Evaluation of the Impact of Wisconsin's Learnfare Experiment on the School Attendance of Teenagers Receiving Aid to Families with Dependent Children
by John Pawasarat, Lois M. Quinn and Frank Stetzer, Employment and Training
Institute, University of Wisconsin-Milwaukee, February 1992
NOTE: THE FULL REPORT IS AVAILABLE ONLINE
The Employment and Training Institute of the University of Wisconsin-Milwaukee was selected
by the Wisconsin Department of Health and Social Services to conduct two of the three state
evaluations of Wisconsin welfare reform initiatives. The Institute prepared a research design
for the evaluation study of Learnfare's impact on school attendance of teens receiving Aid to
Families with Dependent Children (AFDC), which was approved by the Department of Health
and Social Services in July, 1991. For the study the evaluators reviewed school attendance
records for a six-year period for over 50,000 teens in Milwaukee Public Schools who were in
families receiving or formerly receiving AFDC. School records were also analyzed for nearly
6,000 teenagers from five school districts representative of the remainder of the state. Schools
studied included another urban school district with a relatively high minority student enrollment,
two mid-size Wisconsin school districts, and two small rural districts. Samples were not used.
In each school district, the entire AFDC and former AFDC teen population was studied. Major
findings of the study include:
- Using lagged regression models which controlled for differences in age, grade level, sex,
race, and months on AFDC, the school attendance of AFDC teens under the Learnfare policy
was compared to school attendance of teens formerly on AFDC and teens receiving AFDC prior
to the Learnfare experiment. High school teens subject to the Learnfare attendance requirement
did not show improvement in attendance during the Learnfare experiment in any of the six
school districts studied, as compared to control groups of AFDC and former AFDC teen
- To test whether the lack of improvement in attendance attributable to the Learnfare
requirement was resulting from the retention of potential dropouts, a second series of lagged
regression analyses included dropouts in both the control and experimental groups as absent for
the entire semester. Again, no improvement in attendance was found for high school teens
subject to the Learnfare attendance requirement.
- Similarly, the Learnfare requirement was found to have no impact on reducing semester
absences among eighth grade students subject to the policy.
- In Milwaukee
AFDC teens showed a slight, but statistically significant, increase in absences
under the Learnfare policy. Comparing Spring 1989 to Spring 1990, about 7 percent of teens
under Learnfare missed the same number of days of school each semester, 36 percent had fewer
absences the second year, and 56 percent had more absences the second year. For the control
group of teens not under the Learnfare policy, about 6 percent missed the same
number of days, 44 percent had fewer absences the second year, and 50 percent had more
absences the second year.
- The lack of holding power of the Learnfare
sanction was also seen in descriptive statistics for the Milwaukee Public Schools population.
After one year of Learnfare, nearly half (47 percent) of in-school youth whose families received
AFDC sanctions for their poor attendance dropped out of school completely.
- Graduation rates for Milwaukee teens subject to Learnfare who entered high school as
Freshmen in the 1987-88 school year and a control group of low-income classmates were the
same, with 18 percent of each group actually finishing their senior year and graduating in June,
1991. The graduation rates for the next largest urban school district were 48 percent for the
Learnfare group and 49 percent for the control group.
- In Milwaukee the percentage of high school students missing more than a day of school
a week on average increased over the three years of Learnfare. By Spring semester of 1991
over 40 percent of Learnfare teens were missing more than a day of school a week on average.
The next largest urban school district showed similar patterns to Milwaukee with over 30 percent
of high school teens missing more than twenty days in the fall, and more than 40 percent of
Learnfare teens missing more than twenty days of school in two of the Learnfare spring
semesters. One of the rural districts showed over 60 percent of Learnfare teens with forty or
more absences the third year of Learnfare.
- Nearly half of teen parent non-graduates in Milwaukee were never required to attend
school under threat of Learnfare sanctioning. Of Milwaukee Public School teen parents required
to attend school under the Learnfare policy and threatened with financial sanctions, less than half
were enrolled in school. Subsequently, well over half (56 to 64 percent) of this population was
sanctioned each semester.
|Milwaukee Public School Teen Parents on AFDC
Enrolled||Percent not Enrolled||Percent Exempt from
MPS Teen Parents Required to Attend School Under Learnfare |
Enrolled|| Percent Sanctioned|
by Lois M. Quinn, January 1997
University of Wisconsin-Milwaukee Employment and Training Institute completed the
Evaluation of the Impact of Wisconsin's Learnfare Experiment on the School Attendance
of Teenagers Receiving Aid to Families with Dependent Children in February 1992.
After the Employment and Training Institute refused to comply with written instructions from
the Wisconsin Department of Health and Social Services to suppress its research findings and
to revise the evaluation according to new specifications developed by the Department, the
Department canceled all contracts with the Institute, including contracts to evaluate the
effectiveness of case management services offered under Learnfare and analyses of the impact
of the Learnfare policy on AFDC families.1
In May 1992 the Wisconsin Department of Health and Social Services entered into contract with
the Wisconsin Legislative Audit Bureau to conduct a second evaluation of the impact of the
Learnfare experiment on teenage attendance. While Learnfare expenditures had more than
doubled since the period of the first evaluation (see table below), the third semester evaluation
of the Audit Bureau reported similar findings to those observed by the Employment and Training
Institute. The Audit Bureau found no positive effects on the attendance of teenagers in the
Learnfare population studied or for any of the subgroups analyzed, including dropouts, teenage
parents, older teenagers, younger teenagers, teenagers living in Milwaukee County or teenagers
living in other parts of the state.2 Furthermore, the Audit
Bureau reported attendance rates for
Learnfare teens which suggested the same serious and accelerating truancy problems identified
in the Institute's first evaluation study.
Average Attendance Rates of Teenagers Under the Learnfare Requirement
Audit Bureau Evaluation) 3
|Age of Teens || First Semester || Second
Semester || Third Semester|
|14- and 15-year-
|16- and 17-year-
|18- and 19-year-olds
While the Learnfare policy has not shown positive impact on the serious problems of truancy
among low-income youths, the program remains politically popular. Since 1988 Wisconsin has
invested over $70 million in Learnfare-related administrative costs, child care services and
special projects aimed at AFDC teens.
Wisconsin Learnfare Expenditures: 1988-1996 4
| Calendar Year || State Funds || Federal Funds
|| Local Funds ||Total |
| 1988||$983,200 ||$1,229,700 ||$0
| 1989 ||1,391,400|| 1,754,300 ||0
| 1990|| 1,733,000|| 2,315,800||
252,100|| 4,300,900 |
| 1991|| 2,765,900|| 3,422,400||
729,700|| 6,918,000 |
| 1992|| 3,027,200|| 5,095,800||
1,945,700 ||10,068,700 |
| 1993 || 2,939,500|| 5,024,000||
1,965,100 ||9,928,600 |
| 1994 ||2,866,500|| 5,385,400
||1,819,000 ||10,070,900 |
| 1995|| 3,108,500|| 5,955,300||
2,708,200|| 11,772,000 |
| 1996|| 3,591,800||
| Total ||$22,407,000 ||$36,612,000
||$11,579,200 ||$70,598,200 |
The reasons for the ineffectiveness of the Learnfare policy are not readily apparent. Several
possible factors could be evaluated.
- In Milwaukee County Learnfare enforcement has fallen apart with
thousands of teens placed
in monthly "hold status" because their attendance records are not effectively
monitored by state-supported staff. However, record-keeping did not appear to be a problem
in many other areas of Wisconsin where Learnfare impacts were not positive.
some cases teens' truancy problems may be a manifestation of other more serious family
problems. In Milwaukee County, 36 percent of teen parents (570 of 1,562 teen parents) and 15
percent of non-parent teens (757 of 5,050 non-parent teens) sanctioned for poor school
attendance over a sixteen month period had social service or court records indicating prior
reported child abuse or neglect for themselves or other children in their
other cases, parents may lack the will or ability to enforce Learnfare attendance requirements.
The problem is particularly serious for youths who have a history of truancy and who lack
sufficient credits to graduate from high school.
- Case management "services" observed during the preliminary
stage of the
Learnfare evaluation were not impressive. Because of the lack of rigorous evaluation or
accountability for projects funded as "Learnfare-related," a number of anti-truancy
programs selected for funding by the state did not appear to be well-designed or effectively
- For a discussion of the events leading to the University's contract cancellation, see Lois
M. Quinn and Robert S. Magill, "Politics versus Research in Social Policy," Social Science
Review 68 (December 1994): 503-520.
- Wisconsin Legislative Audit Bureau, An
Evaluation of the Third-Semester Effects of the Wisconsin Learnfare Program (Madison, Wis.:
- Ibid., 22-24.
- The amounts for 1988 through 1995 are actual expenditures; the figures for 1996 are
contract amounts, according to the Wisconsin Department of Health and Social Services. The
Department estimates that Learnfare sanctions totaled $12.1 million from March 1988 through
May 1996. Rob Reinhardt, Fiscal Analyst, Legislative Fiscal Bureau, to Representative Barbara
Notestein, 21 May 1996.
- See Lois Quinn, "Using Threats of Poverty to Promote
School Attendance: Implications of Wisconsin's Learnfare Experiment for Families,"
Journal of Children and Poverty 1 (Summer 1995): 5-16, and John
Pawasarat and Lois M. Quinn,
The Impact of Learnfare on Milwaukee County Social Service
Clients, University of Wisconsin-Milwaukee Employment and
Training Institute, March 1990.
& Training Home Page
Employment and Training Institute
University of Wisconsin-Milwaukee
Direct comments to: email@example.com