In a recent post, I wrote about the Admissions Arms Race, and who had come out victorious. The short answer was "almost no one." I rolled up admission rates (percent of applicants admitted) and yield rates (percent of those offered admission who enroll) and showed them over time. These variables are pretty common parlance in college admissions; everyone with experience seems to know them. But I showed them only aggregated by type of institution; averages often mask details contained in them. To add some detail, I've now plotted them for every four-year, degree-granting institution that enrolls freshmen.
In that post I also introduced "Draw Rate," a term few had heard of. It's a simple calculation: You take the yield rate and divide it by the admit rate. So, for instance, Harvard, with a yield rate of about 84% and an admit rate of about 6% (2012) has a Draw Rate of about 14. Given that the industry average is about .6 (not six....point six), you see the market position of Harvard, even in comparison to some of its rivals: Princeton, Yale, and MIT, for instance, all of which hover around the still formidable 8 range.
The beauty of the draw rate is that it can't be fooled: If you get more selective just by generating fake or soft applications, your yield rate is going to go down. Try some numbers for yourself. Reasonable numbers, please, I don't like to argue with absurdity.
Over the last couple of decades, colleges have been pursuing prestige by attempting to get more selective. It's a good example of post hoc, ergo propter hoc thinking: Prestigious colleges are selective, so if we appear to be more selective, we'll become prestigious. (And parents engage in the same behavior when they see that successful people graduate from prestigious institutions, and therefore want a prestigious name on their child's diploma. They think the prestige caused the success, when it's often family success that generates the admission in the first place. Read Gladwell's paragraph on selection effects and treatment effects; it's in Section 3 of this article.)
See for yourself: Select public or private; a Carnegie type; a region, and then, if you want, a state within the region. I started with three years, but you can put in what you want.
As an aside, another thing I like about this is that is shows the problems with IPEDS data, such as missing information and obvious, erratic spikes up or down that suggest data errors. I use IPEDS data a lot and it can be very frustrating.
But mostly, it shows that there have been some winners over time. And they're mostly the ones who have been winning all along.
You can't market your way to the top in higher ed.
In that post I also introduced "Draw Rate," a term few had heard of. It's a simple calculation: You take the yield rate and divide it by the admit rate. So, for instance, Harvard, with a yield rate of about 84% and an admit rate of about 6% (2012) has a Draw Rate of about 14. Given that the industry average is about .6 (not six....point six), you see the market position of Harvard, even in comparison to some of its rivals: Princeton, Yale, and MIT, for instance, all of which hover around the still formidable 8 range.
The beauty of the draw rate is that it can't be fooled: If you get more selective just by generating fake or soft applications, your yield rate is going to go down. Try some numbers for yourself. Reasonable numbers, please, I don't like to argue with absurdity.
Over the last couple of decades, colleges have been pursuing prestige by attempting to get more selective. It's a good example of post hoc, ergo propter hoc thinking: Prestigious colleges are selective, so if we appear to be more selective, we'll become prestigious. (And parents engage in the same behavior when they see that successful people graduate from prestigious institutions, and therefore want a prestigious name on their child's diploma. They think the prestige caused the success, when it's often family success that generates the admission in the first place. Read Gladwell's paragraph on selection effects and treatment effects; it's in Section 3 of this article.)
See for yourself: Select public or private; a Carnegie type; a region, and then, if you want, a state within the region. I started with three years, but you can put in what you want.
As an aside, another thing I like about this is that is shows the problems with IPEDS data, such as missing information and obvious, erratic spikes up or down that suggest data errors. I use IPEDS data a lot and it can be very frustrating.
But mostly, it shows that there have been some winners over time. And they're mostly the ones who have been winning all along.
You can't market your way to the top in higher ed.
Comments
Post a Comment