Chandra Orrill (UMass Dartmouth) with Yasemin Copur-Gencturk (Project PI – Univ of Southern California); Allan Cohen (Univ of Georgia); and Jonathan Templin (Univ of Iowa)
Current Research Assistants: Marty Epstein, Akira Harper, and Hamza Malik
Project Manager: Kym Welty
Past Research Assistants: Senai Sahle
National Science Foundation
The overall goal of this project is to pursue a potentially transformative approach to assessing teachers’ knowledge by developing a measure that is closely aligned with the content and skills taught in PD programs. This instrument will employ emerging psychometric models. Specifically, we aim to blend features of diagnostic classification models (DCMs) and statistical topic models (STMs) to provide informative feedback about teachers’ learning and to identify patterns in their knowledge growth. Using DCMs involves specifying the fundamental components of reasoning in a particular domain, and then constructing test questions systematically to correspond to reasoning with a different combination of those components. In contrast, STMs involve analyzing the patterns in responses toprovide insights into the reasoning teachers use in giving their answers. By developing open-ended items, we will create items that capture the key components of teachers’ content and pedagogical content knowledge in proportional reasoning, and then identify the characteristics of reasoning among teachers with different levels of content and pedagogical content knowledge.
Why Do We Do This Work?
Assessments can help drive high-quality instruction. However, the instruments currently available for support teachers’ learning of proportions are limited in various ways. We aim to use emerging psychometric tools to create feedback for instructors that can help guide instruction. We will make three significant contributions by combining two innovative psychometric models (Diagnostic Classification Models and Statistical Topic Models) with research on teacher knowledge and professional development. First, this assessment has the potential to provide PD organizers with useful information regarding which attributes of understanding teachers lack mastery in, as well as how to inform decisions about what to focus on in their programs and what particular attributes of understanding teachers have developed through PD. Second, this instrument has the potential to provide information about trends in teachers’ responses that will highlight possible characteristics of the thinking patterns of teachers with varying degrees of content and pedagogical content knowledge, thus allowing for more focused PD development. Third, this project promises to be of interest to both psychometricians and mathematics educators.