2024 ASEE Annual Conference & Exposition

This paper presents recommendations for engineering educators considering Social Network Analysis (SNA) to enhance their current research, particularly in large scale networks contexts. SNA is a growing method in the engineering education community, which allows researchers to represent interactions between students both quantitatively and visually through matrices and graphs. With these quantitative and visual representations, researchers can explore how interpersonal connections between students form, evolve, and relate to outcomes of interest. However, despite these virtues, there are several key issues that impact the interpretability and applicability of the data that SNA produces. For example, consolidating interaction data to usable forms and analyzing the results in a manner which accurately represents the underlying network can be extremely challenging. Additionally, these issues may be further complicated by the setting of the study such as how bounded the network should be, if interactions are online or face-to-face, and the temporal resolution of the sampled network. The purpose of this paper is to discuss these issues and disseminate recommendations from the experience of the authors designing and conducting a large scale (1000+ actor) SNA study at a large, public university in the United States over a study period of two years.
Specific points for recommendation in this paper begin with a description of the implications of the desired study network for collecting and consolidating interaction data. This description will also outline the benefits, drawbacks, and implications for online interaction data and face-to-face interaction data. Further, this paper discusses how the bounds of the network can inform data collection efforts and impact open-ended vs. close-ended responses. The benefits and drawbacks of these data collection techniques are also explored. Beyond collecting network data, consolidating networks into usable form presents a major hurdle for large scale studies. This paper presents practices for consolidating networks according to data collection types and resource requirements for each strategy. Finally, the analysis of social networks is an ever-growing field, and includes sophisticated methods which require a threshold of interaction data resolution and confidence. For this reason, this paper begins discussing analysis by presenting fundamental SNA methods and accessible resources for conducting them. The discussion of analysis methods concludes with a presentation of more sophisticated methods for analyzing network data while providing resources and requirements for implementing these methods.
As the importance of interpersonal networks on student outcomes, belonging, and other key aspects of education becomes more apparent, demand for methods which analyze interactions between individual students continues to grow. SNA is a powerful tool for engineering educators to understand interpersonal networks and has to date provided a wealth of understandings for engineering educators. This paper aims to make these methods more accessible to the engineering education community by sharing insights gleaned from our recent experiences.