The 5-year impacts of mandatory welfare-to-work programs on welfare recipients and their children were examined by using a rigorous research design called a social experiment to examine 11 welfare-to-work programs in 6 states (California, Georgia, Michigan, Ohio, Oklahoma, and Oregon). Four employment-focused and seven education-focused programs were examined. Data were collected from administrative records, state and county welfare payment records, and surveys of mothers and children over the 5-year study. In the absence of welfare-to-work programs, approximately three-fourths of single-parent welfare recipients found jobs and more than half left the welfare roles. Although few of the 11 programs improved on that already-high rate of job finding, nearly all the programs helped single parents work more hours during more quarters of the follow-up period and earn more than they would have in the absence of a program. The most effective program used an employment-focused approach that initially assigned some enrollees to very short-term education and training and others to job search. Impacts for children differed more by program site than by welfare-to-work approach. The following items are appended: table and figure notes; supplementary tables; a survey response analysis; a comparison of impacts estimated from different data sources; and selected impacts for various child survey samples. (Contains 121 tables and 282 references.) (MN)