Hiding under the “average” blanket

Higher ed folks often toss around numbers that supposedly describe the quality of a given college or university.  But a funny thing happens on the road to an “average” score.  Although it might approximate everyone, it rarely describes anyone in particular.  So unless a college hires an Institutional Psychic to predict the individual path of each new student (The Nostradamus Endowed Chair of Student Success?), metrics like an average retention rate or a student-faculty ratio don’t tell us as much about a place as we might like – or want – to think.

But this doesn’t mean that the data is useless.  In fact, we can learn a lot about ourselves by looking for differences between subsets of students on a variety of such metrics.  For example, an overall retention rate could – and often does – mask stark differences in persistence between high and lower ability students, high and lower income students, or men and women.  Identifying the nature of those differences could point us toward the solutions that would help us improve what we do.

Over the last several years we’ve increasingly employed this approach to squeeze more useful information out of our student experience data.  Many of you have already seen the way that the experiences of your majors might differ from other Augustana students in your senior survey departmental reports.  Taking the same approach that we use to better understand student retention (dividing students by gender, race/ethnicity, socioeconomic status, and academic preparation/incoming ACT score) reveals a layer of nuance that I believe deepens our understanding of the Augustana experience across diverse students types.  It also helps us use evidence to think about how we might engage specific types of students in specific moments to more carefully mitigate these differences.

As an aside, the differences that we spend most time considering are those that cross a threshold of statistical significance – meaning that there is less than a 5% chance that the likelihood of the observed difference is coincidental (the formula that we used is called a t-test).  In this post I am going to focus on differences between low income and middle/upper income students.  Future posts will consider the differences the emerge across a range of variables including gender, race/ethnicity, and academic preparation.

Comparing low income students with middle/upper income students presents a great example of the complexities this kind of analysis can provide.  We used Pell Grant eligibility as the marker of lower income – its an easy way to categorize financial need, even as it probably over-simplifies the impact of socioeconomic status (SES).  As you look through the items on which differences emerged, think about the possible factors that might produce a statistically significance difference between the two groups’ responses.

Lower income students scored higher than middle/upper income students on several items.

  • My co-curricular activities provided numerous opportunities to interact with students who differed from me in race/ethnicity, sexual orientation, religious beliefs, or social/political values.
  • My out-of-class experiences have helped me connect what I learned in the classroom with real life events
  • In your non-major courses, about how often were you asked to put together ideas or concepts from different courses when completing assignments and during class discussions?
  • My interactions with the librarians helped me improve my approach to researching a topic.
  • Augustana faculty and staff welcomed student input on institutional policy and handbook decisions.
  • When you had questions or concerns about financial issues, were the offices you contacted responsive and helpful?

Conversely, lower income students scored lower than middle/upper income students on one item.

  • My out-of-class experiences have helped me develop a deeper understanding of myself.

The scope of these differences is fascinating.  In some instances low income students’ responses seem comparatively more positive than the rest of the student body.  While this might suggest that some of our efforts are indeed providing a compensatory impact, I think these findings highlight the relative lack of pre-college opportunity that lower income students often must overcome (fewer communal resources like libraries or access to technology, less exposure to some of the ideas fundamental to the liberal arts, etc.).  In other cases, these findings might be evidence of the quality of our effort to be sensitive and inclusive to these students (e.g., the relatively more positive interactions for low income students when asking for help with financial issues).  Understanding the nature of these differences could play an important role in shaping our daily interactions with students who may, unbeknownst to us, come from a lower socioeconomic background.

At the same time, sometimes these differences in responses suggest that some of these students’ experiences are less positive.  Given the small numbers of lower income students at Augustana, it seems likely that they would recognize the extent to which they interact across socioeconomic difference more often than middle/upper income students.  In some cases this might contribute to a sense of marginalization for low income students.  Finally, the difference in responses on the question about “out-of-class experiences develop a deeper understanding of myself” is particularly intriguing.  I’d like to know more about the underlying factors that might influence that difference.

Taken together, these findings replicate the results of many recent studies regarding the impact of social class on college students – an impact that extends far beyond financial constraints.  What have you observed in your interactions with low income students?  Are there things you have done that seem to help these students succeed at Augustana?  As you interact with students this week, I hope these findings expand your sense of the ways in which our students experience Augustana differently, and how our sensitivity to these differences can improve our educational impact.

Make it a good day,

Mark