Project Activities
The researchers located, re-contacted, and consented adolescent students and parents who were initially in the CSRP study as preschoolers. They conducted student follow-up for 4 successive years, approximately covering students’ time in high school. Thus, these four waves of data represent long-term follow-up for the preschool intervention.
The first two waves of follow-up data included assessments of executive function and behavioral regulation and surveys regarding academic achievement. During the first of these two waves of data collection, the researchers randomly assigned students to a purpose for learning intervention (i.e., “dose 1” of the mindset intervention). Approximately 1 year later, students retained their treatment group assignment but received a growth mindset intervention module (i.e., "dose 2” of the mindset intervention). During the next 2 years, students and parents were recontacted and administered surveys regarding student academic outcomes and major life events. The researchers also collected administrative data from participant schools and third-party organizations like the College Board and National Student Clearinghouse. They then conducted extensive impact analyses for both interventions and executed a cost analysis for the potential scale-up of both intervention programs.
Structured Abstract
Setting
Participants for the preschool study were originally recruited from 18 Head Start centers that served low-income neighborhoods in Chicago. These centers were selected for study participation based on three criteria: (1) receipt of Head Start funding, (2) services provided to two or more full-day classrooms, (3) location in a high-poverty Chicago neighborhood. Intervention implementation and study recruitment proceeded across two cohorts, with cohort 1 participating during 2004-2005 and cohort 2 participating in 2005-2006. The adolescent follow-up data for the current study were first collected during the 2015-2016 school year. For the follow-up waves, students largely remained in the Chicago area, though some students participating in the study had left the area during the years since initial recruitment.
Sample
The original preschool study sample (n = 602) included approximately 66 percent of the sample identifying as Black and 27 percent as Hispanic, with a majority of the families living in poverty. At follow-up, 463 adolescents participated in the first mindset intervention module, and 430 participated in the second module 1 year later (i.e., dose 2). The demographic characteristics of the follow-up sample did not notably differ from the characteristics of the baseline sample.
The original preschool intervention provided treatment group teachers with 30 hours of training in classroom management strategies, weekly coaching through classroom-based consultation, and stress reduction workshops to limit burnout. Classroom consultants also worked one-on-one with small groups of 3 to 5 children who exhibited the most challenging behavioral problems. Initial impact evaluations showed that the intervention was successful at meeting its proximal goals. Observations showed that teachers in the treatment condition provided more emotionally and behaviorally supportive classroom environments than teachers in the control condition (Raver et al., 2008). Analyses of child outcomes reported that the intervention led to improvements in children's emotion regulation, executive function, and pre-academic skills (Raver et al., 2009; 2011).
The purpose for learning intervention lasted approximately 20 to 40 minutes. Students were asked consider problems in the world they would like to solve and to reflect on their own goals and values. As part of the module, students were presented with examples of how working hard in school could lead to personal fulfillment and the betterment of their broader community.
The growth mindset intervention (i.e., dose 2) was a module taken from the National Mindset Study (Yeager, Romero, et al., 2016). During the intervention, students were presented with information about how one can grow intellectually and improve in skills through productive struggle. This module gave students examples of how one’s brain can acquire new capabilities by continually engaging with difficult tasks.
Research design and methods
In the preschool study, researchers randomly assigned at the preschool site level after matching sites into pairs based on site characteristics. The random assignment occurred within these pair “blocks.” In the current study, the researchers randomly assigned at the student level for the first dose (purpose for learning), and this assignment persisted into the second dose (growth mindset).
The researchers collected follow-up data in 2015-2016 and 2016-2017, and these waves included direct assessments of self-regulation and survey measures of academic achievement and behavioral functioning. Students also participated in the two-dose mindset intervention during these two initial follow-up waves. During the 2017-2018 and 2018-2019 academic years, the researchers surveyed participants regarding academic outcomes, behavioral functioning, and major life events. At each wave, students and parents were consented for the release of academic records. The researchers also conducted cost analyses for both the preschool and mindset interventions, respectively. They accessed data on intervention implementation, including emails, project budgets, and planning documents to obtain information on intervention costs.
Control condition
The researchers provided a teacher’s aide 1 day per week to for the original preschool study control condition. For the follow-up study, researchers presented students in the control group for the purpose for learning intervention information about the differences between middle school and high school, and information about the importance of completing school assignments to meet performance goals to improve learning. Researchers presented students in the control group for the growth mindset intervention were provided information on promoting a healthy brain through sleep, exercise, and nutrition.
Key measures
The researchers measured student self-regulation in high school with direct assessments. These included the Hearts and Flowers task, the Emotional Go/No Go Task, the Dot Probe Task, and the Behavioral Indicator of Resiliency to Distress task. They measured student anxiety using the State-Trait Anxiety Inventory for Children. The research team also used the PSRA Assessor Report to score adolescents' cognitive and behavioral regulation. Student task persistence and engagement with college planning was measured using the researcher-developed College Knowledge Task. Students and parents reported on behavioral functioning across a variety of measures including the Behavioral Problems Index (parent report only), the Barratt Impulsiveness Scale, the Behavioral Rating Inventory of Executive Functioning, and the Risks and Strengths Scale (student report only). Students and parents also reported on adolescent academic achievement and college planning. The researchers obtained information regarding academic performance, behavioral problems, and school attendance throughout the study from the Chicago Public School district. T also obtained standardized test scores from College Board and ACT, student college enrollment information from National Student Clearinghouse, and information on family socio-economic characteristics from students and parents throughout the follow-up waves.
Data analytic strategy
The researchers conducted psychometric analyses throughout the project on direct assessments and survey reports of behavioral functioning. They analyzed survey measures using factor analyses. They ran key impact analyses using several related approaches for analyzing the multi-level structure of the data. For the treatment impact analyses, they tested sensitivity of results across a variety of modeling approaches. For the models in the current project that examined the later impacts of the preschool intervention, and for the models examining the impacts of the two new mindset interventions, the researchers relied most heavily on regression with fixed effects for the clustering unit of random assignment and clustered standard errors were used to adjust for school-level non-independence. For the final impact analyses, the researchers relied most heavily on hierarchical linear models with fixed effects included for the site-block variable and random effects included for Head Start sites. In their mediational analyses, the researchers relied on structural equation modeling.
Key outcomes
- Exploratory analyses suggested that children in the preschool intervention selected into higher performing schools during high school (Watts et al., 2020).
- The preschool intervention had some lasting positive effects on students’ executive function and academic grades at the beginning of high school (i.e., 10 to 11 years after the preschool intervention). Students in the intervention group also displayed heightened sensitivity to angry and sad emotional stimuli relative to the those in the control group, but there was no effect on student behavioral problems (Watts et al., 2018). Further analyses suggested that the long-term impact of the preschool intervention on students’ academic grades and executive functioning was partially mediated by initial intervention effects on children’s math skills and vocabulary (McCoy et al., 2019).
- The researchers found no effects for the 2-dose mindset intervention on students’ self-regulation, anxiety, critical motivation, and sense of belonging. They found some negative effects on students’ reported grades, but these negative impacts were not consistently observed across models (Gandhi et al., 2020).
- Final impact analyses from measures taken toward the end of high school (i.e., 11 to 14 years after the preschool intervention) suggested no sustained impact of the preschool intervention on executive functioning, emotional regulation, behavioral problems, academic performance, or college enrollment (Watts et al., 2023).
- Final analyses also found no impact of the 2-dose mindset intervention on college enrollment, with some negative effects observed on test scores (though results were largely null across most outcomes tested). Models examining the benefit of assignment to both the preschool and mindset intervention (i.e., 2-way interaction) also suggested no additional benefit for participating in both programs (Watts et al., 2023).
- Additional non-experimental analyses of the dataset explored predictors of college enrollment using child skills and behaviors in preschool. These analyses suggested that an early measure of attention and impulsivity control was the strongest predictor of later enrollment in college (Pan et al., 2023). Analyses also explored the long-term development of self-regulation over the course of elementary school and adolescence, with cross-lagged analyses suggesting that behavioral dysregulation and executive function develop largely independently of one another (Li-Grining et al., 2023). Finally, mixed-methods work explored issues relating to students’ college preparation, identifying organizational skills as a potential mediating factor linking cumulative risk indices to college preparatory actions (Li-Grining et al., 2023).
- Cost analyses (referenced in Watts et al., 2023 and available at this link) suggested that the pre-k intervention would cost approximately $1360 per student (2018 dollars), while the 2-dose mindset intervention would cost approximately $40 per student (2018 dollars).
People and institutions involved
IES program contact(s)
Project contributors
Products and publications
Project website:
Publications:
ERIC Citations: Find available citations in ERIC for this award here and here.
Selected Publications:
Gandhi, J., Watts, T. W., Masucci, M. D., & Raver, C. C. (2020). The effects of two mindset interventions on low-income students’ academic and psychological outcomes. Journal of Research on Educational Effectiveness, 13(2), 351-379.
Li, C., Hart, E. R., Duncan, R. J., & Watts, T.W. (2023). Bi-directional relations between behavioral problems and executive function: Assessing the longitudinal development of self-regulation. Developmental Science, 26(3), e13331.
Li-Grining, C. P., Roy, A. L., Koh, J., Boyer, A., Radulescu, M., & Naqi, Z. (2023). Black and Latino adolescents’ self-regulation: Placing college preparedness in context. Journal of Adolescent Research, 38(3), 423-455.
McCoy, D. C., Gonzalez, K., & Jones, S. (2019). Preschool self‐regulation and preacademic skills as mediators of the long‐term impacts of an early intervention. Child Development, 90(5), 1544-1558.
Pan, X. S., Li, C., & Watts, T. W. (2023). Associations between preschool cognitive and behavioral skills and college enrollment: Evidence from the Chicago School Readiness Project. Developmental Psychology, 59(3), 474–486.
Watts, T. W., Gandhi, J., Ibrahim, D. A., Masucci, M. D., & Raver, C. C. (2018). The Chicago School Readiness Project: Examining the long-term impacts of an early childhood intervention. PloS One, 13(7), e0200144.
Watts, T.W., Ibrahim, D. A., Khader, A., Li, C., Gandhi, J., & Raver, C. C (2020). Exploring the impacts of an early childhood educational intervention on later school selection. Educational Researcher, 49(9), 667-677.
Watts, T. W., Li, C., Pan, X. S., Gandhi, J., McCoy, D. C., & Raver, C. C. (2023). Impacts of the Chicago School Readiness Project on measures of achievement, cognitive functioning, and behavioral regulation in late adolescence. Developmental Psychology, 59(12), 2204–2222.
Available data:
Data for this project have been shared with ICPSR for public use.
- Watts, T. W., Das, S., Li, C., Pan, X. S., Gandhi, J., McCoy, D. C., Li-Grining, C., Roy, A. L., Jones, S. M., & Raver, C. C. (2023). The Chicago School Readiness Project: Adolescent Follow-Up, Illinois, 2004-2019. Inter-university Consortium for Political and Social Research [distributor], 2023-03-06. https://doi.org/10.3886/ICPSR38425.v1
Additional project information
Supplemental information
Co-Principal Investigators: Amanda Roy (University of Illinois-Chicago); Stephanie Jones (Harvard University); Dana McCoy (Harvard University); Cybele Raver (NYU)
Questions about this project?
To answer additional questions about this project or provide feedback, please contact the program officer.
