Background on Data Mining and Strategic Enrollment Management in Colleges and Universities

According to SAS (an organization that produces analytical software), data mining is “an iterative process of selecting, exploring and modeling large amounts of data to identify meaningful, logical patterns and relationships among key variables” (Patel et al., 2010). Strategic Enrollment Management (SEM) is being used at colleges and universities to meet their enrollment goals. According to Lingrell (2012), data and information are what are needed for the “strategic” aspect of SEM.

In traditional data analysis, summary statistics of a group are commonly the result: averages, percentages, etc. In data mining, information about an individual is sought. Chang (2006) says that data mining can identify hidden patterns in data and allow predictions to be made at the individual level, which is what higher education institutions desire to do. Recruiters can then use this information during the recruitment cycle to identify which individuals are more likely to apply and eventually enroll at their institution.

Two data mining processes that are currently being used are SEMMA and CRISP-DM. A description of these two processes follows, along with an example of how each has been used in a university setting.