This study will use a national data set with nonprofit financial variables in a longitudinal scheme from 1997 to 2001 to capture the impacts of welfare reform on nonprofit human service organizations. Two fundamental and unresolved research questions are: 1) Has welfare reform increased or decreased government financial support of nonprofit human service organizations? 2) How have the diverse PRWORA policies influenced the organizations? The nonprofit human service organizations sample for this study is from the IRS/SOI sample files for 1997-2001 of the National Center for Charitable Statistics (NCCS) at The Urban Institute including information from Form 990—Return of Organization Exempt From Income Tax. Welfare reform data of this study includes states' funding data and states' policy data. States' funding data are states' expenditure on Temporary Assistance to Needy Families (TANF) non-assistance of 1997 to 2001 received from Department of Health and Human Service website. States' policy data are from Fender, McKernan, & Bernstein's (2002) research which created factor scores and cluster to summarize TANF and related policies. The dependent variable of this study is percentage of government support in nonprofit human service organizations. The independent variables are states variables including state expenditure, states' TANF policy factor scores, and states' TANF cluster. General Linear Models including one-way ANOVA, correlations, multiple regressions are applied in data of every year, and Hierarchical Linear Modeling (HLM) (Bryk and Raudenbush, 1992, 2002) are applied for data of all years in this study. From the total of 4,983 organizations in the IRS/SOI sample file from 1997 to 2001, there are 2,513 organizations in the category of human service. 1,968 (78.3 %) organizations are qualified for this longitudinal study because those organizations have at least 3 year-data available for Form 990 out of the total of 5 years period. The results show that the percentage of government support in the total revenue of those human service organizations are significant correlated to states' TANF non-assistance expenditure from Federal government, states' MOE expenditure, and the total expenditure, and TANF financial incentive scores from 1997 to 2001. States' cluster significantly predicts the percentage of government support in the total revenue of those human service organizations in 1998, F (4, 1845) = 2.728, p=0.028, in 2000, F (4, 2128) = 3.854, p = 0.004 and in 2001, F (4, 2237) = 3.349, p = 0.01. This empirical study found the percentages of government support were rising from 1997 to 2001. In addition, policy elements of welfare reform applied among states were connected to changes in the financial situations of nonprofit human service organizations. The results confirm that significant changes of government-nonprofit relations exist in post- welfare reform era. Since human service organizations play important roles in providing social service to those welfare leavers, the results urge future studies to continuously pay attentions to the changes beyond 2001.

Fink, B., & Widom, R. (2001). Social service organizations and welfare reform: Manpower Demonstration Research Corporation. Raudenbush, & Bryk (Eds.). (1992, 2002). Hierarchical linear models (2 ed.). Thousand Oaks: Sage Publications.