Why Higher Education Institutions Are Failing With Data

Universities are falling behind with their use of data


One of the major talking points of the democratic primaries that appears to have been forgotten about during the presidential election proper has been the issue of tuition fees, with US students often leaving college in severe debt. At the lower end of the scale, public colleges charge on average roughly $9,000 per year, while at the upper end, Ivy League schools like Harvard demand as much as $60,000 a year for a degree.

This is, by anyone’s standards, a significant sum of money, and students rightly expect value for it. In order to provide this value, universities are increasingly turning to data analytics, with a KPMG study finding that 41% of colleges are engaged in predictive analytics and other recent surveys yielding similar results. While this number is rising though, it is still some way below the 62.5% adoption rate of predictive analytics in the private sector, and new research may provide some answers as to why.

In the 2016 Inside Higher Ed Survey of Faculty Attitudes on Technology, just 27% of faculty members and 34% of administrators said their data efforts have actually improved the quality of teaching and learning at their institutions. Similar numbers said the same about the impact on degree completion rates. Meanwhile, 65% of faculty members and 46% of administrators said their data initiatives had only been carried out to appease external groups, such as accreditors and politicians. This suggests one of three things: 1) they are not committing to data insights because they resent pressure from parties they see as undermining them and trying to infringe on their duties, 2) data does not provide any ROI in a higher education setting, or 3) data initiatives are not being properly carried out because universities are not approaching them in the correct way due to lack of knowledge or mismanagement.

The idea that faculty members would not want to do the best for their students is patently ridiculous, as is the concept that analytics simply cannot work in universities. It is also easily disproved by universities where it is working. Georgia State University, for example, is a shining example. They analyzed 2.5 million grades earned by students in courses over 10 years to build a list of warning signs a student may drop out before graduation. The system is updated daily and includes more than 700 red flags that tell advisors when a student is at risk. For example, in the case of declared political science majors, of those who get an A or B in their first political science class, 75% go on to graduate, while of those who get a C, just 25% do. As a result, graduation rates have gone up up six percentage points since 2013, while these degrees are being achieved an average of half a semester sooner than before - saving an estimated $12 million in tuition.

A more likely explanation for the perceived failure of data is that it is simply being mismanaged. In the Inside Higher Ed report, 54% of faculty said they don’t receive data gathered by their colleges meant to help them improve their teaching, while only 24% do. There also does not appear to be any consensus around how the data they are collecting should be used, with just 38% of faculty members saying meaningful discussions are taking place at their colleges about how assessment data should be used.

Colleges need to instil a culture whereby everyone is sharing data and working towards the same goals, with a Chief Data Officer likely a necessary appointment to ensure this. It is also important that there is an emphasis on action. Again, Georgia State is an example of how it is supposed to work. They hired 42 additional academic advisors alongside their data efforts to ensure that they actually acted on the insights they had garnered, arranging 51,000 in-person meetings between students and advisors over the space of a year.

This is the key principle of data analytics: it needs action. It is all very well investing huge sums on the technologies to gather and analyze data and find problems, but if these problems are not acted upon it is money thrown down the drain. And it appears many universities are willing to do this. Colleges now collect data across a range of touch points, including virtual learning management systems, library usage, attendance, and even personality metrics. Given the vast sums students are paying, it is inexcusable that all of this data is not being used effectively.

Vision small

Read next:

Big Data Forecasting In Pharma