Evaluation of Simulated Learning in Teacher Preparation Across Universities

• In Event: Causal Evidence to Support the Development of an Equitable and Effective Teacher Workforce

Abstract

Objectives
Research suggests coaching can expedite teachers’ skill development (Kraft et al., 2018) but is under-utilized during pre-service preparation (Matsko et al., 2020). We examine the effects of coaching on candidates’ classroom management skills, as well as their perceptions of student behavior, using a mixed-reality simulation platform as a standardized assessment platform and practice space (see Figure 1). We assess whether coaching between simulations leads to more rapid development of candidates’ skills and productive perceptions of student behavior. We also test whether the benefits of coaching replicate across a range of teacher preparation contexts.

Perspective
Research suggests candidates need practice addressing student behavior in ways that preserve student dignity and keep children in the classroom. Preservice teachers report feeling ill-prepared to address off-task student behavior (Le Maistre & Paré, 2010), which may contribute to novices being more likely than experienced teachers to employ exclusionary disciplinary practices (Glock & Kleen, 2019).

Research suggests candidates benefit from extensive practice opportunities (Karpicke & Bauernschmidt, 2011) and “deliberate practice” that couples practice with coaching (Ericcson & Pool, 2016).

Methods
Our sample includes 290 candidates from three diverse preparation programs: 1) a master’s program at a research-intensive university (N=80); 2) an alternative route program, where participants were teachers-of-record (N=88); and 3) an undergraduate program at a teaching-focused university near the US-Mexico border (N=122). Across programs, there is substantial variation in candidates’ characteristics (see Table 1 for baseline balance and descriptive demographic statistics across sites). Outcomes include observable redirection skills and candidates’ perceptions of student behavior, using a modified version of the IOWA Connor’s behavioral rating scale (Waschbusch & Willoughby, 2008). We also asked candidates their likelihood of referring the disruptive student avatar to special education services based on their experience in the simulation.

We employ a multisite block randomized design where candidates within sites are randomized to receive coaching (N=144) or a self-reflection protocol between simulations (N=139). The intervention involves a 4-step coaching model focused on behavioral redirection skills. We use a multisite design to examine the replicability of coaching effects across programs and a block design for examining replicability of effects across candidates’ characteristics (see Figure 2 for overview of study design).

Results

Coaching improved candidates’ skills in redirecting off-task behaviors across all three sites (Table 2). The fixed-effect meta-analytic coaching effect was positive, large, and statistically significant (1.65 SD, p-value < 0.01). Results also indicate that across sites, coached candidates perceived student avatars’ behaviors as less problematic than those who self-reflected between sessions (-.37 SD, p-value < 0.01). Only coached candidates at site 2, the alternative route program, were less likely to refer the avatar to special education services.

Significance
Nearly a hundred teacher preparation programs use this simulation technology, with little empirical evidence to guide decisions about use. Our research capitalizes on a large and diverse sample to better understand whether coaching amplifies the benefits of simulated practice. These findings suggest coaching is an important complement to simulated practice and has positive effects on a range of outcomes across diverse program contexts.

Authors

• Julie Jackson Cohen, University of Virginia

• Vivian Wong, University of Virginia

• Qing Poppy Liu, University of Virginia

• Kristyn Wilson, University of Virginia

• Anna Yonas, University of Kansas