Accountable for Improving Education for All? Tensions in Local Practices of Data Use in Norwegian Municipalities and Schools

• In Event: Enacting Accountabilities in Education (Part 2): Its Effects on Teachers' Work and on the Teaching Profession

Abstract

Objectives
Since national standardized tests were introduced in the Norwegian school system in 2004, superintendents and school leaders are expected to use students’ performance data to improve education on various levels of the school system. This paper investigates the extent to which local educational authorities and school leaders in Norwegian municipalities use such performance data to inform practice, and in what ways practices of data use are linked to improvement efforts, control of schools and accountability mechanisms. Based on the findings, the paper discusses possible limitations of using student performance data which is linked to design and function of the tests. core tasks of the different levels of the local school system, and broader educational goals in the Norwegian context.

Theoretical perspectives
Student performance data, is often considered to be efficient, standardized, uniform, and intuitive measures productive for usage in a range of processes for the development of education, teaching and learning (Porter, 1995). On the other hand, the very same attributes can lead to exaggerated expectations of what can be achieved on the basis of data, and simplification of complex education processes - consequently what can be described as the alluring attributes of data and data use that might mask important aspects, knowledge and nuances in education important for productive developments and for providing high quality education for all students (Authors, 2017).

Methods
Online surveys were administered to superintendents and school leaders in 2017 in one fourth of the municipalities in Norway (stratified sample according to size and geographical location). Various quantitative techniques were applied to analyze the data, such as confirmatory analysis to create latent variables, then structural equation modeling (SEM) is used to examine the data structures. The second step of the analysis includes multilevel analysis which makes it possible to investigate the structural relations between variables both within and between groups of actors (Heck & Thomas, 2009). Moreover, content analysis was applied to the open-ended responses.

Data sources
The analysis draws mainly on data provided by standardized questionnaires, but also qualitative responses provided by the superintendents (response rate: 37%) and school leaders (response rate: 49%).

Findings
Findings show that most of the school leaders use national test data to monitor school quality over time, develop strategies for improvement and benchmark the school with other schools in the municipality. Control and performative accountability are more evident in data use practices on municipal level. School leaders experience tensions between competing ideals; balancing a focus on performance and excellence as well as equal opportunities for all.

Scientific contribution
This paper contributes to generating knowledge on local practices of data use among key actors, such as superintendents and school leaders and discusses consequences which can be linked to competing policy agendas.

Author

• Guri Skedsmo, University of Oslo