction and Outlier Detection

A method to compute perfect values is developed for use with a weighted likelihood approach. It allows to obtain a direct one-iteration approximation for the parameter estimates. We study approximated perfect values applied to logistic regression and demonstrate their usefulness for predictive purposes when the response is missing. We also show that statistics for single-case outlier detection can be deduced. An empirical analysis to determine participation in a federal food-stamp program is presented to illustrate outlier detection and alternatives for estimating the probability of participation in the program. We compare results obtained with the new approach and those obtained maximizing the penalized log-likelihood function.

Elderly food stamp participants are different from eligible nonparticipants by level of nutrition risk but not nutrient intake

Nutrition risk, nutrient intake, and selected socioeconomic characteristics of elderly food stamp recipients were compared with elderly people who were eligible for food stamps but did not participate in the program. Data were collected by survey using a convenience sample of 200 low-income elderly people living in subsidized housing. Although there were no significant differences in food group or nutrient intake between participants and nonparticipants, overall, these low-income subjects did not meet minimum Dietary Reference Intake (DRI) standards for many nutrients and had caloric and food group serving intakes that were below recommendations. Food Stamp Program participants were at significantly higher nutrition risk than their nonparticipating peers as detected by the Nutrition Screening Initiative Checklist (P≤.01) and other indicators of disability. It seems that those already at highest risk were in fact participating in the Food Stamp Program because those eligible but not participating had significantly higher incomes and were at significantly lower nutritional risk. J Am Diet Assoc. 2003; 103:103-107.

Focus Groups Indicate that Vegetable and Fruit Consumption by Food Stamp-Eligible Hispanics is Affected by Children and Unfamiliarity with Non-Traditional Foods

As the largest US food assistance program, the Food Stamp Program has a primary goal of improving the diets of low-income households, historically, though, nutrition education has not been a high-priority area ( (1) Basiotis P.P Lramer-LeBlanc C.S Kennedy E.T Maintaining nutrition security and diet quality the role of the Food Stamp Program and WIC. Fam Econ Nutr Rev. 1998; 11: 4-16 Google Scholar ). To compensate, the US Department of Agriculture began funding statewide projects in 1995 and 1996 to provide nutrition education to food stamp recipients ( (2) Research Triangle Institute and Health Systems Research. Evaluation of Statewide Nutrition Education Networks. Washington, DC: Health Systems Research, 1999. Google Scholar ). The Arizona Nutrition Network, organized in 1995, decided to promote vegetable and fruit consumption as its first nutrition education project for the state’s food stamp-eligible population. Network partners prioritized Hispanic households as the initial target audience for nutrition intervention because Hispanic individuals comprised more than half of the state’s Food Stamp Program participants. The Food Stamp Nutrition Education Program, Food and Nutrition Service, US Department of Agriculture funded this project through the Arizona Department of Economic Security for the Arizona Nutrition Network. The authors thank all of the Arizona Nutrition Network partners, who have contributed countless hours to nutrition education over the past several years.

Explaining Recent Declines in Food Stamp Program Participation

Participation in the Food Stamp Program, a key component of the social safety net, fell precipitously between 1993 and 1998. The strong economy and welfare reform have both been advanced as possible explanations for this decline. We investigate the decline using both administrative and household-level data. We find that among households with incomes less than 300% of poverty, 20% of the decline in FSP utilization can be attributed to lower unemployment, while 30% is due to the implementation of the new Temporary Assistance for Needy Families children. The latter effect is concentrated among single heads and in central cities. We also show that single heads are more sensitive than other households to increases in the transactions costs associated with applying for food stamps. Together, these results suggest that efforts to restore FSP benefits to needy households should focus on reducing application costs for single heads in central cities. The Food Stamp Program (FSP) is one of the largest transfer programs in the United States, having provided nearly $24 billion in benefits to 27.5 million families in 1994 (Committee on Ways and Means, 1998). Participation in the FSP hovered around 20 million persons per year during the 1980s, but rose sharply in the early 1990s to a peak of approximately 27 million persons in 1994. Participation then began to fall dramatically, declining back to 20 million participants by 1998. About three-fourths of this decline occurred between 1996 and 1998 (Castner and Cody 1999). Previous research has shown that most of the fall in the caseload has stemmed from falling numbers of single-parent families that receive food stamps (U.S. Department of Agriculture, 1999). The caseload has also fallen dramatically for two other groups: Able-bodied adults without dependents (ABAWD's) and most legal immigrants. However, since they constitute a relatively small share of the FSP caseload, they account for only 22 percent of the overall caseload decline. Finally, McConnell and Ohls (2000) show that the decline in the caseload has been much more dramatic in urban than in rural areas. Declines in the food stamp caseload may be either good or bad news, depending on the reasons for the decline. A number of hypotheses have been offered. The first is that welfare reform has caused the decline. The Personal Responsibility and Work Opportunity Restoration Act of 1996 (PRWORA) made it more difficult for single mothers to receive cash welfare, and may have had the largely unintended consequence of making it more difficult for them to access food stamps. PRWORA also included provisions limiting the use of food stamps by ABAWDS and legal immigrants. Hence, PRWORA seems to have impacted precisely those groups which have shown the greatest decline in the food stamp caseload. A second, closely related hypothesis is that the climate surrounding welfare reform has 2 increased the stigma and also the transactions costs associated with participating in food stamps. Little direct evidence is available about the importance of stigma and transactions costs, so one contribution of our study is to analyze direct proxies for these concepts. A third hypothesis is that the decline in the caseload is due to the recent economic boom. If families are leaving the roles because they are better off, then this is presumably a good thing. On the other hand, if needy families are leaving because it has become more difficult for them to access food stamps, then most observers would feel that this was a negative consequence of welfare reform, and one that merited some remedial action. Our paper provides some new evidence on these competing hypotheses, using both aggregate administrative and householdlevel data. We find that among households with incomes less than 300% of poverty, 20% of the decline in FSP utilization can be attributed to lower unemployment, while 30% is due to the implementation of the new Temporary Assistance for Needy Families children. The later effect is concentrated among single heads and in central cities. We also show that single heads are more sensitive than other households to increases in the transactions costs associated with applying for food stamps. Together, these results suggest that efforts to restore FSP benefits to needy households should focus on reducing application costs for single heads in central cities. The rest of the paper is laid out as follows: We begin with some background information about the recent decline in food stamp caseloads and the possible explanations. We then present the empirical model and the data in sections II and III. In section IV, we present the main results, while section V discusses several extensions. Finally, conclusions appear in section VI.

Temporary Emergency Food Assistance Program: Perceptions of Benefits and Effect of Welfare Reform

The study evaluated the Temporary Emergency Food Assistance Program (TEFAP) in Larimer County, Colorado. Study participants of various ethnicities completed a survey containing qualitative and quantitative components. Of the 64 respondents, 58% thought they would use TEFAP a year or more; 75% found the quantity of food distributed to be helpful or very helpful. Food security status, Food Stamp program participation, and helpfulness of food quantity provided varied significantly with employment status. Families used the program less often than respondents without children. This research generally supports continuation of TEFAP yet raises policy questions regarding the long-term use of TEFAP.

Modeling Discrete Choice With Response Error: Food Stamp Participation

Abstract Validation of Food Stamp program participation by Marquis and Moore revealed net bias of 13% in mean estimates of participation in the 1984 Survey of Income and Program Participation. We extend the analysis, incorporating demographic and economic covariates in models for underreporting and overreporting. We use the resulting models of response error to refine estimates of participation models using pseudo-maximum likelihood estimation methods. We find that the probability of underreporting rises with increased family income. Estimates of the participation model indicate the presence of substantial bias in income- and asset-related parameters compared to values obtained without incorporating response error.

Modeling Discrete Choice with Response Error: Food Stamp Participation

Validation of Food Stamp program participation by Marquis and Moore revealed net bias of 13% in mean estimates of participation in the 1984 Survey of Income and Program Participation. We extend the analysis, incorporating demographic and economic covariates in models for underreporting and overreporting. We use the resulting models of response error to refine estimates of participation models using pseudo-maximum likelihood estimation methods. We find that the probability of underreporting rises with increased family income. Estimates of the participation model indicate the presence of substantial bias in income- and asset-related parameters compared to values obtained without incorporating response error.

The Impact of Nutritional Knowledge on the Quantity and the Quality of Diet Among the Southern Rural Elderly.

LEARNING OUTCOME: To examine the relationship between diet quality and nutrition knowledge of the rural elderly.Volunteers (n=1555) of elderly residing in counties with 70% or more rural population from 9 Southern states (AL, AK, GA, KY, MD, MO, MS, TN, TX) participated in the study. This study was designed to delineate whether there was a relationship between nutrition knowledge and the quality of diet and nutrient consumption among the rural elderly, one of most neglected segments of our population. The sample consisted of 47.8% Blacks and 51.9% Whites, 15.8% male and 84.2% female. The age ranged from 60 to 102 yrs. (mean ± SD: 72.8 yrs. ± 7.6) and 57.6% lived alone; only 34.9% of the elderly completed high school or higher education. 43% of the elderly had a per capita monthly income less than $450, and 27.5% were Food Stamp Program participants. Information collected from these participants through face-to-face interviews included: socioeconomic factors, health history, 2 non-consecutive days food intakes by 24-hr recall, and nutritional knowledge, practice, and attitudes. The nutrition knowledge and practice questions were adapted from the USDA Diet and Health Knowledge Survey. Nutrient intakes were computed from the 24-hr recall data by using Nutritionist IV (N-Squared, San Bruno, CA). The Diet Quality Index (DQI) was computed based on R. Patterson's method (JADA,. 94: 57-64, 1994). Knowledge scores on dietary fiber and fat/cholesterol were trisected - high, medium and low. One-way factorial ANOVA technique was used to discern the difference in DQI and nutrient consumption as related to their knowledge scores. The DQI increased significantly with the knowledge scores on dietary fiber ((p<0.001) and on fat/cholesterol (p<0.0001). Those with higher knowledge scores on fiber consumed significantly more fiber (p<0.001), and those with higher scores on fat/cholesterol consumed significantly lower amounts of cholesterol (p<0.0002) and fat (p<0.05), particularly saturated fat (p<0.003). This study clearly demonstrated the importance of nutrition knowledge to quality of diet and higher consumtion of fiber as well as lower fat consumtion. (Supported by USDA, CSREES, RR6)

Relationship of Socioeconomic, Demographic, and Behavioral Variables to Changes in Children's Fat Intakes Between 1989-91 and 1994

LEARNING OUTCOME: To better understand factors associated with the decline in fat intakes by children 2-17 years of age between 1989-91 and 1994.The U.S. Department of Agriculture's Continuing Survey of Food Intakes by Individuals (CSFII) 1989-91 and 1994 data indicate that fat intakes by children 2-17 years of age declined from 34 percent of calories in 1989-91 to 33 percent of calories in 1994. This study investigated whether this decline was associated with specific socioeconomic, demographic, or behavioral characteristics. The sample included 3,025 children 2-17 years of age who provided one 24-hour recall in either CSFII 1989-91 or CSFII 1994. One child was randomly selected from each survey household to avoid household clustering effects. Multiple regression analysis was used to assess the effects of selected independent variables on children's fat intakes. These variables included socioeconomic status (SES) as defined by household income and educational level of the household meal planner, region and urbanization level of residence, age, sex, and race of the child, food stamp program participation, school food consumption, meal planner employment status, season and day of week of reported intake and survey time period (1989-91 vs. 1994). Interaction terms assessing differences in the effects of SES between the two time periods were also included. The interaction of time with higher SES (household income above 185 percent of the Federal poverty level and meal planner with at least some college education) was significant (p<.05) and negative, indicating that between 1989-91 and 1994 higher SES was associated with a significant decline in children's fat intakes. Other factors significantly associated with lower fat intakes were being of a race other than black or white and living in the Northeast region of the United States as opposed to the South. School food consumption was significantly associated with higher fat intakes. Results suggest that fat intakes by lower SES children are slower to approach Dietary Guidelines recommendations than those of higher SES children. Nutrition interventions directed toward these children may be needed.

Variables Selection Using the Wald Test and a Robust C P

SUMMARY A new variables selection criterion is presented. It is based on the Wald test statistic and is defined by Tp = Wp- K + 2p where K and p are the numbers of parameters in the full and submodel respectively, and Wp is theWald statistic for testing whether the coefficients of the variables not in the submodel are 0.'Good' submodels will have Tp-values that are close to or smaller than p, and, as with Mallows's Cp, they will be selected by graphical rather than stepwise methods. We first consider an application to the linear regression of the heat evolved in a cement mix on four explanatory variables; we use robust methods and obtain the same results as those from the more computer-intensive methods of Ronchetti and Staudte. Our later applications are to previously published data sets which use logistic regression to predict participation in the US federal food stamp program, myocardial infarction and prostatic cancer. The first data set was shown in previous analysis to contain an outlier and is considered for illustration. In the last two data sets our criterion applied to the maximum likelihood estimates selects the same model as do previously published stepwise analyses. However, for the food stamp data set, the application of our criterion using the robust logistic regression estimates of Carroll and Pederson suggests more parsimonious models than those arising from the likelihood analysis, and further suggests that interactions previously regarded as important may be due to outliers.

How Much Can Informational Outreach Programs Increase Food Stamp Program Participation?

AbstractData from the Panel Survey of Income Dynamics (PSID) indicate that one‐fourth of the households eligible for the Food Stamp Program (FSP) do not know they are eligible. Using an FSP participation equation that incorporates a household's perceived FSP eligibility, it was estimated that informational outreach programs directed at these households have the potential of increasing the overall FSP participation rate by a maximum of 18% and benefit payments by 12.6%. These results can be useful in establishing realistic goals for FSP informational outreach programs.

Risk Factors for Delayed Immunization Against Measles, Mumps, and Rubella in Colorado Two-Year-Olds

1) To examine a wide range of potential risk factors for delayed childhood immunization against measles, mumps, and rubella and 2) to determine the parental response to proposed solutions regarding the problem of delayed childhood immunization. A case-control study in which subjects had (controls) or had not (cases) received measles, mumps, and rubella vaccine (MMR) by 2 years of age. Parents of subjects were also surveyed to find out whether they thought selected proposals to improve immunization rates would be successful. Maternal education of high school or less at the time of the child's birth, more than or equal to two moves between birth and age 2, maternal age < 21 years of age at the time of the child's birth, more than or equal to two older siblings, participation in The Special Supplemental Food Program for Women, Infants, and Children (WIC), participation in the federal food stamp program, and incorrect knowledge of the recommended age for MMR significantly elevated the odds ratio (OR) for delayed immunization. At least two older siblings (OR = 3.2), maternal age < 21 years of age at the time of the child's birth (OR = 2.8), and incorrect knowledge of the recommended age for MMR (OR = 2.7) remained significant risk factors in a multivariate logistic regression model. Insurance status and cost factors were not significant risk factors for delayed immunization. Parents though that reminders for immunization and a central record system would have made obtaining immunizations easier. Based on our findings of the importance of immunization knowledge and demographic characteristics as risk factors for delayed immunization, we suggest that a message (the recommended age for immunizations) and a target group for that message (families who move frequently, have older children, and are headed by younger parents) be evaluated as an intervention to improve immunization rates.

Ethnicity and Food Stamp program participation: Effect upon dietary intakes of low-income mothers served by a north Florida family practice center

Ethnic differences in self-reported dietary intake between blacks and whites (participants and nonparticipants in the Food Stamp program) were studied in a group of 102 low-income mothers presenting themselves for treatment at an urban family practice center. The study population included 51 whites, of whom 21 were participants in the Food Stamp program, and 51 blacks, of whom 29 were Food Stamp program participants. The mean age of white and black mothers was approximately 29 years. Mean numbers of children per family were 2.1 for whites and 2.82 for blacks. Food and nutrient intake data were collected using a 24-hour recall and a food frequency questionnaire. Mean intakes of calcium, iron, riboflavin, and thiamin were found to be less than the RDAs for both groups. Whites additionally had mean intakes of vitamin A and niacin that fell below the RDAs. Blacks had significantly higher mean intakes of niacin and protein and a significantly higher 24-hour mean intake of soft drinks than whites. Being a member of the black race was positively associated with consumption of chicken, hot dogs, collard/mustard greens, and white rice. Being a member of the white race was positively associated with consumption of broccoli, potatoes, and whole wheat bread. Among participants in the Food Stamp program, mean intakes of protein, niacin, and soft drinks were significantly higher for blacks than for whites. This study suggests that ethnic differences in food preferences do exist but that those preferences do not translate into major differences in daily reported nutrient intakes. J Am Diet Assoc 88:1081, 1988.

Food stamps, food costs, nuctrient availbility, and nutrient intake

The Food Stamp Program (FSP) is designed to increase the food expenditures and the diet status of low income households. However, suprisingly little is known about the relationships among income, FSP bonus, FSP participation, availability of nutrients in household food supplies and individual nutrients intakes. Data from the special Low Income Component of the 1977–1978 Nationwide Food Consumption Survey were utilized to estimate effects of FSP participation on food expenditures, nutrient availability, and nutrient intake. The estimation procedure utilized a principal components index for characterizing availability and intake of the included nutrients. Results showed that FSP participation and the bonus increased food expenditures and improved the nutritional status of diets of low income household members. Nutrient availability and intake levels for FSP participants were increased by as much as 10–15% per week for selected nutrients.

Food stamps and the energy status of the U.S. elderly poor

The energy status of poor elderly individuals in relation to the nonpoor elderly was examined using data from NHANES I and II. Within the poor population, the status of Food Stamp Program participants was compared with that of nonparticipants. The indicators of energy status used were weight (adjusted for height), triceps skinfold, subscapular skinfold, and energy intake. Energy status was examined by analyses of variance; the effects associated with sex, race, age, geographic region, and other potentially confounding variables were included in the model. In both surveys, there were interactions of income and food stamp use with sex. That is, poor women had greater means for weight and triceps skinfold than the nonpoor. Among the poor women, the Food Stamp Program participants weighed more than the nonparticipants, and their triceps skinfold mean was greater in NHANES II than in NHANES I. The poor men had a lower mean for triceps skinfold than the nonpoor, but there was no difference between the program participants and nonparticipants. Hence, for the poor elderly in NHANES I and II, food stamp use was not consistently associated with differences in energy status, as indicated by weight, skinfold, and energy intake.

Food stamps and the iron status of the U.S. elderly poor

With the use of data from NHANES I and II, the iron status of poor elderly persons was examined by comparing it with that of nonpoor elderly persons. Within the poor group, the iron status of Food Stamp Program participants was compared with that of nonparticipants. The poor were those below the poverty line; the nonpoor were those with incomes at least twice the poverty line. The indicators of iron status were: hemoglobin, hematocrit, free erythrocyte protoporphyrin, transferrin saturation, total iron-binding capacity, serum iron, and iron intake. Iron status was examined by analyses of variance. The effects associated with sex, race, age, geographic region, and other potentially confounding factors were included in the model. In both surveys, there were interactions with race. That is, the poor whites had lower transferrin saturation and iron intake means than the nonpoor whites, while the Food Stamp Program participants were no different from the nonparticipants in transferrin saturation. Within the nonwhites, the poor had a lower hemoglobin mean than the nonpoor, and the program participants had a lower hemoglobin mean than the nonparticipants. Therefore, the Food Stamp Program was not consistently associated with better iron nutrition of the elderly poor in NHANES I and II.

Cash Equivalence, Welfare Stigma, and Food Stamps

The Food Stamp Program (FSP) is one of the largest domestic redistributive transfer programs with approximately 22 million participants at a cost of $12 billion in 1983. The stated objective of the FSP is raise the levels of nutrition among low income by providing such households with an opportunity to obtain a low-cost nutritionally adequate diet [29,5]. For program evaluation and cost forecasting, accurate estimates of the determinants of eligible households' program participation and food expenditure decisions are of particular concern in light of the elimination of the purchase requirement (EPR) and other recent structural changes in the program. The problem of obtaining these estimates has three dimensions. First, observations on the behavior of eligible households under the post-EPR institutional setting are required. Second, the roles of welfare stigma and the cash equivalence of the in-kind food stamps must be clarified. Finally, because program participation is a household choice that affects food expenditures, the interdependence of eligible households' FSP participation and food expenditure decisions must be accommodated.' Emphasizing an important component of this interdependence, the connection between cash equivalence of stamps in the participation decision and in expenditures for food, this paper tackles all three dimensions. The purpose of this study is to determine the extent to which food stamps are cash equivalent and to determine which program and household characteristics, including the degree of cash equivalence and stigma, affect post-EPR Food Stamp Program participation

Effects of Food Stamp Program Participation and Other Sociodemographic Characteristics on Food Expenditure Patterns of Elderly Minority Households

An Engel-type regression model was used to examine the relationship between Food Stamp program participation, socioeconomic characteristics, and household food expenditures. Findings suggest that Food Stamp subsidies, money income levels, and housing subsidies have a strong relationship with household food expenditure patterns. Food Stamp subsidies were significantly more effective in enhancing food expenditures at the lower household income ranges. Food Stamp subsidies also interacted strongly with housing rental subsidies, feeding programs, and money income to improve the quantity and quality of food consumption. Findings suggest that multiple-program benefits are important policy instruments for providing an adequate economic safety net for the elderly poor.

Food Energy and Nutrients Per Dollar's Worth of Food From Available Home Food Supplies

Data from the Nationwide Food Consumption Survey, 1977–78, were used to assess effects of household size and composition, household income, and eli gibility/participation in the Food Stamp Program (FSP) on the food energy and nutrients per dollar's worth of food from the household food supply during one week. Mean food energy, protein, calcium, iron, magnesium, and vitamin A per food dollar were calculated for selected sample partitions. Household size, in come, FSP participation, and other socioeconomic variables were regressed on food energy and nutrients per food dollar. Results indicated that household size had a statistically significant and positive impact on nutrients per dollar's worth of food except for vitamin A. Household income had a statistically significant and negative impact on nutrients per food dollar. Households that participated in the FSP had greater food energy and nutrient return per dollar than households eligible but not participating in the FSP. However, these differences for FSP par ticipation were statistically significant only for calcium.

The Impact of Federal Transfer Programs on the Nutrient Intake of Elderly Individuals

In this paper we analyze the determinants of elderly nutrient intake. Our focus is on the impact of food assistance and income transfer programs, particularly the Food Stamp Program (FSP). Our statistical analysis is of a model with endogenous FSP participation and nutrient intake decisions. The insights gained regarding the effects on the behavior of the elderly caused by interactions among FSP, Supplemental Security Income, Social Security, and elderly feeding programs should prove useful in designing future transfer policies.