I’ve spent the past six and a half years analyzing higher education data for more than sixty institutions, from large public universities to small private liberal arts colleges. The most common data challenge I find is institutions deleting data on students who were admitted, but ultimately decided not to enroll. I’ve found many institutions do well at recording any data they can get their hands on, but then preserve only the data for enrolled students. One common example is need-based aid, which often gets cancelled, without being archived, for non-enrolled students.
In the aggregate, your final enrollment, discount rate, diversity and SAT/ACT numbers don’t tell the whole story. It’s often important to dig deeper, analyzing results by subpopulation to better understand what worked and what didn’t. Frequency reports, cross-tabulation reports, financial aid summaries, and regression models are all powerful tools to help you understand what factors influence enrollment behavior. However, in order to properly use these tools, you need the data for both enrolled and non-enrolled students.
The best advice I can pass on to our readers is don’t delete your data... ever!
If something was important enough to record in the first place, it’s important enough to preserve forever. Everything doesn’t need to stay active in your “live” database, but before anything gets purged from the “live” database, export a copy of the data into an “offline” database for future analysis. As a researcher, I would love to have easy access to bi-weekly snapshots of historical data, but at a minimum I would recommend that institutions prioritize preserving an annual snapshot of their enrolled and non-enrolled data.
Remember disk space is cheap; it doesn’t cost you much to hold on to your non-enrolled data. You never know if it might come in handy down the road.
Image © iStockphoto.
About the author: Don Gray served as a Research Analyst and IT Manager for Scannell & Kurz from 2005 through April of 2013. His responsibilities included analyzing and reporting on client data, building predictive models, and maintaining the FAST and SKORE software environments.
Don has a B.A. in Mathematics from Cedarville University and is working on a Masters degree in Theology at Dallas Theological Seminary.
Connect with Don on LinkedIn.