The survey was fielded to teens in seven counties in Ohio: Cuyahoga (Cleveland), Franklin (Columbus), Hamilton (Cincinnati), Lawrence, Lucas (Toledo), Muskingum, and Stark.
Between 1989 and 1991 a total of 7,017 pregnant women and custodial parents under 20 years old who were receiving Aid to Families with Dependent Children (AFDC) and did not have a high school diploma or a GED were randomly assigned to the LEAP evaluation—80% to the intervention group and 20% to the control group. Due to changes in the program structure and implementation, 4,050 teens were excluded from the analysis. First, 1,442 older teens who experienced LEAP only during its start-up phase, when the program was undergoing problems operating under rules and procedures different from those that were eventually adopted, were excluded. Second, 2,608 teens
who were randomly assigned during the first year of program operations were also excluded so that the analysis sample was limited to teens who were enrolled after the bonus and sanction process was functioning smoothly. The remaining 2,967 teens were chosen as the research sample because they were more representative of participants in an ongoing LEAP program. Because exclusions were based on age at random assignment or date of random assignment, treatment-control equivalence was not disrupted. A survey was fielded to 1,178 of the remaining teens (all control group members and a random sample of 25% of the intervention group members in seven counties) three years after random assignment. The analysis was conducted on the 913 respondents (446 intervention and 467 control) to this survey.
More than 50% of the 913 teens in the three-year survey sample began LEAP when they were 17 or 18 years old, 33% entered the sample when they were 16 or younger, and 11% percent were 19. The average age was just over 17.5. Most participants entered LEAP with one child (71%) or were pregnant with their first child (21%), while few (8%) had two or more children. Just over half (58%) of the survey teens reported that they were enrolled in a junior high, high school, or GED program when they entered LEAP, while 42% were out of school. The average highest grade completed was 9.54. Nearly all teens in the survey sample are female (99%) and had never been married (94%) when they were randomly assigned. More than half (54%) headed their own welfare cases at the time of random assignment, with 40% on a parent’s AFDC case and 6% on another’s AFDC case. The sample was 67% African-American, 31% white, 2% Hispanic, and 1% other.
The program had a three-tiered incentive structure: grant increases ($62 for proof of enrollment plus $62 for each month in which they met attendance requirements), grant reductions ($62 for each month they failed to attend an initial assessment interview, failed to verify enrollment in school, or exceeded the allowed number of excused absences), and unchanged grants (exceeded the allowed number of total absences but not the allowed number of unexcused absences). Teens’ enrollment and attendance were monitored by case managers, who also offered guidance and authorized assistance with child care and transportation for teens complying with the rules. The $200
bonus for school completion was not provided during the period of the study.
Teens in the control group received normal cash benefits, with no bonuses paid or sanctions imposed for school enrollment and attendance.
Outcomes in each of the domains are included in this study: the measure of staying in school is defined as ever or currently enrolled in high school or a GED program; the measure related to progressing in school is ever completing grade 11, because it is the latest measure of progress; and the completion indicator is a combined measure of ever graduated high school or received a GED. All measures are taken from the survey fielded three years after random assignment.
Support for implementation
Information on staff training was not available.