Author ORCID Identifier

0000-0001-9893-962X

Defense Date

2024

Document Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Psychology

First Advisor

Bryce D. McLeod, Ph.D.

Abstract

Teachers have difficulty delivering Tier-2 intervention programs for preschoolers at elevated risk for socioemotional and behavioral disorders with high intervention fidelity (Durlak, 2010; Sutherland et al., 2018). Several proximal factors that impact program fidelity have been identified, which may undermine program effects (Domitrovich et al., 2008; McLeod et al., 2022; Sutherland et al., 2018). Therefore, researchers have identified strategies, such as practice-based coaching, to improve intervention fidelity and support teachers (Hemmeter et al., 2015; Synder et al., 2015). Most of the field's knowledge of these topics is limited to variable-centered approaches (Fettig et al., 2022; Low et al., 2016). However, variable-centered approaches, as opposed to person-centered approaches, ignore the possibility that there may be underlying subgroups of teachers that can be identified based on their level of intervention fidelity. As such, the current study aimed to utilize a person-centered approach to identify whether there are distinct profiles of teachers based on their patterns in the level of intervention fidelity throughout practice-based coaching. Additionally, the study aimed to understand whether proximal factors and program outcomes are associated with profile membership. To achieve these aims, a secondary analysis of a multi-site cluster randomized control trial of the BEST in CLASS program, a Tier-2 teacher-delivered intervention for children at risk for emotional and behavioral disorders, was conducted (Conroy et al., 2018; Sutherland et al., 2018a). Data was drawn from teachers (n = 85) and children (n = 195) assigned to the BEST in CLASS study condition. To identify whether there are underlying subgroups of teacher-child dyads based on teachers’ patterns of intervention fidelity during coaching, two Repeated-Measure Latent Profile Analyses (RM-LPA) were conducted. The first RM-LPA using subscale-level adherence data as indicator variables across five time points revealed two distinct subgroups: Average Adherence and Above Average Adherence. Likewise, results from the second RM-LPA using competence subscale scores at the same five time points also found two distinct subgroups (“Average Competence” and “Above Average Competence”). Notably, no significant differences were found between the two adherence subgroups based on proximal factors or intervention outcomes. In contrast, subgroup membership based on competence scores was significantly associated with classroom climate and initial levels of challenging child behavior. Additionally, the “Above Average Competence” subgroup displayed significantly lower levels of challenging child behavior and teacher-child conflict at post-test than the “Average Competence” subgroup. However, no differences in teacher-child closeness were observed. The current study's results support the existence of distinct subgroups of teachers who vary in their patterns of intervention fidelity while receiving practice-based coaching. Findings also suggest that proximal factors and intervention outcomes are associated with subgroup membership, mainly based on levels of competence. These findings may help inform efforts to tailor coaching and professional development for teachers based on their unique needs.

Rights

© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

11-21-2024

Download

Share

COinS