The term e-Learning Analytics (e-LA) refers to the set of techniques aimed to extract useful information from existing online education datasets.
The final goals of E-Learning Analytics fall within one the following categories:
- Educational: Targeting to improve online education impact and student’s performance, such us:
- Reducing students’ drop outs
- Improving students’ understanding and learning
- Deciding which content is relevant for a given user
- Improving training materials
- Enhancing tutoring capabilities
- Business: Targeting to improve the return of investment (ROI) of educational initiatives, such us:
- Helping on marketing courses among the right target audience
- Reducing tutoring costs
E-Learning analytics comprise the use of data science techniques over data coming from multiple sources. In a typical online education environment, the most common data sources are:
- Learning Management System (LMS) activity records describing the users’ interaction with the online platform and training content.
- Learning Management System (LMS) performance records describing the users’ results over the proposed evaluation tests.
- User profile information, specially those characteristics that could impact the way students learn.
In environments where online content is used as a supporting tool and student interaction also happens outside the Learning Management System, teachers may collect data that could be used as an additional input to the analysis. In this case, educators’ engagement is key to make the overall process relevant.
Collected data is processed and analyzed, using advanced data science techniques, and a set of relevant analytics are obtained as a result. These analytics provide insights into the learning process, that should be used to perform whatever actions are required to meet the educational or business goals.
There are two main types of analytics:
- Descriptive: Provide insights about the past and allow to make decisions aimed to impact future learning processes.
- Predictive: Perform predictions about elements and variables that could impact ongoing learning processes. These type of analytics allow educators to take proactive actions.
When the volume of data to analyze is considered “big”, traditional data processing applications cannot deal with it in a reasonable time. Then, we have not only a problem of obtaining valuable insights from data, but also the challenge of processing this data at volume. This requires the use of Big Data computing techniques.
Sometimes, the terms analytics and big data are misused. It is true that obtaining analytics, most of time, requires processing important volumes of data, but it is not always the case.
When facing an e-Learning analytics project, we should have in mind the model defined by Dr. Mohamed Amid at Aachem University. This model describes the most relevant questions you should ask yourself in order to set the ground for a successful project:
- What? What kind of data does the system gather, manage, and is available for the analysis?
- Who? Who is targeted by the analysis?
- Why? What are our objectives (educational and business)?
- How? How does the system perform the analysis of the collected data?
On top of those, I would add:
- How much? Do we have the required budget to pursue the project?
- Success? Do we have a clear set of predefined success criteria? Are we in a position to measure the return of our investment (ROI)?
If you look at the e-Learning analytics related literature, you will also find the term “Educational Data Mining” (EDM). Many posts out there are trying to establish a distinction between Educational Data Mining and e-Learning Analytics. In my opinion, this distinction is, somehow, “artificial”. No matter how you call it, both try to understand educational data patterns in order to visualize them (descriptive) or to predict future outcomes (predictive).
In summary, e-Learning Analytics are about obtaining descriptive or predictive insights from online education data, using data science techniques and having a clear set of educational or business goals in mind.
The final objective of e-Learning Analytics is facilitating action that brings us closer to our goals.