The Strategic Value of Using Student-Related Data to Achieve Student Success
Student success in Higher Education has risen in prominence over the last 10 years. The growth of dedicated student success ‘centers’ in many universities (primarily in the US) is testament to this, as well as the increased focus by governmental organisations at national and regional levels on higher education’s ‘value for money’ and student ‘outcomes’. Long gone are the days when student success might be viewed simply as a function of the individual student – higher education institutions provided the resources, infrastructure and teaching leaving everything else up to students.
A major driver in this attitudinal sea change has been the worrying growth in student debt. The impact of student debt has been profound at a political level. It cost a leading UK politician (Deputy Prime Minister) his seat in Parliament in the 2016 General Election. In the US, Democratic politicians with their eyes set on the White House are jockeying for position to see who can offer the most appealing deal on reducing (or even removing) student debt.
While I cannot address the current level of student debt, what I can do is show how student success (i.e. successful student outcomes) can be improved radically. This, in turn, will reduce the likelihood of student debt accrual as a consequence of student dropout or extensions of periods of study. The key to this is staggeringly simple and we’ve had it at our fingertips for a very long time: vital student-related data. The challenge we have faced is knowing what to do with it or how to handle it effectively.
Predictive analytics in higher education has been around since at least the end of the 20th century. I was using it in a very generic way in the late 1980s and by the mid-2000s, I was using it in earnest to identify the characteristics of students who were failing or struggling badly with their studies. The data and the analyses were extremely powerful because it highlighted which groups of students were most likely to succeed and, conversely, which were more likely to drop out or fail to progress or graduate.
This information was used to produce recommendations on general and program-specific recruitment policies. Critically, it included key recommendations on the development of specific interventions to help clearly identified groups of students (i.e. those most likely to struggle) to achieve better outcomes. This was the predictive analysis of its day; extremely valuable for influencing the ways in which future cohorts of students would be supported through their studies. The strategic value of this information to the institution was obvious.
However, one huge downside was that, although it focused effort to support future cohorts of students, it did absolutely nothing for those students who had been in difficulty and those who had failed and/or dropped out during the academic year under review. In simple terms, the report lacked the critical time imperative to be of benefit to current students. There were two main reasons for this:
- Key data resided in a range of organisational silos outside student information system (in a range of teaching departments and/or support services) that were often difficult and sometimes virtually impossible to access;
- There was no means of processing the data such that the critical, timely interventions so essential to improving successful outcomes could be achieved.
The good news is that was then. Today we can overcome these major obstacles with relative ease. The application of SEAtS cutting-edge technologies means that data extraction and data processing can be achieved in virtually real-time. Better still, when combined with a range of other real-time data such as proxies for student engagement (including attendance at lectures/tutorials/laboratory classes, access to and use of the Virtual Learning Environment), there is a real step-change in our ability to help students. With the SEAtS Student Success Solutions platform in place, the critical conversations and interactions with students can take place within a timeframe that ensures that students receive the help they need when they need it most: now!
The application of SEAtS machine learning technology delivers another major leap forward in the application of these analytics. Using historic student outcomes and achievements, as well as current achievement outcomes and statistical data on real-time engagement, machine learning can reveal patterns of activity that contribute to success and, conversely, to failure or withdrawal. Lessons learned can then be applied to deliver routine analytics on individual students based on their record and activity profiles in real-time. This underpins further the crucial early interventions that ensure the appropriate help, advice, and guidance is provided to students when they need it.
As the image below illustrates, it is a very straightforward process when the 21st-century SEAtS Student Success technology is applied.
SEAtS Student Success Solutions deliver real added value to institutions through cloud-based solutions that minimise the impact on existing IT infrastructures. SEAtS can access multiple data sources seamlessly, overcoming organisational and data silo problems. Most importantly, SEAtS real-time processing delivers the critical real-time interventions customised to meet specific, institutional and student requirements for success.
In short, SEAtS will help you achieve a real student success win-win for everyone.
By Philip Henry
Philip is a former U.K. University Registrar and Secretary with almost 40 years’ experience in higher education in the UK and overseas.