Network Effects on Early Career Teachers' Use of Mathematics Assessment

• In Event: Studying Early Career Elementary Teachers’ Instruction at Scale: Factors Related to Planning and Teaching Mathematics

Abstract

Objective: This paper examines how early career elementary teachers’ school-based social networks affect their mathematics assessment practices in the context of state mathematics curricular standards.

Perspective: Many early career teachers (ECTs) acquire some knowledge of student assessment and develop initial mathematics assessment practices during pre-service preparation. But assessment often receives disproportionately little attention in teacher preparation (Darling-Hammond & Bransford, 2006) and it is difficult for ECTs to develop assessment skills through trial and error. Consequently, there is large, unexplained, variation in ECT’s math assessment practices. The theoretical framework underlying this study posits that ECTs can access knowledge and skills related to math assessment in their schools through their formal social networks (i.e., mentors, math instructional coaches) and their informal social networks (teacher colleagues).
This is consistent with findings that certain types of mentoring, collaboration with colleagues, and instructional coaching are associated with effective instructional and assessment practices (Coburn & Russell, 2008; Glazerman et al., 2010; Stanulis & Floden, 2009). Further, social network members may provide knowledge and support in exchange for compliance with school norms of teaching. Thus, our framework contends that ECTs’ assessments practices can be affected by the math assessment practices and expectations (i.e., instructional norms) of their formal mentors, colleagues, and instructional coaches.

Data Sources and Methods: In 2015-16 and 2016-17, the EMT project collected classroom observation and survey data from 150 teachers in grades K-5 who were in their first five years of teaching. We used the Teaching for Robust Understanding in Mathematics (TRU Math) classroom observation instrument (Schoenfeld, Floden et al., 2014) to observe each ECT four times; in addition, each ECT completed two surveys about their student assessment practices and their access to assessment-related support from formal mentors and other teacher colleagues and a survey about their mathematical knowledge for teaching (MKT) in number and operations. We also administered surveys to the mentors, colleagues, and instructional coaches who worked with the ECTs to learn about their practices, expectations, and MKT levels. We used multivariate regression and hierarchical linear models to study ECTs’ enactment of mathematics assessment practices as a function of several independent variables, including those representing social network effects.

Findings: Our findings indicate that ECTs were especially likely to increase their use mathematics assessment practices (as measured by dimension 4 of TRU Math) when their mentors and colleagues incorporated assessment into their own teaching. The effects were especially pronounced in the final episodes of a lesson, providing a key link between ECTs’ learning and the transition between lessons.

Significance: This study is significant for three reasons. First, it integrates a cognitive component (focused on teachers’ MKT) and a sociological component (focused on norms in their social context) in order to explicate factors that influence ECTs’ math assessment practices. Second, our study used social network analysis to explore the effects of access to knowledge and instructional norms on ECTs’ math assessment practices. Third, our findings have key implications for how schools and districts organize mathematics professional development for their teachers, including ECTs.

Authors

• Kenneth A. Frank, Michigan State University

• Kristen Bieda, Michigan State University

• Serena Jean Salloum, Ball State University

• Yincheng Ye, Michigan State University

• Sihua Hu, Northwestern University

• Peter A. Youngs, University of Virginia