Achieving Change for Texans, 1997-2002
Achieving Change for Texans: Statewide, the following were key program provisions: (1) time limit of 12 to 36 months for able-bodied adults, depending on work experience and education, with exemptions for severe local economic conditions and personal hardships; (2) establish detailed Personal Responsibility Agreement, as an eligibility condition, requiring JOBS participation, child support cooperation, medical screening and immunization and school attendance for children, and prohibiting job quits, and substance or alcohol abuse; (3) sanctions for failure to comply with JOBS and child support at $78/month for one-parent and $125/month for two-parent families, and $25/month for all other mandatory provisions (maximum $75 per month); and (4) transitional child care and Medicaid extended from 12 to 18 months for families exhausting benefits. In addition, pilot sites tested "fill-the-gap" budgeting combined with disregarding assets in Individual Development Accounts, and "One-Time” voluntary grant diversion payment of $1,000 to TANF eligibles agreeing not to reapply for 12 months were evaluated by quasi-experimental or survey methods.
The major research questions include: Policy impact questions were: (1) Was self-sufficiency increased according to various measures of employment, earnings and income? (2) Was welfare use reduced according to various measures of the duration and amount of welfare receipt? (3) Was participation in program activities and use of program services increased, according to such measures as attendance at, and completion of, assigned training or employment activities? Questions related to policy impact on family structure and stability, child well-being, and health benefits were also explored.
The evaluation employed a random assignment experiment operated in various counties, which implement different sets of policies. The experimental group was subject to all provisions described above, except the "One-Time" grant diversion, which was evaluated with a survey. Implementation, operations and impacts were observed using administrative records, site observation, staff and client interviews. The evaluation included process, impact and cost/benefit analyses.
ACF provided multi-year grants for 23 evaluation projects in 21 States that continued or modified evaluations initiated under welfare reform demonstration. These demonstrations were begun under waivers prior to implementing TANF. Nine States (Arizona, Connecticut, Florida, Indiana, Iowa, Minnesota, Texas, Vermont and Wisconsin) received funding as Track 1 projects to continue unchanged or with limited modifications, the evaluations of the welfare reform demonstrations that they had begun operating prior to the implementation of TANF. Thirteen States (California, Illinois, Iowa, Maryland, Minnesota, Montana, Nebraska, New Hampshire, North Carolina, North Dakota, Ohio, South Carolina and Virginia) received funding as Track 2 projects to continue, modify or replace evaluations of welfare reform demonstration they had begun operating prior to implementing TANF. Although there was significant variation among the States in the actual policy strategies that they adopted, the policies had common goals related to promoting self-sufficiency. The policies included time limits on assistance, stronger requirements to participate in work or work-related activities, stronger sanctions for non-cooperation, making work pay, and promoting personal responsibility.
By the end of FY 2003, most of these evaluation projects had been completed. Evaluation work continued on five projects in four states (California, Indiana, New Hampshire and Ohio).
The major research questions include: Common research questions about impacts included: (1) Was self-sufficiency increased according to various measures of employment, earnings and income? (2) Was welfare use reduced according to various measures of the duration and amount of welfare receipt? (3) Was participation in program activities and use of program services increased, according to such measures as attendance at, and completion of, assigned training or employment activities? Also, most States included policies to improve family stability and child well-being, and these were reflected in their impact research questions. In addition to these impact issues, each evaluation assessed program strategies and problems in implementing and operating welfare reform programs, including such major factors as planning, staffing, training, organizational structure, management and supervision, and application of program rules and delivery of services to clients.
Some projects (particularly Track 2) are limited to process evaluations related to assessing program strategies and problems in implementing and operating welfare reform programs, including such major factors as planning, staffing, training, organizational structure, management and supervision, and application of program rules and delivery of services to clients.
Some examine the breadth of a mix of welfare reform policies related to time limits, stronger work requirements and personal responsibility and their effects on self-sufficiency, welfare receipt and participation in program activities. Meanwhile, others focus on more specific issues such as post-employment services or front-line assessment and placement in job activities where policy choices have devolved to the county level.
Although States varied in particular elements of their design, all States funded under Track 1 used a random assignment experiment to answer impact questions. Under this approach, experimental group cases are subject to welfare reform policies as modified by waiver policy and control group cases are subject to AFDC policies in place prior to welfare reform. Each evaluation also utilized record data, surveys, site observations, staff interviews, and program documents to obtain information about the research population, and completed process, impact and cost-benefit analyses.
All Track 2 projects provide for process analyses on the implementation of welfare reform policies. Some include impact analyses based on either experimental or non-experimental research designs.