Distance learning has become a common mode of education in the pandemic era, but how can educators and researchers make the most of the data generated by online platforms and tools? In this article, we will explore the concepts and applications of educational data mining and visualization, and how they can help improve the quality and effectiveness of distance learning.
Educational data mining (EDM) is the process of analyzing large and complex datasets from various sources of educational activity, such as learner interactions, assessments, feedback, and content. EDM aims to discover meaningful patterns and insights that can inform educational decisions and interventions, such as curriculum design, personalized learning, student engagement, and learning outcomes. EDM uses various methods and techniques from data science, such as machine learning, statistics, natural language processing, and network analysis.
In my experience, this type of multi-layered analysis that merges both quantitative and qualitative data is invaluable in helping faculty advance higher levels of student success. EDM provides a more holistic picture of what works and what is not as effective, especially for community college faculty who teach in open door institutions that do not use selective admissions criteria. Designing courses that will effectively engage diverse learners at various levels of preparedness is complex. EDM can provide critical analysis to guide course design that addresses the diversity of learners and informs both structure and content to successfully engage more learners.
Educational data visualization (EDV) is the process of presenting and communicating the results of EDM in a visual and interactive way, such as graphs, charts, maps, dashboards, and animations. EDV helps to make the data more accessible and understandable for different audiences and purposes, such as teachers, students, researchers, and policymakers. EDV can also facilitate exploration and discovery of new patterns and relationships in the data, as well as comparison and evaluation of different scenarios and interventions.
One example of the importance of EDV is the use of dashboards to provide comparisons of both leading and lagging success metrics. Dashboards provide visuals of current data and also allow you to compare data over time, creating quick insight into what interventions work. At our college we also use EDV through dashboards to help us compare traditional versus online course success and to see what clases are working best in an online format. Once we know the online courses with the highest success rates we can dig into why those classes are doing better and use those findings to improve the course design for courses that do not have high success levels.
Word clouds are also powerful in terms of taking qualitative data and transforming it to quickly show trends. You can immediately see what was said must often and quickly dive into that data. The real power of EDV is that it enables us to quickly synthesize data to make better, informed, decisions. It's absolutely critical that EDV is leveraged across platforms and research areas to ensure that people can take away meaningful insights from the data collected. One thing to note about accessibility: although EDV enables a larger group to interpret large sets of data quickly, it may not be accessible to all individuals. It's important to still use captions, proper labels, and alt text (alternative text) to ensure greater accessibility.
EDM and EDV can support distance learning in a variety of ways, such as monitoring and measuring student progress, performance, and engagement in online courses and activities, as well as identifying and predicting student needs, preferences, and difficulties. These technologies can also be used to analyze and improve the quality and effectiveness of online content, pedagogy, and assessment. Additionally, EDM and EDV can be used to explore the social and emotional aspects of online learning, such as collaboration, motivation, and satisfaction. Finally, they can be employed to generate hypotheses and research questions about online learning phenomena and outcomes, contributing to the advancement of educational theory and practice.
Actually I agree that EDV can assist students that require a variety of modalities to understand data. This is especially true in adult education where students need to see their progress visually represented to help them track their own progress in their learning journey. This aids in retention of students as it builds persistence in learners.
I agree. I believe the greatest benefit of EDM and EDV for online learning success is realized when the people analyzing and using the data understand the importance of a culture of curiosity where data provides the basis for more questions. EDM and EDV can help faculty and course designers easily see which online classes are not as successful as traditional seated class options and then they can delve deeper into the data to understand why. The key however is a healthy culture where faculty do not fear their data but rather see data as critical for continuous innovation.
Distance learning data research is a growing and evolving field that offers ample opportunities and challenges for EDM and EDV. Currently, there is a need to develop and apply new methods and techniques for data collection, analysis, and visualization, such as multimodal data, learning analytics, artificial intelligence, and augmented reality. Ethical, legal, and social issues related to data privacy, security, ownership, consent, and bias must also be addressed to ensure transparency, accountability, and fairness of data practices and policies. Additionally, collaboration and communication among different stakeholders involved in distance learning data research should be enhanced to foster a culture of data literacy and inquiry. Lastly, it is important to bridge the gap between research and practice to ensure the relevance, validity, and impact of data findings and recommendations for distance learning contexts and goals.
In my experience, transparency and accountability are the two greatest hurdles that can become insurmountable barriers to creating a culture of inquiry or curiosity. Great care must be used to ensure transparent data sharing is positively received as a starting point for achieving greater student success versus being negatively perceived as punitive. Additionally, simply sharing data without the expectation that the data will inform practice and be used for continuous innovation is a waste of time and effort.
The push for data aggregation and visualization will ultimately move distance learning research forward. The gap that exist across institutions (educational and corporate) is simply having systems that lend themselves to extracting meaningful metrics. It's easy to get lost in data, and expecting individuals (faculty, administrators, employees) to extract large data sets and use tools like Python or R (which have a steep learning curve) to visualize data isn't scalable. Instead, institutions need to prioritize getting systems/tools that EDV features -- or, at least, push existing tools to do a better job of presenting data. Make sure you work with stakeholders across the organization to figure out what data they need to filter out the noise!
If you are interested in learning more about EDM and EDV for distance learning, there are a few steps you can take. Examples of EDM and EDV tools and resources for distance learning include MOOCdash, RISE, LearnSphere, and LAK conference. You can also consider enrolling in online courses or workshops on EDM and EDV through DataCamp, Coursera, or EdX. Joining online communities and networks of EDM and EDV practitioners and researchers, such as EDM Society, SoLAR, or ISTE is another option. Finally, you can start your own EDM and EDV project or experiment with your own distance learning data using open source or free software and platforms like R, Python, Tableau, or Power BI.