Client evaluations of the quality of welfare state programs.

Abstract

Client evaluations are examined as research studies generating, directly from clients, data about program quality: effectiveness, efficiency, effort, and unintended consequences. Data from the University of Michigan, 1983 Detroit Area Study are used to investigate citizens' perceptions and clients' evaluations of the quality of welfare state programs. Results show that clients are more likely than others to believe--there's little or no waste in programs; government aid helps people to become independent; but, welfare and social security benefits are inadequate. Race, education, and marital status are, however, found to be better predictors of attitudes about the quality of welfare state programs than client status. An explanation for this is offered that focuses on personal exchange resources and citizen's relationships with non-governmental dependency targets. Also, hypotheses and models explaining client evaluations of welfare state programs are developed from two theoretical perspectives: power-dependence and gap/discrepancy. These are tested and estimated on a random sample of clients. Results show that outcomes, getting less than one expected, the extent of negative departures from justice, as well as the perception of influence, income, and program-type have a substantial effect on client evaluations of program quality. It is also found that being unmarried, a single mother, or female directly and negatively affects client outcomes and expectations and thus indirectly client evaluations of welfare state programs. Reestimation of the models, stratified by program type, discloses that only evaluations of public assistance programs show the effects of discrepancies and that program type masks the effects of several client attributes on evaluations of public assistance programs as well as the effects of race and marital status on evaluations of social insurance programs. It is suggested that these differences reflect the duality of the United States' welfare system and indicate a problem of perceived differential treatment assistance programs. Implications of the findings, research issues and policy questions for future work are discussed.