Analysis of Two Studies of Quality in New York City's Aid to Families With Dependent Children Program
HRD-80-48: Published: Feb 7, 1980. Publicly Released: Feb 21, 1980.
- Full Report:
An analysis was made of two studies of quality in New York City's Aid to Families with Dependent Children (AFDC) program to determine: (1) if the difference between the error rate estimates was significant, and (2) which estimate was most accurate. These studies were made by the State Quality Control (QC) staff and the State Office of the Welfare Inspector General (IG). AFDC, one of the largest federally aided public assistance programs, is administered by the States in cooperation with the Department of Health, Education, and Welfare (HEW). Both HEW and Congress have been concerned for several years with the high incidence of erroneous payments to AFDC recipients. To reduce the error rates, HEW has encouraged States to implement quality control programs, which all States now operate. The statistically valid sample of AFDC cases which the State selects every 6 months is the most important part of quality control. The IG and QC made independent studies in 1976 and reported different eligibility error rates for New York City.
In the IG study of August 1976, sample cases showed that about 21.5 percent of the welfare recipients in New York City were ineligible for benefits. The QC study based on sample cases from July to December 1976, revealed that 10.2 percent of the city's AFDC recipients were ineligible for the benefits they received. However, these studies were not directly comparable, because they covered different time periods and different welfare programs. The 1,200 QC samples were selected from a 6-month period and the 228 IG samples were selected from a 1-month period. Both studies covered the month of August. Error-causing conditions that were prevalent in August could have influenced the IG error rate more than the QC error rate. While specific error causes more prevalent in August were not identified, it was found that the QC AFDC ineligibility error rate for August was higher than the overall 6-month QC error rate. The QC cases were selected only from the AFDC register and the IG cases were selected from the State and locally financed home relief welfare program registers in addition to the AFDC register. The error rates of both studies were compared using only the August AFDC payments selected, the only payments that were comparable. Differences in the two studies' ineligibility error rates were much smaller for the August AFDC cases than for all cases on the two samples. Both studies treated financial support from legally responsible relatives differently, which made the IG error rate higher.