Higher Ed Data Sources

[polly_mer]

Over the years, people have expressed their appreciation for my writing lengthy posts reporting data on the state of higher ed. 

I'm sharing the wealth by starting a thread where we can put links to data sources so we'll all have them in one convenient place.  I'm still going to write lengthy posts using data on relevant threads; this is more like an annotated webography stored in public instead of on my computer.


First up: Payscale's Return on Investment (ROI) for attending specific colleges and universities at https://www.payscale.com/college-roi.

Drawbacks:

* Money isn't everything and many successful people in life aren't maximizing on profit
* No breakout by majors
* Payscale is often self-reported data, although this information is from IPEDS and other nationally reported data, which is subject to other biases like first-time, full-time students as the only people who count in graduation rates.


Information provided:

* By sorting on different categories and applying various filters, one can see which colleges are likely to be struggling and why.  For example, a very low graduation rate combined with a negative 20-year ROI bodes poorly in all directions.  However, a very low graduation rate at an average of 6-years-to-degree combined with a strongly positive 20-year ROI may mean a focused institution doing great things with struggling students.  It could mean other things like a huge transfer population with a handful of full-time, first-time students to track through IPEDS.

* One can filter the elite institutions to look more at the middle.  Again, this might be useful to job seekers in terms of either going to a place that is investing in a particular mission (e.g., high grant money, even with lower graduation rates, and a higher loan value) or picking out places that have a possible disconnect between what they claim their mission is and what the numbers indicate their achievements are.

[ciao_yall]

Yes, education pays off, and for the broader population. This chart hasn't changed much in many years.

https://www.bls.gov/emp/chart-unemployment-earnings-education.htm

And if you ask "But what about some schmuck who got a BA and is still a schmuck?"

I will answer "I would imagine that schmuck is still better off with a degree than without one."

[polly_mer]

College Scorecard (https://collegescorecard.ed.gov) puts some useful IPEDS data into a friendlier format for comparing institutions.  Some of the categories of data suffer from the widely known drawbacks of following only first-time, full-time student cohorts and only having financial data from those who graduated with a student loan balance.

Again, one can find outliers who are being measured on the wrong metrics (e.g., those valiant institutions who admit the toughest cases, allow them to attend part-time, and consider successful completion of courses more realistic than degrees in 4 to 6 years), but one can also differentiate between institutions that are healthy and institutions with many red flags.

I like the ease of investigating up to 10 institutions side-by-side for a variety of data categories.  IPEDS has far more categories and depth, but for a quick overview, I like this interface.

Since I'm here: the main IPEDS data use page is https://nces.ed.gov/ipeds/use-the-data .

[mamselle]

Wasn't Octo or someone working on this quite awhile back?

Trying to recall...

M.

[polly_mer]

Wasn't Octo or someone working on this quite awhile back?

Trying to recall...

M.

If you find something, then let me know.  I can remember some fabulous one-time posts by a variety of people (Spork, I know you're one of them), but I don't remember a dedicated thread.

[spork]

Jon Boeckenstedt's work at Higher Ed Data Stories is incredibly informative.

[polly_mer]

Jon Boeckenstedt's work at Higher Ed Data Stories is incredibly informative.

<stern look>

How long have you known about this website and refrained from sharing?

<big smile>

This is incredible; thank you for sharing on this thread.

[spork]

I've known about it for at least four years given that it was mentioned on this website back in 2015.

[polly_mer]

I've known about it for at least four years given that it was mentioned on this website back in 2015.

That's a really disturbing graph for the draw rate on the website.

[marshwiggle]

Did anybody other than me have to rearrange the algebra on draw rate?

D.R. = (enrolled)(applicants)/(admitted)^2

Then it's easier to see. If you're getting a higher number of applicants OR a higher number who actually enroll, compared to how many you admit, that's better.

[polly_mer]

Did anybody other than me have to rearrange the algebra on draw rate?

D.R. = (enrolled)(applicants)/(admitted)^2

Then it's easier to see. If you're getting a higher number of applicants OR a higher number who actually enroll, compared to how many you admit, that's better.

Shhh!  I spent too many hours arguing against wasting time and energy trying to increase the number of applicants we couldn't possibly admit or who would not enroll with us to be happy about just doing the math in isolation.


Attempting to game the system by having a good draw ratio that doesn't result in a higher enrollment along with a much larger percentage of that enrollment being full (or nearly full) pay is the kind of thinking that accelerates the demise of smaller institutions.

A possibly useful article for perspective on selective/non-selective admissions is https://www.pewresearch.org/fact-tank/2019/04/09/a-majority-of-u-s-colleges-admit-most-students-who-apply/

[Hibush]

That Pew report is good to see.

In terms of selectivity, the effect among schools that admit 20% or less is so small that one can hardly make it out. The very large shift is for schools that used to admit 70 to 90% of applicants now admitting 40 to 70%.

Even that shift is probably due in large part to the increase in applications per applicant.


One graph captioned says that the number of admission has increased, but not enough to keep pace with the "soaring number of applications." 
Of course, they shouldn't keep pace with the change in applications but the change in applicants. Since the number of applications per applicant has increased a lot that is a significant difference. Their measure was an increase from four to seven applications per enrolled student, which ignores the few who never enrolled.

I'd like to see the total number of applicants, how many were admitted nowhere, how many were admitted somewhere but did not enroll, and how many were admitted somewhere and enrolled. Basically taking the whole of the 4-year nonprofit higher-ed world as one big unit.

[polly_mer]

https://www.educationdive.com/news/how-many-colleges-and-universities-have-closed-since-2016/539379/ has a good overview of why colleges are closing.

The non-profit-only list of college closures and merging is https://www.educationdive.com/news/tracker-college-and-university-closings-and-consolidation/539961/.  That list includes consolidations like the announced changes in Wisconsin.

[marshwiggle]

An interesting note from one of the articles referenced:

These issues were more pronounced for smaller private colleges, measured as those with less than 1,400 full-time enrollees.

Does anyone know whether this is mostly a US-specific phenomenon? In Canada, as far as I know, institutions that small would mostly be private religious organizations. I'm not sure there's any non-religious degree-granting place that would qualify. Do other countries have lots of tiny places like this?

[Parasaurolophus]

Does anyone know whether this is mostly a US-specific phenomenon? In Canada, as far as I know, institutions that small would mostly be private religious organizations. I'm not sure there's any non-religious degree-granting place that would qualify. Do other countries have lots of tiny places like this?

There are a few institutions that small or smaller, but not many. University of King's College springs to mind, as do a few of the small francophone universities like Université Sainte-Anne, Université de Saint Boniface, and IIRC the Université de Sudbury and Université de Hearst (unless we don't count these two because they're federated parts of Laurentian). First Nations University of Canada, too, as well as Royal Roads University and Université du Québec en Abitibi-Témiscamingue.