“The Feds are tipping off colleges?” Really?

August 30, 2015

In about a week, US News will release it’s all important college rankings to inform the public about colleges that lost or gained a tenth of a point and dropped or increased five spaces in the rankings as a result. I have written countless times about the absurdity of the arbitrary weighting of certain characteristics in the rankings, so I’ll not take time to do that here. But rankings fever and the competition it breeds among colleges (let alone giving incomplete short-hand information to students), contributes to ridiculous headlines in the wake of the decision by the Department of Education to hide from colleges’ view the list of colleges to which students are submitting the FAFSA.  The assumption here is that if colleges see their position on the list (number 1 through 10), this will influence an admission or financial aid decision.

On August 17, my friend Lynn O’Shaunessy penned a story that cbsnews.com titled, ” Feds to stop tipping off colleges on student choice.”
(http://www.cbsnews.com/news/feds-to-stop-tipping-off-colleges-on-student-choices/). While Lynn used the phrase “tipping off” in her story, her account was mostly balanced. The headline, however, implied widespread abuse of this information. Having served as the chief enrollment officer at three institutions over the past 27 years, I can say that abuse of “FAFSA position” is hardly widespread, as far as I know.

First of all, the vast majority of the nation’s 2400 non-profit private and public undergraduate universities are not selective. Perhaps no more than 5 percent of them admit fewer than half of the students who apply. Most cannot afford to deny admission or give less aid to a student who positions them low or high on the FAFSA list, because they need the student, no matter the odds of enrollment.

Secondly, schools with strong demand and strong applicant pools do not need to use their position on the FAFSA to impact a particular admission or aid decision. They have plenty of demand at just about any price point.

Over the past fifteen years, students have applied to more and more colleges because of perceived competitiveness and the need to have an affordable option. Colleges have aided this frenzy to be sure, in an effort to be more selective and to improve in the rankings, but mainly to assure that they can meet their enrollment goals.

During this time, predictive modeling has emerged to help enrollment leaders predict how many students (not individually whom) to admit at what level of financial aid in order to meet enrollment and net revenue goals. In some cases, if there is predictive value, FAFSA position may be used as one characteristic among many that — in the aggregate — can predict total enrollment. But an individual characteristic alone cannot predict behavior, and even taken in the aggregate, characteristics can only assign a probability of enrollment between zero and one for an individual.

When I say “characteristics,” I am referring to things students can impact (such as grades, test scores, campus visits, FAFSA position, leadership experience, strength of curriculum) and things that the student cannot control (gender, race, state or country of residence, socio-economic status).

To suggest that institutions make individual decisions based on their listed position on the FAFSA, and to remove this data because of a suspicion that a handful of colleges may be using the information inappropriately, would be like requiring students to visit campus under an alias, so schools would not know who is really interested. Interest, after all, predicts enrollment. Or since lower grades predict enrollment at a selective college, perhaps we admit all with low grades, since they are likely to enroll,  and deny those with high grades as they are least likely to enroll.  Absurd, right?

The point is, using a single characteristic to make an enrollment or aid decision raises not only ethical questions, but demonstrates a lack of understanding on how to use predictive models. These models should never be used for individual decisions — only to inform in the aggregate how many students can be admitted at what level of financial aid — in total — in order to reach our goals.

It is a shame that the Department of Education yielded to uninformed pressure to remove this data element from the FAFSA.  Of course, this is not the end of the world, as a college’s position on the FAFSA is just one element in a predictive model. But the implication that colleges are sinister in using this information against students is as absurd as it is insulting. My colleagues and I deserve better.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: