Changes to the probation / good standing policy

For a number of years, we’ve been collecting data on the efficacy of Honors probation as a tool to help students recover from a bad semester, or a bad class.  This fall was the first time we had a full set of data since the current 3.3 minimum GPA was instituted in Fall of 2013.

We’ve long had anecdotal evidence that the first semester of a students’ first year in college was a weak predictor for their overall success–in fact, the probation policy was originally designed to help students who first semester was uncharacteristic of their overall academic record. But we until this fall, we didn’t have a robust enough dataset to make any policy decisions.

I started the investigation by trying to find how well a students’ GPA at the end of ‘n’ semesters correlated with their overall success.  To begin, we defined ‘overall success’ as a combination of the total semesters the student completed at Ferris and their final cumulative GPA, plus a ‘bonus’ for any degrees earned.  After a few different models, I determined that simply adding the semesters and GPA was the best measure, so long as we compensated somehow for the students in graduate school, whose cumulative GPAs were lower than their counterparts in undergraduate.

The following charts show the distribution of students in ‘Good Standing’ or not in ‘Good standing’ who go on to be successful at completion.  The first one shows what it would be like if we measure ‘good standing’ after 1 semester (the current practice), 2 semesters or 3 semesters.

After checking the correlations between the cumulative GPA after 1 semester, 2 semesters and 3 semesters, I determined that higher the number of semesters, the greater the correlation with the ‘success’ rating.

Well, Duh. That was stupid– guess what, if we wait until semester 8 to check, we get a correlation of 1 with the final GPA.

So I chucked that approach and started over.

This time, I started with students who were successful according to the measure I previously described.  I decided, somewhat arbitrarily, that to be ‘successful’, one needed to be in the top three quartiles of the ‘success’ measure.  So then the question becomes: at which point in their careers, can we confidently say that 75% of our students will end up being successful?

This heat map plots semesters enrolled (x) versus cumulative GPA.  Each box represents the percentage of ‘successful’ (i.e. top three quartiles at graduation) students with that GPA or higher at that point in their career.

HeatMap-GooStanding-75Pct

As you can tell, the first semester is a bit of a mess—it simply does not fit the pattern of the rest of the data.  You may notice that semester 11 has a similar blip. That is the first semester in graduate school for those who go on.

So this indicates that the anecdotal evidence we’ve collected is basically correct: the first semester GPA is not a strong indicator of overall success.  But interestingly, the point on the 2nd semester chart at which 75% of students go on to be successful is a GPA of 3.3.

We’ve presented this data to the Dean and Provost, and both have agreed to let us pilot a change to the probational policy this year, whereby the requirement to maintain the minimum GPA of 3.3 will begin after the first year for first time in any college (FTIAC) students, rather than the first semester.  We will continue to offer support services to students who have rough first semester, but it will not count as a ‘probation.’

The Honors council will keep an eye on the practice this year, and if everything works as we expect it to, we’ll forward a recommendation to the Provost to make it the formal policy of the program.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s