Sustaining mathematics education research : a secondary data analysis framework

Abstract

Secondary data analysis (SDA) is the process of transforming and further analysing previously collected data or information for new research questions and/or with new techniques or theoretical models. The aim is to generate additional interpretations, conclusions or knowledge that are different from those reported in the main study. Whilst SDA within mathematics educational research is not new, it is underutilised within the field as a genuine form of research, since funded primary research projects are considered the benchmark in educational research. However, such funding is becoming increasingly difficult to source and government agencies are looking to leverage the funding they do provide for research purposes. As such there are opportunities for SDA to contribute to sustaining mathematics educational research into the future. SDA is considered a methodology. However, there are very few models that explicitly describe how to undertake SDA. This thesis advocates and applies a methodological, process-based framework that can be utilised for secondary data analysis in mathematics education across both quantitative and qualitative paradigms. The Knowledge Discovery in Databases (KDD) framework is a systematic, iterative and generative process that involves a series of sequential steps, each with corresponding decision-making components. This thesis utilised this framework across several different data sets to illustrate how the framework supported SDA. Across the course of this candidature, there was a need to broaden the original framework so that it encompassed a wider scope required for educational research. Hence, a modified version of the KDD framework was developed and is described within the thesis. This modified KDD (MKDD) framework was also used to analyse data from a secondary perspective. This thesis is presented as a series of six publications that advocates for the increased use of SDA in mathematics education research. Each of the publications adds to the narrative that SDA and the use of the KDD and MKDD frameworks are valuable to mathematics education. The first paper adds to the context of the study, describing the availability of data sets for secondary data analysis in mathematics education. The second paper describes the MKDD framework and presents an illustrative example of how the KDD framework was used to manage and analyse a qualitative data set. The third, fourth, fifth and sixth papers all show how the KDD and MKDD frameworks were used as a process-based approach to SDA that promoted a systematic way of organising and managing data sets for analysis. Each of these four application papers present new insights into mathematics education concerning the areas of spatial reasoning and the impact of digital technologies on mathematics engagement. An analytic architecture was developed as a mechanism for explaining the respective dimensions that were considered in terms of research manuscript development and the application of the KDD and MKDD frameworks. This architecture incorporates three elements, namely, (1) paradigm and methodological approach; (2) data source and context; and (3) data composition. Such an architecture provides researchers with an overarching understanding of how various data sets can be considered for SDA. Several implications for practice and policy emerged from the research. From a practice perspective, there needs to be an increased uptake of SDA through the MKDD framework in mathematics education. This would provide opportunities for existing data sets to highlight new theoretical and conceptual perspectives and lessen the burden on schools and education systems to support primary data collection. From policy perspective, undertaking SDA would increase the value proposition of the research funding provided to researchers, which may in turn, increase the number data sets made available for SDA in national repositories. If SDA is used more broadly, there will be less reliance on empirical data collection, more opportunities for early career researchers and HDR students, better use of government funds and new knowledge and insights from data that may have previously been rendered dormant once research projects are completed.

Date of Award	2019
Original language	English
Supervisor	Robyn Jorgensen (Supervisor) & Robert Fitzgerald (Supervisor)