Texas officials have decided not to participate in a proposed federal study of the effectiveness of new education and job-training programs for welfare recipients, on the grounds that the study could unfairly deny such benefits to some people.
The 1989 federal welfare-reform law mandated a study of the new Job Opportunities and Basic Skills training program, and required that the progress of participants be compared against a “control” group of welfare recipients who did not receive services.
Citing ethical questions posed by randomly assigning to the control group welfare recipients who might otherwise receive benefits, the Texas officials said last month that they did not want to be among the 10 states that will be chosen this fall for the study.
But welfare officials in other states, including Ohio and California, have indicated that they are willing to be a part of the evaluation.
Backers of the study say use of a control group is essential to achieve valid results on the value of the as-yet-untested education and training programs. Moreover, they say, the 10,000 people to be assigned to the control group will be part of a considerably larger number of welfare recipients who will not receive the services because of lack of funding.
So far, 18 states have indicated that they would like to participate in the study.
Ethical Issues
Sally Griffin, director of service delivery and client self-support services in Texas, said her state declined to participate in the program “based on the whole issue of taking people out of availability of services.”
“I think this method is an accepted practice,” Ms. Griffin said. “But I think it does boil down to some ethical issues.”
She said state officials made the decision in the wake of a controversy over a state experiment that could have denied certain transitional Medicaid and child-care benefits written into the new welfare-reform law to some recipients in order to continue an ongoing study.
That study was killed after an outcry in the state.
“I think coming out of that situation,” she explained, “we decided it was not good to take thousands of people out of the running for the jobs program.”
But “the jobs program is not an entitlement,” said Judith Gueron, president of the Manpower Demon8stration Research Corporation, the federal contractor for the study. “Not everyone eligible will be served.’'
Most states will serve 10 percent or less of those eligible for the education services in the first years of the program, she noted.
“It’s one thing to deny Medicaid benefits to someone in order to place them in a control group,” Ms. Gueron said. “Medicaid is an entitlement, and we can be sure that it is a good thing.”
“But the jobs program is not an entitlement, and we don’t know if it works,” she added. “We’ve never tried mandatory education for benefits in this country before.”
“I think [random assignment] is a spurious issue,” said Paul Offner, deputy director of Ohio’s department of human services. “We are running various demonstration programs all over Ohio. But you don’t hear people complaining that we shouldn’t because the teen parents in one county might get services that teen parents in others won’t.”
