Martin David

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.

Alternative Methods for CPS Income Imputation

Abstract The U.S. Bureau of the Census imputes missing income items in the income supplement of the Current Population Survey (CPS) by a technique commonly known as the CPS hot deck. This article compares CPS hot deck imputations of wages and salary amounts with alternatives based on regression models for the logarithm of wages and salary and for the wage rate. Comparisons are effected by comparing imputations with an Internal Revenue Service (IRS) wages and salary amount found by an exact match of CPS data to IRS records. Although limitations in the matching and in the comparison variable preclude a definitive conclusion, we find that (a) the CPS hot deck does not underestimate income aggregates to any serious extent; (b) model-based alternatives have slightly smaller mean absolute error than the hot deck, when comparable data bases of respondents are used to carry out imputations; and (c) multivariate models for imputing recipiency, weeks and hours worked, and earnings need to be developed to provide re...

Estimation with Response Error and Nonresponse: Food-Stamp Participation in the SIPP

Error in survey data originates from failure to contact the sample and from false answers to verifiable questions. These errors may be systematic and associated with uncooperative or unreliable respondents. Zabel modeled attrition in the Survey of Income and Program Participation and found systematic demographic and design effects. Bollinger and David modeled response error and identified correlations to income per capita. In this analysis, we link missing interviews in a panel and response error through a trivariate probit analysis. Robustness of the correlation between attrition and response error is examined by comparing variants of the model. The joint model of response error and attrition becomes the first stage of a pseudolikelihood estimate of a model of food-stamp participation. The model is significantly different from naive probit on the survey data.

Transferable discharge permits and economic efficiency: The fox river

Recent emphasis on reforms of environmental regulation has led to suggestions for strategies which maintain environmental standards but allow the needed flexibility and cost effectiveness. The transferable discharge permit (TDP) is one such strategy for water pollution control recently adopted in Wisconsin. In this article, the potential for substantial cost savings from trading TDPs is demonstrated using data on the Fox River in Wisconsin. A simulation model of water quality (Qual-III) and a linear programming model of abatement costs determine the optimum pattern of discharge. Reaching that optimum from proposed pollution abatement orders is shown to be feasible. Varying conditions of flow and temperature can be accommodated using trade coefficients which can be accurately estimated through interpolation. The calculations demonstrate the value and feasibility of flexible regulations governing water pollution abatement.

Education and Income

Introduction, 423. — The problem of method, 424. — Proposed procedures, 425. — Statistical method and source of data, 426. Explanatory factors in addition to education, 427. — Search for interaction effects, and more detail on education, 430. — Estimated lifetime earnings, 432. — Qualifications, 435. — Summary, 436.

The Validity of Income Reported by a Sample of Families Who Received Welfare Assistance during 1959

Abstract The author collaborated on a nationwide study of low-income families conducted by the Survey Research Center of the University of Michigan. One of the primary purposes of this study was to determine some of the factors which are responsible for the low-income position of the 16 per cent of the nations families who receive less than

Changes in Cohort Wealth Over a Generation

2,000 a year.1 Information on the incomes of low-income families with heads of working age was collected from a nationwide sample by personal interview techniques. Any bias in the reporting of income creates serious problems for such a study. Underreporting of income by low-income families would lead the analyst to believe that a higher proportion of families live on low incomes than is actually the case. The investigation reported here provides a check on one particular form of bias in the reporting of incomes—underreporting of income from public welfare assistance by families whose head is of working age.2 Bias in reporting of that income resulted in approximately ...

Welfare, Income, and Budget Needs

Empirical computation of expected wealth is hampered by two problems: mortality risks vary in the population and over time; and observation of net estates for most cohorts is truncated, as some individuals in a cohort survive the calendar date on which observation is terminated. These two problems are solved in estimating cohort wealth for a sample of Wisconsin taxpayers. Hazard rate models of differential occupational mortality risks were estimated from the occupational information on the tax records. Values of net estate are simulated for individuals in each birth cohort who survived. Survivors have characteristics that imply greater wealth holdings than the deceased in every birth year covered by the study (1890–1924). Because of this, estimates of wealth-age relationships produced by the estate multiplier method for any given year will have a serious downward bias. Longitudinal data imply that dissaving does not occur after age 65.

The Wisconsin Assets and Incomes Studies Archive

R EAL disposable income is the measure generally accepted by economists as an indicator of the welfare of a household. Were we living in a world in which each individual formed a household and in which everyone required exactly the same goods to maintain himself, real disposable income might be an adequate measure. However, households are generally complex, and the income of a household must support a number of individuals whose daily requirements range from the nominal demands of a young child to the ample needs of an adolescent. Thus the welfare of a household must be measured not only by its income but also by the size of the household and the variety of needs of its members. This paper discusses a measure of welfare which relates the resources of a household to its needs. Estimates of welfare levels in the United States population based on this measure show that disposable income and per capita disposable income are extremely crude measures of welfare. Both overstate or understate the standard of living of an appreciable fraction of the United States population. The paper concludes with a few notes on the need for more accurate measures of welfare.

Systems for metadata: documenting scientific databases

It is a commonplace that the inability to conduct controlled experiments (amongst other things) makes the social sciences more difficult fields of scientific inquiry than the physical and even the medico-biological fields. It has only relatively recently become recognized that this can be overcome, using modern statistical analysis, provided that a sufficient number of significant variables is available for analysis. As yet it is almost exclusively the professional social scientist (who wishes to analyze such data) who recognizes the implications of these statements, while those in a position to provide these data are largely ignorant or unconcerned about these implications. If it can be said that the survey is the social scientist’s test tube, then surely the vast bodies of data recording human behavior that exist in the files and administrative records of governmental and business organi-