Content area
Full Text
The purpose of this article is to utilize an analysis of African American students' retention rates at a regional public university in the Midwest to discuss enrollment management and retention behaviors as they pertain to academic success and failure.
Many colleges and universities build elaborate tuition discounting models to enhance enrollment; some also build student success models to drive the post-enrollment strategic enrollment management (sem) process. When these models fail to lead to the desired outcomes and/or modeled results, the tendency is to attribute blame to external events that "couldn't possibly have been controlled for." It is easy to adopt the role of armchair quarterback. Rare is a rigorous quantitative analysis of models and their impact on our programs.
Alternatively, when our strategies prove successful and/or our models perform as anticipated, we tend to be quick to take credit. A better strategy would be to conduct a post-mortem examination to ensure that credit is deserved. Predicting the probability of student success is extremely difficult-particularly at the individual student level. Thus, caution must be exercised regardless of the failure or success of our enrollment strategies.
The university that serves as the focus of this discussion is the same one that Davis and Burgher (2013) discussed in a previous article. The institution has in place a rather aggressive sem process; utilizes data extensively; and is focused on retention and graduation. African American students' retention rates varied substantially across the classes of 2009, 2010, and 2011. Specifically, a significant decrease in retention from 2009 to 2010 was followed by an 8.5 percent increase in 2011. During this time, many new programs were initiated, and recruiting efforts were modified. Even while administrators were celebrating the increase, managers had to decide which programs were responsible for the increase and thus were deserving of continued funding. Was this result random? Was it a result of programmatic decision making (if so, then which programs should continue to be funded?)? Or had the input (/.<?., class demographics) been altered so as to affect change ?
Regardless of whether their intended outcomes are attained, college and university administrators need to determine which effects are causal. Determining the importance [i.e., the impact) ofvarious independent variables is not an easy process; even the best models have...