But the idea that struck me last evening has nothing to do with these. Rather, it focuses on the problem of how to more efficiently manage a university's enrollment while it tries to raise standards. We are in the middle of such a transformation right now. The usual route is to raise admissions standards. This seems like the obvious thing to do. But HS GPA and SAT/ACT are blunt instruments that explain less than half the variance in first year grades. For less selective institutions this is particularly acute because the high non-completion rate represents a huge waste of time and money for those students who don't finish. At the same time, it represents lost opportunity for the university, because those seats potentially could have been filled by others who tested lower but would have performed better. Finding better predictors is one approach, and we are developing non-cognitive instruments to do that.
However, there is another idea I'd like to share. What if an institution were to accept that predicting success is not all that effective, and adopts the "try and see" approach? This is how it might work:
- Adopt lower admissions standards, but still try to predict as best as possible
- Admit double the number of freshman you would actually want to continue
- Charge half price for the first year
- Advertise and implement high expectations in the classroom, and foster high quality teaching and support programs: give students every chance, but accept no compromises or excuses for non-performance
- Accept a 50% attrition rate for Freshman to Sophomore years. For those who are leaving, try to identify them early and have a Plan B outlined (community college perhaps)
- Maintain high standards throughout the rest of the curriculum.
- Advertise all of the above with a banner like: we give you the chance, make it affordable, and expect great things.
This would be a radical change, and would require a faculty that endorses the plan, upholds high standards, and constantly works to improve teaching and learning. The advantage of this approach is that you don't have to whittle down enrollment to increase standards--just the opposite.