EPJ Data Science Highlight - Academic performance and behavioral patterns

Photo by Nathan Dumlao on Unsplash

In an article just published in EPJ Data Science, Valentin Kassarnig, Sune Lehmann and Andreas Bjerre-Nielsen look into smartphone data of undergraduate students to assess factors influencing social behavior and educational performance.

(Guest post by Valentin Kassarnig, Sune Lehmann & Andreas Bjerre-Nielsen, originally published on the SpringerOpen blog)

Everyone has one, everyone uses them: smartphones. In the last decade, technology has fundamentally changed the way we communicate, socialize and interact with each other. But what are the consequences of these changes? How is our behavior affected? What role does the social environment play? An interesting venue to explore these questions is educational research, where the goal is to understand which factors influence academic performance – and how.

In order to shed light on those questions we investigated data collected from 538 students’ smartphones. As part of a larger project (and with explicit consent from everyone involved), we recorded a rich dataset of social behavior: Bluetooth scans, call and text message meta-data, Facebook activity logs, and mobility traces. These data were complemented by personality tests filled by the participants, as well as and administrative data. (More information about the larger project and the data collection process can be found here.)

In our new study, we analyze this dataset in order to shed light on how different aspects of behavior impact the students’ academic performance. We place a special focus on the students’ social environment which was captured across five different channels: calls, text-messages, physical proximity (from Bluetooth scans), Facebook interactions, and Facebook friendships. These ways of interacting are arguably some of the most important channels of communication.

Our analyses revealed that the mean grade point average (GPA) of peers is highly correlated with academic performance. That is, students with high performing friends are very likely to also perform well. Although this effect was consistently observed across all investigated channels, it was most pronounced for calls and text messages, which are considered to be proxies for strong social ties. When we divided students into three groups according to their GPAs we could observe that the dominant fraction of text messages was exchanged among members of the same group:

Distribution of exchanged text messages among students of different performance groups.

So, if you’re not getting good grades maybe it’s because you have the wrong friends?

Unfortunately, things are not that simple. Whether these effects are driven by selection (people select friend that are similar to themselves) or adaption/peer-effects (people grow to be more similar to their friends) still remains an open question.

Social environment strongly influences academic success

However, due to our rich dataset, we were able to compare the importance of various features on academic performance (features about personality, behavior and social network). In our models, the network factors had consistently higher explanatory power than individual factors, underlining the importance of the social environment for academic success.

Among the considered individual effects, class attendance and the personality traits “conscientiousness” and “self-esteem” were found to have significant impact on academic performance.

When we performed a thorough analysis of attendance behavior, we once again found a strong social component, as attendance levels among friends tend to be highly correlated. (A detailed discussion of that topic can be found here.)

Finally, we achieved the highest explanatory power when combining all individual and network features. In a supervised learning experiment with the full feature set we achieved state-of-the-art performance without using any information on past performance. Usually past performance is the most important feature in predicting academic success, so that result is quite remarkable.

In summary, our findings emphasize that academic performance depends to a considerable extent on personality, behavior, and social environment. We hope that our work will inspire further studies of the impact of the social environment on academic success, as well as the interplay of individual and network factors.

This was our first experience of publishing with EPJ Web of Conferences. We contacted the publisher in the middle of September, just one month prior to the Conference, but everything went through smoothly. We have had published MNPS Proceedings with different publishers in the past, and would like to tell that the EPJ Web of Conferences team was probably the best, very quick, helpful and interactive. Typically, we were getting responses from EPJ Web of Conferences team within less than an hour and have had help at every production stage.
We are very thankful to Solange Guenot, Web of Conferences Publishing Editor, and Isabelle Houlbert, Web of Conferences Production Editor, for their support. These ladies are top-level professionals, who made a great contribution to the success of this issue. We are fully satisfied with the publication of the Conference Proceedings and are looking forward to further cooperation. The publication was very fast, easy and of high quality. My colleagues and I strongly recommend EPJ Web of Conferences to anyone, who is interested in quick high-quality publication of conference proceedings.

On behalf of the Organizing and Program Committees and Editorial Team of MNPS-2019, Dr. Alexey B. Nadykto, Moscow State Technological University “STANKIN”, Moscow, Russia. EPJ Web of Conferences vol. 224 (2019)

ISSN: 2100-014X (Electronic Edition)

© EDP Sciences