Data use for what and for whom?: a close look at the policies and teacher practices that shape data-driven decision making at an elementary school

In this thesis, I present an analysis of how educators practiced data-driven decision making (DDDM) at Greenbrook Elementary, a school that has historically struggled to facilitate equitable learning outcomes for students. The findings from this study suggest DDDM may look quite different in practice than how it is described in the literature. The literature often characterizes DDDM as a practice where teachers are active participants who systematically collect data, interpret these data, formulate action plans, and continuously evaluate and adjust their plans based on further data (Coburn & Turner, 2011; Mandinach, 2012). At Greenbrook, this was not typically what was observed. Teachers were often passive recipients of data and directives on how to interpret and use these data. In this context, I offer an examination of how particular policies and practices mediated teacher data-use. Specifically, I present three essays on practices or policies associated with DDDM at Greenbrook. In the first essay, I describe teachers’ engagement with color-coded students’ performance data. While the color-coding of data are meant to support teachers’ interpretations of data (Love, 2004; Marsh 2012), I argue that at Greenbrook, students’ color-coded data was primarily used to sort students into different educational offerings. In the second essay, I examine aims for data use. Borrowing the concept of “matchmaking” (Oakes & Guiton, 1995), I describe how educators’ data use targeted matching students to pre-determined educational programs. I argue that matchmaking promoted particular data-use conversations and decisions while stifling inquiries into other issues that might have merited attention, like inequities in the learning environment. In the third and final essay, I present an overview of the political mandates that governed teachers’ work at Greenbrook. I argue that teachers had little autonomy to respond to students’ data in meaningful ways