The EdTech data analytics hierarchy: The fuel for powering intelligent experiences

Higher education leaders and those working in education technology (EdTech) often grapple with the topic of data analytics, and the industry continues to navigate how to gather, measure and intelligently analyze data to support decision-making within the institution. We have seen that data analytics has the potential to transform learner experiences when approached with a focus on creating actionable insights from reliable, privacy-compliant data. But to truly understand how to get the most out of EdTech data analytics, it is important to first consider what data analytics is and what it is not.

Data analytics refers to the act of sifting through high quantities of data to identify patterns and make inferences based on trends, but that isn’t the whole story. People tend to equate any type of data with the promise of effective decision-making and instant outputs; yet not all data analytics-powered technology functions are created equal. There must be additional steps taken to guarantee the usefulness of this information to support transformative outcomes. For example, no matter how elegantly prepared, basic reports that aggregate the results of a query are not intelligent resources in and of themselves.

Technology deployment in education has increased tremendously after the onset of the coronavirus pandemic and this has created a greater need for knowledge sharing and collaboration. This has resulted in the creation of a huge pile of learning data. 

The National Education Policy (NEP) 2020 emphasizes a few important areas in its quest for ensuring inclusive and equitable education, driving employability and focusing on learner experience, specifically:

1. Promoting learner well being
2. Developing energized and capable faculty
3. Setting up the Higher Education Commission of India (HECI) with 4 independent functions of regulation, accreditation, funding and academic standard setting

Each one of these areas will require data analytics to predict the need or the outcomes. For example, earlier identification of academic distress can lead to proactive academic advising interventions to promote well-being and help the learner get back on track toward their goals. 

The key now is to mine this precious data and use it for the benefit of the learner. This approach can be used further to improve the quality of teaching in India, while also giving an idea of the struggles faced by learners in a classroom.

For higher education, the best way to view data analytics is as a maturity model for which tools and features exist, with the end goal of using the data to positively impact learners. Leveraging this framework can assist organizations in distinguishing genuine data insight from reports that are packaged as raw numbers and nothing more. The stages represented in this hierarchy are best illustrated as follows:

In this format, data is the basis for creating meaning, which evolves from information into knowledge through context. However, knowledge and wisdom in the next stage are not enough to create an intelligence-driven experience. It is when analysis becomes part of the picture that the path to action becomes clear which will ideally create the desired impact for learners as identified on the left side of the pyramid. 

To better illustrate each of the stages, let us look at an example in action. Suppose you had a spreadsheet with information on learners' first-semester GPAs (stage 1). This report is only an informational output (stage 2) or knowledge artifact (stage 3) with no intended purpose. However, if you know that learners with a GPA below 2.5 are at a greater risk of dropping out, you might use the data to identify learners who need further support (stage 4).

With this information (stage 5), your academic support team may be automatically notified through triggering mechanisms to develop an outreach effort (stage 6). If you monitor how many learners with this GPA range continued after their first year, considering other variables that may have affected learner performance, you might infer that this outreach helped the majority of these learners successfully re-enroll (stage 7) or determine that additional support was needed for those that didn’t remain with the program.

When approached with these stages in mind, data analytics can be the fuel for powering intelligent experiences that lead to improved outcomes and support for learners, instructors and administrators.

Parents can be huge beneficiaries of the greater use of data analytics in education. Predictive modeling and analysis could be especially useful in India’s smaller towns and rural areas where parents are likely to have lesser insights into their child’s learning modes, if schools and colleges are able to use digital tools in a more consistent manner. If we can improve the performance of learners in this fashion, by putting data to good use, then we will be able to check dropout rates. Use of digital tools can also help learners identify their interests early enough and help them to channelize their efforts in that direction.

In summary, education institutions have lots of data around leaners and faculty. Insights gleaned from this data can help the institution improve operational effectiveness, quality of faculty and research and more importantly foster learner engagement (amongst many other things). Institutions which harness the power of data will be the education super-brands of the future.

The article has been written by Raj Mruthyunjayappa, President, India, Anthology Inc.