Implementation of a Collaborative Data Use Model in a U.S. Context

• In Event: The Complexity of Data-Based Decision Making: Evidence From the United States, Norway, New Zealand, and Belgium

Abstract

Objectives
Over the past two decades, K-12 public school districts across the nation routinely use data to inform teachers’ pedagogy. However, the literature reveals teachers do not reap the full benefits from the practice. One of the reasons for this is the uneven implementation of models that guide their practice. This study is an examination of the factors that hindered or enabled the implementation of a successful European evidence-supported model for collaborative data use. Our two research questions are: What ought an evidence-based model look like? What does research suggest about how a model must be implemented to reap desired benefits?

Theoretical frameworks
Several concepts and theories support this study of one Texas elementary school’s implementation of the evidence-based model. Adult education (e.g., Merriam, 2002) supported the study because the collaborative nature in which adults learn is central to the practice of data use. Hallmarks of effective professional development, such as socially rich engagement (Guskey & Yoon, 2009; Wei et al., 2009), coherent design (Desimone et al., 2002; Wei et al., 2009), and appropriate duration (Garet, et al., 2001; Yoon et al., 2007) are also essential to the practice. Furthermore, the theory of co-construction of policy implementation (Datnow, 2006; Honig, 2006) allowed for the consideration of potential competing interests between the model implementation with school and district policies. Last, was the attention to the different contexts in which these models existed—the US and The Netherlands.

Methods and data sources
This is a qualitative study of one school in Texas. Data sources included audiotapes of ten 40-minute data team meetings during the 2017-2018 school year and ten memo-to-self recordings by the lead researcher that captured issues, concerns, and initial impressions of the meetings. Additionally, each participant in the experiment completed an assessment before and after the project and engaged in two interviews (midyear and end of the project). In all, over 450 pages of documentation were collected and analyzed over the course of the project. Other data sources included peer-reviewed journals and books on the practice.

Results
Analysis suggested factors influencing the fit of the data team implemented at the Texas school fell into two broad categories: General/team-oriented factors, and leadership factors. Two enabling and four hindering factors fell into the general/team-oriented category—meaning that they were less person-dependent/team-oriented category and more institutionalized (or at least dispersed throughout the system/team). Beyond such factors, the leader(s) of the campus personally seemed to engage in some actions during the process that pushed the data team process forward and, in a few cases, held it back. Three enabling and three hindering factors fell in this leadership-oriented category.

Scholarly significance
This study is significant because it revealed new ways to approach the practice. The first, is a bifurcated model for data use (quick data cycles for less complex issues, and slow data for more complex ones. Another approach is nesting early data team experiences within well-resourced, high-priority carrier initiatives for stronger implementation of the data use model in similarly situated US-based school contexts.

Authors

• Jo Beth Jimerson, Texas Christian University

• Vanessa B. Garry, University of Missouri - St. Louis

• Cindy Louise Poortman, University of Twente

• Kim Schildkamp, Universiteit Twente