You are here

Professional Opinion

Making sense of ‘big data’ in higher ed admissions

A deeper analysis of student behavior strengthens recruitment strategies
University Business, July 2016
Robert Miller is vice president for enrollment management of Centenary University.
Robert Miller is vice president for enrollment management of Centenary University.

If you work in enrollment management you’ve probably heard the term “big data.” Ubiquitous in the business world, big data is being adopted by higher education, particularly in the area of recruitment.

With stagnant or declining recruitment budgets and increased competition for students, leveraging the data most colleges capture is a cost-effective approach that can yield significant results.

Predictions offer guidance

It is important to provide a concise description of what “big data” means. Big data refers to high-volume, disparate information that requires powerful statistical programs to garner insights that aid in decision-making.

Enrollment managers collect a wealth of data from a variety of sources, including College Board and ACT demographic information, admissions office outreach, prospective student behavior captured in a CRM, and socio-economic data from financial aid filings.

Combine these traditional data sources with click-through rates on a website, number of Instagram posts and Facebook “likes,” and you can see exponential growth in the data points and possible interactions between them. This vast amount of data requires statistical techniques beyond traditional descriptive dashboard metrics.

Descriptive statistics are still useful, of course, and can help identify common characteristics within your prospective, admitted and ultimately enrolled student population. The amount of data available allows for much more powerful analyses.

Predictive analytics in particular allows enrollment managers to identify key variables in the likelihood of a particular student enrolling, allowing for a tailored recruitment strategy. Enrollment managers should look to “uplift modeling” that connects the institution’s actions to student behaviors that indicate an increased likelihood of enrollment.

Answering the key questions

Whether conducting the analysis in-house or contracting with an outside party, there are steps to be taken before any analysis can commence. An enrollment manager must understand what type of data is currently captured and identify what is most relevant now and down the road. This should be a collaboration with offices outside admissions, including the bursar, financial aid and student affairs.

To capture important data elements, some key questions must be answered, such as: How are the website analytics tracked? Who is responsible for capturing social media data? Is it in a usable format? Are counselors entering every student interaction into the CRM? Do potential residential students interact with student affairs and housing? Where is that information collected? Without proper procedures, you will end up with inaccurate data and hurt the integrity of the analysis.

In-house resource

The institutional research office is an important resource during the process. It can assist with the modeling if you choose to do it in-house using SAS (Statistical Analysis System), SPSS (Statistical Package for the Social Science) or a similar statistical tool. However, the office’s expertise can also assist in the data collection process and help various offices capture and code data more effectively.

The interpretation and utilization of the data is where many of these analyses fall short. The analysis will yield many tables, charts and statistical measures, and can be overwhelming. By working with the research office, enrollment managers can take time to dive deep and fully understand the findings.

Identifying the institutional activities that increase the likelihood of enrollment will allow you to enhance your recruitment strategies. You can, for example, alter your name-purchasing process and travel plans based on student characteristics that show a high propensity to enroll.

This type of analysis will provide a more complete understanding of your student population and the effectiveness of your recruitment activities. Continual improvement in the area of data collection can improve the effectiveness of future modeling initiatives.

Robert Miller is vice president for enrollment management of Centenary University.

Register now for UBTech 2018

Register now for UBTech 2018, June 4-6 at the Mirage, Las Vegas. At UBTech 2018, you’ll network with a dynamic community of higher ed leaders who are shaping the future of campus technology and explore topics like cyber security, distance learning, campus learning space design, communications, personalized learning and more. Your UBTech registration also includes a free pass to the InfoComm exhibit hall.

Register now>>