Study Details for:
Kemple, James, Daniel Friedlander, and Veronica Fellerath (1995). Florida's Project Independence: Benefits, costs, and two-year impacts of Florida's JOBS program, New York: MDRC.

Study URL: http://www.mdrc.org/sites/default/files/florida_project_independence_beefits_costs_fr.pdf (Link not working?)

Evaluation: Project Independence

Program Studied: Project Independence
See Study Characteristics tab below for more information about this program.

Strength of Evidence: 1-High High (3 of 3)

Populations Targeted Parents, Single parents, Welfare population
Setting Mixed (urban and rural)
Services Provided Case management, Education, Financial incentives or sanctions, Supportive services, Training, Work readiness activities
Outcome Domains Examined Favorable Impacts FoundShort-term employment, Long-term employment, Favorable Impacts FoundShort-term earnings, Long-term earnings, Education and training, Favorable Impacts FoundShort-term benefit receipt, Favorable Impacts FoundLong-term benefit receipt

Findings

Toggle Short-term employmentShort-term employment
Outcome Treatment Group Comparison Group Impact Findings Strength of Evidence Study Sample Sample Size Data Source and Timing

Ever employed, quarter 4 of year 1, %
Adjusted mean = 36.7 Adjusted mean = 34.6 2.2 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida UI earnings records, year 2; AFDC records, year 2; Food Stamp records, year 2

Employed in all quarters, quarters 2-5, %
Adjusted mean = 19.2 Adjusted mean = 17.4 1.9 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida UI earnings records, year 2; AFDC records, year 2; Food Stamp records, year 2
Toggle Long-term employmentLong-term employment
Outcome Treatment Group Comparison Group Impact Findings Strength of Evidence Study Sample Sample Size Data Source and Timing

Ever employed, quarter 4 of year 2, %
Adjusted mean = 38.3 Adjusted mean = 37.8 0.4 No Effect 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida UI earnings records, year2; AFDC records, year 2; Food Stamp records, year 2

Employed in all quarters, quarters 6-9, %
Adjusted mean = 23.1 Adjusted mean = 22.1 1 No Effect 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida UI earnings records, year 2; AFDC records, year 2; Food Stamp records, year 2

Employed at end of year 2, %
Adjusted mean = 38.3 Adjusted mean = 38.9 -0.7 No Effect 1-High High (3 of 3) Survey sample randomly assigned between September 1990 and May 1991 1,029 24-month survey
Toggle Short-term earningsShort-term earnings
Outcome Treatment Group Comparison Group Impact Findings Strength of Evidence Study Sample Sample Size Data Source and Timing

Total earnings, year 1, $
Adjusted mean = 2,548 Adjusted mean = 2,401 146 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida UI earnings records, year 2; AFDC records, year 2; Food Stamp records, year 2
Toggle Long-term earningsLong-term earnings
Outcome Treatment Group Comparison Group Impact Findings Strength of Evidence Study Sample Sample Size Data Source and Timing

Total earnings, year 2, $
Adjusted mean = 3,219 Adjusted mean = 3,138 80 No Effect 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida UI earnings records, year 2; AFDC records, year 2; Food Stamp records, year 2
Toggle Education and trainingEducation and training
Outcome Treatment Group Comparison Group Impact Findings Strength of Evidence Study Sample Sample Size Data Source and Timing

Had a GED or high school diploma by the end of year 2, %
Adjusted mean = 57.5 Adjusted mean = 56.2 1.3 No Effect 1-High High (3 of 3) Survey sample randomly assigned between September 1990 and May 1991 1,029 24-month survey

Had a trade certificate by the end year 2, %
Adjusted mean = 31.9 Adjusted mean = 31.6 0.3 No Effect 1-High High (3 of 3) Survey sample randomly assigned between September 1990 and May 1991 1,029 24-month survey

Had an associate's degree by the end of year 2, %
Adjusted mean = 4.4 Adjusted mean = 3.6 0.8 No Effect 1-High High (3 of 3) Survey sample randomly assigned between September 1990 and May 1991 1,029 24-month survey

Had a bachelor's degree by the end of year 2, %
Adjusted mean = 2.2 Adjusted mean = 1.9 0.4 No Effect 1-High High (3 of 3) Survey sample randomly assigned between September 1990 and May 1991 1,029 24-month survey
Toggle Short-term benefit receiptShort-term benefit receipt
Outcome Treatment Group Comparison Group Impact Findings Strength of Evidence Study Sample Sample Size Data Source and Timing
1 Negative impact is favorable.

Ever received any AFDC payments, year 1, %1
Adjusted mean = 85.2 Adjusted mean = 86.7 -1.5 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida AFDC records, year 1

Total AFDC payments received, year 1, $1
Adjusted mean = 2,196 Adjusted mean = 2,348 -152 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida AFDC records, year 1

Ever received any Food Stamp, year 1, %
Adjusted mean = 88.6 Adjusted mean = 89.1 -0.5 No Effect 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida Food Stamps records, year 1

Food Stamps value, year 1, $1
Adjusted mean = 2,148 Adjusted mean = 2,213 -66 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida Food Stamps records, year 1
Toggle Long-term benefit receiptLong-term benefit receipt
Outcome Treatment Group Comparison Group Impact Findings Strength of Evidence Study Sample Sample Size Data Source and Timing
1 Negative impact is favorable.

Ever received any AFDC payments, year 2, %1
Adjusted mean = 68.6 Adjusted mean = 71.3 -2.7 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida AFDC records, year 2

Total AFDC payments received, year 2, $1
Adjusted mean = 1,832 Adjusted mean = 1,945 -113 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida AFDC records, year 2

Ever received any Food Stamp, year 2, %
Adjusted mean = 72.6 Adjusted mean = 73.9 -1.3 No Effect 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida Food Stamps records, year 2

Food Stamps value, year 2, $1
Adjusted mean = 1,855 Adjusted mean = 1,925 -70 Favorable 1-High High (3 of 3) Full sample randomly assigned between July 1990 and August 1991 18,233 Florida Food Stamps records, year 2

Study Characteristics

Toggle Participants & Program Details Participants & Program Details
Participant Detail The research sample consisted of single parents (the vast majority were women) who were Aid to Families with Dependent Children (AFDC) applicants or recipients and who had at least one child age 3 years or older when the study began. About one-third of the research sample was white non-Hispanic, one third was black non-Hispanic, and 22.2 percent was Hispanic. The average age was about 32 years. About half of study participants had two or more children at the time the study began. More than half of study participants had a high school diploma or GED at the time of random assignment.
Program Services Eligible AFDC applicants and recipients attended an orientation at which they were assigned to one of two service tracks depending on their educational and work histories. In the first service track, participants who had been deemed "job-ready"—those who had completed at least grade 10 or worked at least 12 of the past 36 months—conducted a two-week job search that included contacting at least 12 employers. Those who were unable to find employment through the job search were assigned to a two- to three-week job club in which staff provided guidance on best practices in finding a job, including help in developing resumes and interview skills. If the participant was still unable to obtain a job, a case manager conducted a formal assessment of the participant’s employment goals and educational needs, helped the participant develop an employability plan, and referred the participant to additional education or training activities available in the community.

In the second service track, participants deemed "not job-ready" worked with a case manager who conducted the formal assessment, helped develop an employability plan, and referred participants to basic education or training programs consistent with the plan.

Participants had to fulfill participation requirements or else risk losing part or all of their AFDC grants for noncompliance. All participants were provided with or referred to support services as needed, including child care, tuition assistance, transportation, tools, and uniforms. Financial support for child care ended in January 1991 because of budget cuts, at which time participants were excused from employment and training requirements if child care services were unavailable. People assigned to the treatment condition who did not participate in services could lose part of their AFDC grant until they began to participate.
Program Duration Though not explicitly mentioned, authors suggested that clients could receive Project Independence services through the full follow-up period (two years).
Comparison Services In the comparison condition, AFDC recipients continued to receive benefits but did not have access to the Project Independence program and were not subject to the participation mandates that could lead to welfare benefit sanctions for two years. They were provided with a list of employment and training services available in the community and were potentially eligible for subsidized child care and tuition assistance programs.
Toggle Study Design Study Design
Strength of Evidence Description High
Group Formation The study was a randomized controlled trial in which evaluators randomly assigned welfare applicants within participating counties to the Project Independence treatment condition or the control condition between July 1990 and August 1991. Welfare applicants were eligible for the study if they were applying (or reapplying) for AFDC welfare benefits and potentially subject to the participation mandate (they did not have a child under age 3 years, were younger than age 60 years, were working fewer than 30 hours per week, were not in the second or third trimester of pregnancy, and were not permanently ill or incapacitated).

Evaluators interviewed a random subsample (who were randomly assigned between September 1990 and May 1991 and were from the public assistance units serving 10 of the largest urban centers in the nine research counties) for a 24-month follow-up survey.
Effect Calculation The effects reported by the authors and displayed on this site are the raw mean differences between the groups, adjusted for baseline characteristics.
Notes on Reported Outcomes The authors reported employment, earnings, educational attainment, and public benefit receipt outcomes for time points that are not the focus of this review.
Subgroups Studied Random assignment cohorts; age of the youngest child; job-readiness status; history of prior AFDC receipt
Other Domains Examined None
Toggle Study Setting Study Setting
Setting Details The study took place in nine counties in the state of Florida, which accounted for 58 percent of Florida's AFDC caseload as well as its eight largest cities. The nine counties were Bay (Panama City), Broward (Ft. Lauderdale), Dade (Miami), Duval (Jacksonville), Hillsborough (Tampa), Lee (Ft. Myers), Orange (Orlando), Pinellas (St. Petersburg), and Volusia (Daytona Beach).
Timing of Study This study reported the results of a two year follow-up for single parents who were randomly assigned to research groups from July 1990 through August 1991.
Study Funding The study was conducted under a contract with Florida's State Department of Health and Rehabilitative Services, with support from the Ford Foundation and the U.S. Department of Health and Human Services.
Study URL http://www.mdrc.org/sites/default/files/florida_project_independence_beefits_costs_fr.pdf