Measuring Preservice Teachers' Instructional Practice

• In Event: Examining Shifts in Teachers' Student-Centered Practices as a Result of Practice-Based Teacher Education

Abstract

Objectives and Perspectives
Teacher preparation programs (TPPs) have a responsibility to help pre-service teachers (PSTs) develop practices that can provide students with meaningful educational experiences. However, it is challenging to determine whether TPPs have met this goal because few programs use valid and reliable tools for measuring PST’s burgeoning skills (NCATE, 2010). Supervisor ratings during clinical placements are the most common approach used to inform programmatic decisions (Diez, 2010; Goldhaber, 2019). However, these ratings are subject to substantial rater bias, making it difficult to make valid inferences about whether PSTs have cultivated quality practices (Bartanen & Kwok, 2020). This study explores two alternative methods of assessing PST development in their enactment of quality practices. Our research questions are: 1) what is the relationship between the two approaches to measuring PST practices? and 2) what are the affordances and constraints of each?
Methods
We use two observational measures of teaching: 1) Classroom Assessment Scoring System (CLASS) scores (Pianta & Hamre, 2009) from clinical placements, and 2) researcher-designed rubrics for assessing instructional practice in a pair of standardized simulation scenarios focused on classroom management and quality feedback. In these simulations, PSTs practice teaching virtual students, providing opportunities to “approximate” teaching practice (Author et al., 2009) and serving as a standardized assessment platform. We use multivariate regressions to estimate changes in PST practice over time, as well as to measure the relationship between CLASS and simulation scores.
Results
We find a small but significant relationship between average scores on the Instructional Support CLASS domain and scores on the feedback simulation (Table 1). However, we find no relationship between scores on the Classroom Organization CLASS domain and those on the classroom management simulation. Instead, we find a significant negative relationship between simulation scores and the Behavior Management CLASS dimension (Table 2).
While CLASS provides a broad picture of classroom interactions, there are multiple challenges to interpreting CLASS scores as valid measurements of changes in PST practice over time. In particular, this approach fails to account for the influence of classroom and mentor teacher characteristics on PST observation scores, making it difficult to isolate growth over time. Additionally, given CLASS’s broad focus, PSTs likely don’t make large enough changes in their practice to differentiate true change from measurement error. The simulation-based measures capture a much narrower slice of PST practice, helping TPPs focus on honing specific skills. In addition, their standardized nature allows for insights about PSTs that are not confounded by the content they are teaching, the students in their classrooms, or the mentors with whom they work (Campbell & Ronfeldt, 2018).
Significance
Employing validated frameworks, such as CLASS, with trained raters who do not know the PSTs they rate, offers one way to combat the challenges of current measurement approaches in teacher preparation. Simulation-based measurements are another promising alternative. However, more work is needed to develop additional simulation-based measures to capture the full spectrum of quality practices and how measures influence shifts in classroom practice.

Authors

• Arielle Boguslav, University of Virginia

• Julie Jackson Cohen, University of Virginia