Blackboard is a leading education technology company with 100 million users and 16,000 customers using its solutions across 90 countries. They’ve seen the adoption of mobile for learning becoming commonplace and have helped drive the debate on BYOD and big data in universities.
Higher education is just one of the many sectors that has heard talk of big data and wants to move ahead quickly to make the most of it. However, ‘doing’ big data is not what it’s about. If big data can be defined as more than one terabyte of data stored over many machines or billions to trillions of records of millions of people—all from different sources, is it even fair to call what educational institutes are attempting ‘big data’? No university can boast that much data but that doesn’t mean they can’t use analytics for unprecedented institutional and academic improvement. The key is how the data they have is interpreted, analysed, and how those insights are applied in strategic and operational decisions.
Big data is, by definition, complex. And, when it’s multi-sourced, it’s challenging to detect patterns that are meaningful. All universities monitor student activities; reporting the number of admissions, the applications accepted and the number of students who achieve academic qualifications. Yet, the annual reviews required by most Governments are ineffective when it comes to developing a good understanding of the bigger and smaller picture: The big picture of how successful a particular course is across all students, and the small picture of how an individual is coping day to day.
Evaluations performed at the end of the semester, module or course can highlight areas of concern but it may be too late to provide remedial support to individuals who have struggled to progress. Fortunately, there are universities who have realised that the data at their fingertips can provide a real-time view on students’ performances and likeliness to succeed and they don’t need to wait for the official assessments before they know about any teaching or learning issues.
The majority of teachers and professors rely on their own interaction with students to monitor their progress and recognise if a student is falling behind. But, with the growing diversity of students and increasing class sizes, they will struggle to stay up to speed on each individual’s success rate and react in time to provide steps to help. Technology can help and Learning Management Systems can provide an alert if assignments are late or attendance drops. However, this doesn’t offer the insight that could be achieved if the student data was analysed more discerningly.
The idea of initiating an analytics project can be daunting, but it is possible to start in a smaller yet effective way. First and foremost it is important to have clarity on what, how and why the data is being monitored. Student success, for example, is the goal of many analytics projects. Retention, progress, and academic achievement all determine success and universities need to know how success is defined for them, what are the data points, and how they will collect and analyse data.
It is worth remembering that regardless of how many data points a university holds about a student, they need to identify initially only those factors that have the most significant influence on student success.
Once the influencing factors have been decided, the next most important aspect is to have the data points for these available for the required period, without gaps. For example, if student attendance in class is an influencing factor, then the data should be stored and be accessible for every occurrence of the class. Or, if assessment grades are an influencing factor, then these grades should be stored and accessible for every assessment. In that way, from the first lesson and the first assignment to the last, the teacher and faculty will know if a student is on track to succeed. It’s also a way of highlighting any issues early enough to address them.
Data analytics is a science but big data doesn’t have to be ‘big’ to benefit universities and their student body. The key is to identify the relatively few influencing factors that are useful indicators of student progress and make use of that information to support their learning. So far, many universities who have tried data analytics are still working project by project and only a minority has initiated a plan that can deliver the scale, scope or frequency to make the results of any report meaningful. But we see that changing.
The objective is that data analytics will become the backbone of university administration and learning development. If universities can positively affect student retention and qualification success, this will help them attract good teaching staff, future students and research project sponsors. It’s more than worth the investment – even if the big data projects have to start quite small.
Dr Demetra Katsifli
Latest Posts:
Tags: Big Data, Education, Industries