|Title:||An Integrative Data Analysis of the Coping Power Program: Exploration of Academic Outcomes, Subgroup Effects, Mechanisms, and Implementation Factors|
|Principal Investigator:||McDaniel, Heather||Awardee:||University of Virginia|
|Program:||Social and Behavioral Context for Academic Learning [Program Details]|
|Award Period:||3 years (07/01/2022 – 06/30/2025)||Award Amount:||$1,293,703|
Co-Principal Investigators: Bradshaw, Catherine; Morgan-Lopez, Antonio
Purpose: The purpose of this project is to conduct an integrative data analysis (IDA) which combines individual-level student data collected across 11 randomized controlled trials (RCTs) of the Coping Power Program (CP), an evidence-based, school-based, preventive intervention effective in reducing externalizing behaviors in children demonstrating aggressive, disruptive behaviors. Adoption and implementation of evidence-based behavioral interventions in schools continues to be a multifaceted challenge, with inconsistent evidence around whether evidence-based behavioral interventions impact academic outcomes; moreover, it is unclear "what works, for whom, how, and under what conditions". Any single RCT is rarely adequately powered to detect important moderation ("for whom") and mediation ("how") effects and often do not capture enough contextual heterogeneity in implementation or participant engagement to comprehensively answer important implementation-related questions. To address these gaps, the current project will explore "how" and "for whom" CP is associated with academic outcomes and understanding of how implementation and engagement factors are associated with student outcomes.
Project Activities: The researchers will harmonize data collected across 11 RCTs of CP, in an IDA framework. They will (1) explore the crossover effects of CP on academic outcomes, as well as other school-related behavioral outcomes, with consideration of malleable subgroup effects to better understand "for whom" CP is most effective; (2) explore how CP impacts academic and related school outcomes through CP mechanisms of action, as well as reductions in externalizing difficulties; (3) investigate how implementation of CP (e.g., clinician adherence; youth and parent dosage; youth engagement) relates to academic and other school-related behavioral outcomes; (4) explore how study, individual, and program characteristics relate to parent and youth dosage and engagement in CP.
Products: This research will provide rigorous evidence about potential crossover effects of CP on academic outcomes, as well as information about subgroup effects, mechanisms, and implementation of the program that may enhance future implementation of CP in schools. The study team will disseminate the findings to diverse groups of researchers and stakeholders to inform training, practice, research, and policy in education. They will also produce a single dataset that harmonizes individual-level data from 11 RCTs of CP and prepare it for public archiving so the data can be used by other researchers.
Setting: The 11 RCTs of CP were conducted in elementary, middle, and high schools in the Southern region of the United States.
Sample: The team will integrate data from 11 RCTs of CP which includes approximately 3,100 students in over 180 elementary, middle, and high schools. Participating students demonstrated elevated disruptive, aggressive behavior based on teacher screeners.
Factors: The Coping Power program is an evidence-based, targeted, school-based preventive intervention for youth demonstrating disruptive, aggressive behavior. RCTs included in the present IDA include the standard CP program, as well as variations of the original program, including individually-delivered CP, optimized CP (e.g., adding mindfulness components) and developmentally-adapted CP (i.e., developmental adaptations). The research team will explore student-related (e.g., gender, race, baseline inhibitory control) and study-related (e.g., CP variation) moderators of CP impacts, as well as mediators of program impacts (social-cognitive skills, parenting, externalizing problems) on student academic outcomes. In addition, they will explore how implementation factors (i.e., implementer adherence, parent and student dosage and responsiveness) are associated with student outcomes.
Research Design and Methods: The researchers will use a state-of-the-art methodological approach, Integrative Data Analysis (IDA), which will allow the integration of individual-level data from 11 RCTs. In contrast to standard meta-analysis which pools study-level data, IDA facilitates pooling of individual data from multiple RCTs, offering benefits of a large sample size, including increased power (e.g., for moderation and mediation analysis) and sample heterogeneity (e.g., student characteristics as well as intervention implementation and engagement), as well as the ability to answer new research questions.
Control Condition: Each of the 11 included RCTs has a control condition (i.e., school-as-usual) or compares an augmented or adapted CP to the standard CP.
Key Measures: The focal outcomes in this study are academic and school-related behavioral outcomes. School-related behavioral outcomes will be assessed by teacher- and parent-report on broadband social, emotional, and behavioral health assessments. Archival school record data has been collected across studies for participating youth, including grades in core subjects, disciplinary referrals, and standardized test scores; however, few prior studies of CP examined academic outcomes as the studies were largely focused on mental health, drug, or violence-related outcomes (as per the funding sources were NIJ, CDC, NIMH, NIDA). Mediators, moderators, and other key covariates will also be based on student-, teacher- and parent-reported surveys. Measures will be harmonized across studies.
Data Analytic Strategy: Across all aims, the research team will use a general, flexible, multilevel structural equation modeling framework. First, the research team will logically harmonize data across RCTs. For latent constructs, they will use moderated nonlinear factor analysis (MNLFA) to harmonize data analytically and estimate scale scores that account for differential item functioning across many predictors. Subsequent modeling will account for the nested dependencies in the data using multilevel modeling, combined with propensity score weighting, mediation, moderation analyses, and mixture modeling, as relevant for each aim.
Related IES Projects: Testing the Efficacy of a Developmentally Informed Coping Power Program in Middle Schools (R305A140070); Coping Power for Rural Middle Schoolers: A Tiered Approach to Increasing Behavioral and Mental Health Supports and Reducing Disparities (R305A190116); The National Center for Rural School Mental Health (NCRSMH): Enhancing the Capacity of Rural Schools to Identify, Prevent, and Intervene in Youth Mental Health Concerns (R305C190014)