Milwaukee Showcase

The Milwaukee Mathematics Partnership: A path model for evaluating teacher and student effects

Description

Alternative statistical models have been proposed for evaluating the effects of the Milwaukee Mathematics Partnership (MMP) on teachers and students. These models have been articulated in the form of hierarchical linear models that incorporate various combinations of student, teacher, classroom, and school-level variables. For example, one HLM approach would nest students within teachers, within schools to estimate the effect of MMP activities on student achievement.



Each attempt to estimate the impact of the MMP on teachers and students, however, share a common, underlying hypothesis: MMP activities, e.g., teacher professional development are impacting teacher mathematical knowledge for teaching (MKT), which is in turn impacting classroom practice, leading to improved student achievement. This paper steps away from the prior HLM analyses and explores these relationships using structural equation modeling techniques.



The path model proposed by this paper hypothesizes relationships between teacher education, teacher experience, professional development hours, mathematical knowledge for teaching (MKT), classroom practice, and student achievement (see Figure 1). We hypothesize that education, experience, and professional development hours are predictors of MKT and classroom practice. We further hypothesize that MKT and classroom practice are predictors of student achievement.



Given the exploratory nature of this work, the purpose of this paper is to present the results from an initial attempt to apply path analysis techniques to the problem of linking MMP activities to teacher performance and student achievement. We expect this work will be refined in the remaining years of the MMP evaluation. As such, in addition to presenting the results from this work, recommendations for improving the path model, as well as the implications of this work for evaluation are offered.