Grantee: University of Chicago. The full study report is linked here.
Description of the Intervention. Working on Womanhood (WOW) is a two-year program for high school girls, delivered within schools by masters-level educated, clinically trained social workers and counselors. Each of these staff members works with 50 – 55 girls across four to five groups in a school. Groups meet once per week, during the school day as part of an elective course in students’ schedules. The curriculum focuses on five core themes: self-awareness, emotional intelligence, healthy relationships, visionary goal setting, and leadership. For needs that cannot be addressed in a group setting, WOW counselors provide individual counseling and referrals to other services to a small subset of participating girls. WOW is rooted in Cognitive Behavioral Therapy (CBT), Acceptance and Commitment Therapy (ACT), and narrative therapy, and key learnings from other school-based mentoring programs. The program costs about $2,300 per participant.
Study Design. In fall 2017, the study randomly assigned 3,749 girls in 9th to 11th grade in 10 public high schools in Chicago to (i) a treatment group that was offered WOW participation, or (ii) a usual-services control group. This sample comprised over 95% of all 9th — 11th grade girls in these schools. 85% of the girls were Black or Hispanic; 95% were low income; and nearly 30% had personally witnessed someone being attacked, stabbed, shot at, hurt badly, or killed during their lifetime. 65% of girls in the treatment group attended at least one program session. For a random subsample of 2,011 girls, the study assessed mental health outcomes one year after study entry, using well-established, validated survey measures. 1
The study also measured educational and criminal justice outcomes for the full sample, using school and police records, over the two years following study entry.
Impact on the Preregistered Primary Outcomes. One year after study entry, the study found modest reductions in rates of clinical depression and PTSD at one year, but these effects are only suggestive due to lack of statistical significance (for clinical depression: 9% control vs 7% treatment, p=0.26; for clinical PTSD, 27% control vs 23.5% treatment, p=0.15). 2 The study found no statistically significant effect on the rate of the clinical anxiety (4% control group vs 4% treatment group).
The above findings are based on the study’s pre-registered primary analysis. The study measured several exploratory outcomes, of which there were suggestive impacts on broadly defined “clinical or at-risk” outcomes and symptom severity scores. The published study report contains additional details about the exploratory findings.
Study Quality. Based on a careful review, we believe this study was a well-conducted RCT that produced valid findings. 3
-
1
A planned second survey, to be administered two years after study entry, was cancelled due to COVID.
↩︎ -
2
The study may not have had a sufficiently-sized sample (i.e., “power”) to detect effects on clinical levels of anxiety, depression, or PTSD, given the unexpectedly low rates of these outcomes in the one-year survey.
↩︎ -
3
For example, the study had successful random assignment (as evidenced by highly similar treatment and control groups), valid analyses that were publicly pre-registered, and low-to-modest sample attrition in the one-year survey.
↩︎
