Author:
Dominik Froehlich | University of Vienna | Austria
Social network analysis and multiple and mixed method research are two important methodological trends in education and learning research (Hall & Preissle, 2015). The increased usage of combinations of both (e.g., Rienties & Hosein, 2015), what we call mixed methods approaches to social network analysis (MMSNA), has proliferated a plethora of approaches and designs. Since MMSNA are rather resource-intensive (Froehlich & Harwood, 2016), a guide for researchers is needed of how methods of social network analysis may be combined effectively and efficiently. Hence, we develop a map of methods used in MMSNA in the domain of education and learning based on published articles. This map informs future research about the potential of mixing and integrating multiple types of data and multiple types of methods of analysis. Furthermore, it is a tool that contributes to the objective of making the method more accessible also for researchers with more limited resources. This is done by moving away from typologies derived from theory that "have become almost too refined" (Bryman, 2006, p. 98) and staying close to research practice.
Data is generated through a systematic literature review across empirical education and learning journals published featuring MMSNA. For the selected articles, we code the methods used for data collection and analysis. The temporal order of the methods is then used to build a relational dataset. At the time of writing, the data collection and analysis is ongoing, the final dataset will be ready as of April 2017.
Quantitative social network analysis is used to analyze the relational dataset created. We analyze the network map that shows how different methods are linked with each other by extracting three centrality measures: indegree, outdegree, betweenness. Here, for instance, methods high in indegree may hint at a potential for preceding component preparation methods (Schoonenboom & Froehlich, 2016). Depicting a network graph and the network measures associated with it (e.g., measures of centrality) offer more versatile ways of finding and discussing patterns in previous research than typologies or mere counts of certain methods used (e.g., Bryman, 2006).
In conclusion, MMSNA is an important way of creating insight and moving the field of education and learning research forward. We provide a map of how MMSNA designs were implemented in previous empirical education and learning research. This can trigger an informed debate about what designs have proven to be useful, informs about the variety of potential designs, the ways of integrating qualitative and quantitative methods in social network research, and how to achieve the research goals given economic constraints. Furthermore, next to just reviewing what has been done and what not, the social network analysis used as a review tool allows the identification of new approaches to mixing quantitative and qualitative methods.