How much could we realistically improve retention?

While we consider a variety of measures to assess our educational effectiveness, we focus on our retention rate (the proportion of full-time first year students who return for a second year) for some pretty crucial reasons.  First, it’s a legitimate proxy for the quality of our educational and socially-inclusive environment.  Second, as a tuition-dependent institution every student we lose represents lost revenue; and there is real truth to the old adage that it costs more to recruit students than it does to retain them.  So every year we calculate our retention rate, hold it up next to the last five or ten years-worth of numbers and ask ourselves:

Did we do a good job of retaining students?

Most of the time, we end up telling ourselves that our retention rate falls somewhere between “decent” and “pretty good” – especially considering all of the things we can’t control.  But this conversation always leads us to the next question; one that is substantially more difficult to answer:

What should our retention rate be?

And that is where people in charge start to daydream and folks in the trenches start to cringe.  Because it’s all too common for a small group of folks – or even one folk – to arbitrarily decide on the institution’s goal for 1st-to-2nd year retention without any sense of whether or not that number is a reasonable goal.  And there’s nothing more corrosive to an educational organization’s long-term quality than assigning an unrealistic goal to the people you depend on to accomplish it.  So over the last few months, I’ve been wondering how we could get closer to figuring out what Augustana’s ideal retention rate should be.  I don’t know if I have an answer yet – or if there really is a right answer – but I’d like to share some numbers and consider their implications.

Since research on retention suggests that a primary predictor of student success is a student’s incoming academic ability or preparation, it seems reasonable to use our students’ ACT score as a starting point to test whether or not we could realistically expect to improve our retention rate.  If most of the students that we lose are also those who enter with low ACT scores, it suggests that the students we lose depart because they are academically unprepared and it’s therefore more likely that we’re already pretty close to our optimum retention rate.  However, if most of the students we lose enter with ACT scores comparable to our average freshman ACT score, then it’s likely that we still have room to improve.  And if this latter possibility proves to be so, we could consider a few additional factors and come closer to identifying a “ceiling” retention rate from which we could begin to choose a plausible goal.

To begin this process, we took the two most recent cohorts for which we can calculated retention rates (2010 and 2011) and broke down the students who departed before the beginning of their second year by incoming ACT scores.  The table below shows the number of students in each of three different categories – the bottom quartile (<22), the middle 50% (22-28), and the top quartile (>28) – that departed before the second year.


<22 ACT

22 – 28 ACT

>28 ACT









Clearly, in both of these cohorts the majority of the students who left entered with ACT scores in the middle 50% rather than the bottom quarter.  Thus, to the degree that ACT score is a proxy for pre-college academic preparation, it appears that there might be some room for us to realistically improve our 1st-to-2nd year retention rate.

However, ACT score doesn’t necessarily reflect the degree to which a student has the personality traits and personal habits (persistence, time management, motivation, etc.) to succeed in college.  And there are plenty of students who enter with low ACT scores and thrive at Augustana.  So another way to explore this data is to consider the number of students who left in good academic standing.  Even though good academic standing at Augustana is a 2.0, in an effort to be conservative in this analysis, I set the bar at a GPA of 2.5.

From the 2010 cohort, 48 of the students who left departed with a GPA above a 2.5.  From the 2011 cohort, 58 students fit into this category.  Again, both of these numbers suggest some degree of opportunity for improvement.  I emphasize caution here because there are many reasons why students depart that are beyond our control (health issues, financial exigency, or family emergencies).  In addition, some students leave for non-academic reasons that aren’t accounted for in this rudimentary analysis.  So we would be wise to estimate a number substantially below the 48 or 58 students noted above.

Where does that leave us?  Well, I would suggest that a reasonable starting point would be to build out from the 2010 cohort.  As it stands, our retention rate with that group was 87.6% – the highest on record.  If we assume that, with some combination of improved programming , advising, and student support, half of those 48 students could have been retained, that means that we could estimate an additional 24 students – or an increase of about 3 percentage points in our retention rate.  That would put us at an optimum retention rate – a best possible scenario – of between 90% and 91%.

How does that compare to colleges like to us?  A 90% retention rate would be significantly higher than colleges like Augustana that enroll a similarly student profile.  What kind of financial investment would this require?  Although that is an even more difficult question to answer, the comprehensive effort necessary to improve our relatively strong retention rate would not be free and would likely require some tradeoffs.

Two final thoughts stick out in my mind.  First, while we might have some room to improve, I’d suggest that the we aren’t that far away from our optimum rate.  Second, since there are as many moving parts in this equation as there are students at risk of departure, effective change may result from subtle shifts in institutional culture just as much as it might be influenced by a new program or policy.

So can we improve our average retention rate? Probably.  Will it be easy?  Probably not.  Is it the right thing to do?  Of course.  But we had better not assume that we will see a surge in revenue even if we are successful.

Make it a good day,



Leave a Reply

Your email address will not be published. Required fields are marked *

Spam Protection by WP-SpamFree