CPRE - Consortium for Policy Research in Education at the University of Wisconsin Madison

Previous Experimental Research on FAST (Families and Schools Together, Inc.)

Description of FAST Experimental Design Data Sets
Federal Funder/
Location of FST schools/ethnicity of sample Inner City Milwaukee AfroAm/Latino Rural WI
Indian Reservation
(3 tribes)
Madison, WI 
Mixed race
Inner city NewOrleans AfroAmerican Madison, WI
# of FST sample
485 families/
10 schools
100 families/
3 schools
  9 schools 400 families/
9 schools
40 families
community based
# of FST cycles 20 cycles 7 cycles 9 cycles 16 cycles 4 cycles
Follow-up 2 years 1 year 1 year 1 year 6 months
Control Group FAST vs. FAME (mailings of parent ed) FAST vs. No-FAST FAST vs. No-FAST FAST vs. No-FAST Wait-list
Research Design-
Universal invitation of 2nd grade;
Randomized Classrooms prior to recruitment into study and program
K-2 Universal Invitation & recruitment;
Teacher CBCL pretests used to match pairs prior to Randomization
Mixed recruitment: universal plus some teacher identified, At risk prior to
Randomization of matched pairs based on teacher CBCL
Teacher identified at-risk list;
of list prior to recruitment into the study and into FAST
Recruitment across schools
Randomization of matched pairs based on parent CBCL assessment

National Institute on Drug Abuse (NIDA)

Summary: This RCT involved randomly assigning second-grade classrooms to either FAST or a comparison condition called FAME in 10 inner-city elementary schools serving at-risk, low-income communities (McDonald et al., 2006). The study included a 2-year follow-up. The participation rate among those who agreed to join the study was 89%, and the rate of retention for at least five sessions was 78% among those who participated, for an overall completion rate of 69%. An ITT HLM analysis of 2-year outcomes found that teachers blind to condition gave higher ratings of academic competence to children assigned to the FAST condition (effect size = .23) than to children assigned to the comparison condition (Moberg, McDonald, Brown, & Burke, 2002). In the sample as a whole, findings for behavioral outcomes were non-significant (Moberg et al., 2002). However, an HLM analysis that examined the program impact on Latino children in the sample (N = 130) found that at the 2-year follow-up, teachers gave Latino children in the treatment group significantly higher scores on academic competence and social skills and significantly lower scores on aggression than Latino children in the comparison group (McDonald et al., 2006). Moreover, rates of participation, retention, and overall completion were especially high among Latinos—at 90%, 85%, and 77%, respectively. Similarly, while 82% of FAST graduates (program completers) attended at least one FASTWORKS session, and the average number of FASTWORKS sessions attended was 7.1 over 2 years, 91% of Latino FAST graduates attended at least one FASTWORKS session, and the average number attended was 9.9. (This is the only study in which FASTWORKS participation was tracked.) These results are particularly important in light of the current study’s focus on Latinos and may reflect high levels of family social networks among Latinos, possibly compounding the benefits that accrue from adding school and community ties to family networks.  

Office of Educational Research and Improvement  (OERI) Grant
Families and Schools Together: An Experimental Analysis of a Parent-Mediated Early Intervention Program for At-Risk American Indian Children

Summary: Researchers at Wisconsin Center for Education Research performed this randomized controlled trial using a sample of students from three American Indian Nations in rural Northern Wisconsin. The goal of the study was to test whether participation in the FAST program could better the academic performance while reducing the problematic behavior of American Indian children. Another aim of the study was to create and test the effectiveness of a FAST program adapted to suit the needs, norms, and values of the three participating tribes.

This school-based experiment had a sample size of 100 students, who were spread across three schools. These students were paired based on tribal heritage, school, grade level, gender, and teacher’s rating. Non-response for one student in the pair resulted in the whole pair being dropped.  Forty of the fifty families graduated from seven cycles of FAST over three years. Using multiple indicators of academic and behavioral performance, parents and teachers scored children in a pretest, post test, and follow up. There were also qualitative observations in the classroom to assess behavior and social interaction of students performed by an American Indian research staff. 

Overall, this adaptation of FAST proved very successful. Teachers reported a statistically significant decrease in aggression, delinquency, and social conflict among students who had participated in FAST, as opposed to the students in the control group.  In addition, they perceived an increase in academic competence.  Furthermore, both teachers and FAST parents showed general satisfaction with the program.

Teacher Ratings of Academic Competence

Office of Special Education and Rehabilitation Services (OSERS) Grant
Families and Schools Together: An Experimental Analysis of a Parent-Mediated Early Intervention for Elementary Children

Summary: Designed as an intervention for children with emotional disabilities or high risk for special education services, this study aimed to test FAST”S ability to strengthen family relationships and build a support network for families within the school.

This randomized controlled trial, conducted with 67 families in an urban school district, recruited students both universally and through teacher referral.  Students were matched into pairs based on their grade, gender, and how their teachers rated them on the Child Behavior Checklist.  Students and families in the control condition received ongoing services provided by the school, while students in the experimental condition received the FAST intervention. The retention rate for this study was 90%.

Researchers measured family involvement in the special education evaluation process, the emotional/behavioral status of the child, and the child’s academic and social performance.  Data was collected through a pretest and posttest, both of which used parent and teacher ratings on the CBCL and SSRS.  Academic achievement was assessed using standardized test and state-mandated assessment data, while family relationships were evaluated using the Family Environment Scale. 

Overall, FAST had a positive impact on reducing the number of special education referrals for students who participated with their families.  FAST families scored higher on the Family Adaptability Scale, and FAST parents rated their children as being less aggressive. Furthermore, fewer FAST students ended up receiving special education services than students in the control condition, which indicates that there are cost saving benefits of using FAST as a prevention tool.

Madison FAST Experimental Study

Administration of Children and Families (ACYF) ABT: National Evaluation of Family Support Programs

Summary: ABT performed this randomized controlled trial in New Orleans to test the impact of FAST on children with behavior or academic problems.  Overall, there were 400 students and families, predominantly African-American, who participated in these trials.
This sample was split into two cohorts, which cycled through the experiment at different times.  Students were assigned to either the experimental group, which received eight weeks of FAST with the option of participating in FASTWORKS, or the control group, which received weekly pamphlets on parenting.

Researchers collected data using primary caregiver interviews, teacher evaluations, and report cards administered pre-experiment, immediately post-experiment, and one year after the experiment had ended. Specifically, researchers were measuring children’s social activities and behaviors, academic progress, learning environment, and literacy activities. In addition, they also examined family environment and parenting, parent’s social support and connectedness, parent’s mental health, school-family connections, and community participation of parents.  Besides these measures, there was a qualitative evaluation of the implementation process at each school site to check for program integrity.

The results indicate that FAST did have a significantly positive impact on its participants. Parents in FAST rated their children’s behavior more positively, did more volunteer work within the school, and were more likely to be in a school leadership position one year after FAST ended.  However, there was no impact on behavior or academic performance as evaluated by teachers, nor was there a difference in family environment, or parenting outcomes.

New Orleans FAST Experimental Study