Longitudinal Analysis of How Professional Development Can Promote Changes in Teaching and Learning

• In Event: The National Writing Project: Networked Development and Use of Knowledge to Support Teachers’ Professional Learning

Abstract

Empirical evidence on how professional development (PD) can strengthen professional learning community and promote instructional change can provide direct guidance on supporting teachers’ learning. Our prior study found that beyond the direct effect on changing teachers’ instructional practices, participants in high-quality PD programs had spillover effects, indicated by the fact that participation in PD programs with high-quality features could promote the number of colleagues teachers helped with writing instruction (e.g. β=0.23) and the fact that colleagues’ instructional practices could be changed through help from PD participants (e.g., β=0.46; Authors, 2011). These high-quality PD features include sustained duration, broad content foci, and active learning strategies (Desimone et al., 2002; Garet et al., 2001; Penuel et al., 2007). In this study, we extend our prior work of three-year analysis to a four-year dynamic analysis of how collegial interactions will augment the direct PD effect on individual teachers’ instructional practices. This analysis will provide empirical evidence on the extent to which both PD and professional learning communities can serve as the infrastructure for teachers’ learning. Moreover, our new analysis will include students’ writing performance data to examine the effects of this infrastructure on student learning outcomes.

The proposed study will continue drawing on data from the larger experimental evaluation of the National Writing Project partnerships. Data were collected on roughly 2,000 teachers at four time points in 2007-08, 2008-09, 2009-10, and 2010-11 using surveys. In addition, in 2011, students’ performance data were collected through on-demand writing prompts and from local sites records. Data analysis includes two steps: 1) run growth models and year-fixed effect models to examine the possibly differential impacts of PD on teachers’ instructional practices across years; 2) run value-added models to gauge the extent to which the differences in teachers’ participation in both PD and their engagement in collegial interactions are associated with the variations in their students’ writing performance. These analyses will be performed within experimental conditions—that is, performing separate analysis for treatment and control groups—to separate the treatment effect from the PD effects, the focus of our study.

We anticipate that the differences in both direct and spillover effects of PD between treatment and control schools would increase across years, because teachers in treatment condition received more sustained content-focused PD that promotes peer collaboration than did teachers in the control condition. We also anticipate that within treatment schools, the direct PD impact in 2010-11 can be larger than that in prior years because more teachers had been exposed to longer duration of high-quality PD. Moreover, the effect of spillover effects in 2010-2011 would increase because more teachers would offer help to other colleagues and actively engage in professional learning communities, which increases the amount of expertise available to teachers in school networks and promotes supportive collegial norms (Bryk et al., 2010). In short, this analysis will expand our understanding of how PD should work in schools and how school mechanism can mediate the effects of external interventions on improving teaching and learning.

Authors

• Min Sun, Virginia Polytechnic Institute and State University

• H. Alix Gallagher, SRI International