Collaboration Analytics explained
Given the growing amount of data on learning, analysis of learning inform learners and educators to make better decisions. Collaboration analytics is a promising subfield of learning analytics focused specifically on collaboration. In this article we want to introduce the concept of collaboration analytics and illustrate its application on examples mostly from the perspective of education, however the principles may be applied in different settings too.
What is Collaboration Analytics
Collaboration analytics is measuring, collecting, and analysing data from groups of learners.
Getting insights of how group of students works together on a shared goal is collaboration analytics. Analysing online activity of students forming collaborations to solve specific problems is collaboration analytics. Developing processes and interfaces to present the gained insights for teachers or students to help them effectively is also a part of collaboration analytics.
What we mean by collaboration shapes what might be analysed. Collaboration varies on different scales from small groups all the way to huge communities, organisations and social networks.
In network perspectives, collaboration may be more about building and maintaining relevant relationships with shared goals. In specific problem solving perspectives, collaboration means specific interactions in solving the problem ahead and so on.
Who we consider to be collaborator is important alike. Although technology often acts as a middleman, environment or a tool, it is increasingly closer the role of our partner.
Collaboration analytics seek to
- support individual students as well as “the group of students” in collaboration,
- support educators in providing feedback to students and set up learning environment and design,
- support researchers in forming new theories and gain new understanding of group communication.
The insight gained by analysing data might give suggestions
- to students such as whether to change their collaboration strategies,
- to educators such as which students groups collaborate effectively or which students are at risk of low collaboration and social presence,
- to researchers such as what conditions help collaboration and vice-versa. (Martinez-Maldonado et al., 2021)
The insights may play different role such as
- monitoring — providing ways to see the important aspects on current state of collaboration, mostly for educators and researchers.
- reflective — mirroring the current state of collaboration to students so they may decide how to change their strategies,
- normative — providing students the guidance and next-steps to meet desired state,
- formative — providing students the guidance while considering their characteristics.
Meeting the normative and formative roles is particularly difficult as it assumes the systems infer next steps to meet desired state of cooperation while being highly context-aware to adjust based on collaboration dynamics, its users and groups differences (Schneider et al., 2021).
Examples
We cherry-picked specific examples of collaboration analytics from research to real-world applications and implementations based on their illustrative potential.
Developing automatic analysis of Face-to-Face interaction
In their paper Lamsa et. al. (2021) study the use of machine learning algorithms to analyse cooperation within 11 groups of students during one programming task a the course. The analysis was to classify the various phases of inquiry-based learning, such as orientation, conceptualisation or research during group interaction. Regardless of the specific results of the study, this enables us to peek into the options and considerable difficulty of automating collaboration analysis. They recorded the interaction of each group and then transcribed their communication to get the data for the analysis. The classification models were based on deep neural networks, a popular machine learning technique, enhanced of two layers. The first, word-embedding layer obtains textual features that capture the semantic and syntactic relationships of the words. The second layer is an attention layer that extracts substantial parts of the words sequences. The human classification was used as a benchmark for various configurations of classification models.
Reflecting group collaboration
An example of collaboration analytics implementation that focuses on student monitoring and reflection is an application called BLINC. The application is designed with an aim to support student collaboration literacy. It allows students to visualise the characteristics of their collaboration regards the key dimensions based on audio recording of the collaboration. They identified the dimensions of collaboration such as the climate, communication, context or contribution by mixed methods survey study made on 131 university students specifically asking about their collaboration experiences. As we can see, this application focuses on the listed features of discussion so it is very general and less specific regards to collaboration in concrete type of problems (Worsley et al., 2021).
Helping students at risk
The network monitoring perspective on collaboration is present in the study of Dawson (2010) who analysed differences of high and low performing students networks sizes and interaction with teachers. The analysed network was based on online forum discussion interactions with total 1026 students and staff of first-level chemistry course at Canadian University. Besides finding that higher-performing students according to their course grade had bigger networks and connection with teachers, the study mentions the opportunity of network visualisation for teacher interventions.
Seeing workplace collaboration
In the similar paradigm as in the previous example, companies that aim on more general audience than education introduce analytics dashboards to support healthy workplace collaboration. Tools such as Microsoft Viva Insights or Webex People insights are designed to provide insights to both managers with overview on organisations groups activity and individuals by showing highly personalised insights to them.
Conclusion
Generally speaking collaboration allows for solving bigger problems at the sake of bigger friction. Collaboration analytics promises to help us collaborate better by leveraging all the possible data. As we can see, there are examples of analytics helping or exploring to help individual students and groups, teachers and researchers looking at collaboration from different perspectives. For more detailed but definitely not fully comprehensive view on the concept of CA, we recommend you to read the paper ‘What Do You Mean by Collaboration Analytics? A Conceptual Model’ (see Martinez-Maldonado et al., 2021). They also list cases of CA applied on collaborative problem solving which we didn’t included.
References
Dawson, S. (2010). ‘Seeing’ the learning community: An exploration of the development of a resource for monitoring online student networking: Monitoring online student networking. British Journal of Educational Technology, 41(5), 736–752. https://doi.org/10.1111/j.1467-8535.2009.00970.x
Lämsä, J., Uribe, P., Jiménez, A., Caballero, D., Hämäläinen, R., & Araya, R. (2021). Deep Networks for Collaboration Analytics: Promoting Automatic Analysis of Face-to-Face Interaction in the Context of Inquiry-Based Learning. Journal of Learning Analytics, 8(1), 113–125. https://doi.org/10.18608/jla.2021.7118
Martinez-Maldonado, R., Gašević, D., Echeverria, V., Fernandez Nieto, G., Swiecki, Z., & Buckingham Shum, S. (2021). What Do You Mean by Collaboration Analytics? A Conceptual Model. Journal of Learning Analytics, 8(1), 126–153. https://doi.org/10.18608/jla.2021.7227
Microsoft. (2022). Collaboration — Microsoft Viva Insights Docs. https://docs.microsoft.com/en-us/viva/insights/personal/use/collaboration
Schneider, B., Dowell, N., & Thompson, K. (2021). Collaboration Analytics — Current State and Potential Futures. Journal of Learning Analytics, 8(1), 1–12. https://doi.org/10.18608/jla.2021.7447
Simon Kerrigan, S. K., Shihui Feng, S. F., Rupa Vuthaluru, R. V., Dirk Ifenthaler, D. I., & David Gibson, D. G. (2019). NETWORK ANALYTICS OF COLLABORATIVE PROBLEM-SOLVING. Proceedings of the 16th International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2019), 43–50. https://doi.org/10.33965/celda2019_201911L006
Worsley, M., Anderson, K., Melo, N., & Jang, J. (2021). Designing Analytics for Collaboration Literacy and Student Empowerment. Journal of Learning Analytics, 8(1), 30–48. https://doi.org/10.18608/jla.2021.7242
