Endowments of Top 25 Research Universities

I already posted the figures for top 25 liberal arts colleges, so I might as well post these as well.

USNWR	School	Endowment	Enroll	End/Stu	5%/stu	Rank
1	Harvard	$31,728,080	19,627	$1,616,553	$80,828	3
1	Princeton	$17,109,508	7,802	$2,192,964	$109,648	1
3	Yale	$19,374,000	11,701	$1,655,756	$82,788	2
4	Columbia	$7,789,578	22,283	$349,575	$17,479	15
5	Chicago	$6,575,126	12781	$514,445	$25,722	10
5	CIT	$1,772,369	2,175	$814,882	$40,744	6
5	MIT	$9,712,628	10,566	$919,234	$45,962	4
5	Penn	$6,582,029	19,842	$331,722	$16,586	16
5	Stanford	$16,502,606	19,535	$844,771	$42,239	5
10	Duke	$5,747,377	14,983	$383,593	$19,180	12
11	Dartmouth	$3,413,406	6,141	$555,839	$27,792	8
12	Northwestern	$7,182,745	19,389	$370,455	$18,523	14
13	Johns Hopkins	$2,598,467	21,092	$123,197	$6,160	21
14	Wash U	$5,280,143	13,820	$382,065	$19,103	13
15	Brown	$2,496,926	8,695	$287,168	$14,358	17
15	Cornell	$5,059,406	20,939	$241,626	$12,081	19
17	Rice	$4,451,452	5,879	$757,178	$37,859	7
17	Vanderbilt	$3,414,514	12,714	$268,563	$13,428	18
19	Notre Dame	$6,259,598	11,992	$521,981	$26,099	9
20	Emory	$5,400,367	13,381	$403,585	$20,179	11
22	Georgetown	$1,160,291	16,871	$68,774	$3,439	24
23	Carnegie Mellon	$1,017,338	11,618	$87,566	$4,378	23
23	USC	$3,517,173	36,986	$95,095	$4,755	22
25	Wake Forest	$1,058,250	7,162	$147,759	$7,388	20

The six columns above are the school's US News & World Report ranking, the school name, the endowment size (in thousands), the school's enrollment (as reported by USNWR), the endowment per student, a ratio of 5% of the endowment per student (which is roughly how much a school has to spend per student solely from tapping into the endowment), and the rank (within this sample) of the endowment per student.

I dropped public schools because I wasn't sure which endowments listed were for one particular campus versus which ones were for a system of schools.  As expected, there's a strong correlation between endowment size and ranking (-.68 between endowment and ranking and -.70 between per pupil endowment and ranking).

Notable overachievers include Penn and Columbia, which both have longstanding reputations of excellence given their placement in both the Ivy League and large, historically prominent American cities, and Johns Hopkins, which has some very prominent programs in medicine and hard sciences.

Notable underachievers include Rice, which is located in Houston, which has only recently become one of the largest cities in the nation, and Notre Dame, which has strong ties with the Catholic church.

Endowments of Top 25 Liberal Arts Colleges

The release of the latest endowment data piqued my curiosity.  One can look at them a million ways, but I'm particularly interested in liberal arts colleges and decided to compare the endowment figures to US News & World Report rankings.

As you can see in the table below, the higher-ranked schools generally have both larger endowments and more endowment money per student (correlation of -.53 between endowment and USNWR rank and -.58 between endowment/student and UWNWR rank), which is, of course, not a surprise (both because they money allows schools to afford items that increase prestige and because the endowment itself is used by UWNWR as one component of their rankings).

Before I get into the discussion, let me display the data.  In the six columns are the USWNR ranking, the school name, the endowment size (in thousands), the school's enrollment (as reported by USNWR), the endowment per student, a ratio of 5% of the endowment per student (which is roughly how much a school has to spend per student solely from tapping into the endowment), and the rank (within this sample) of the endowment per student.

USNWR	School	Endowment	Enroll	End/Stu	5%/stu	Rank
1	Williams	$1,784,305	2,083	$856,603	$42,830	5
2	Amherst	$1,641,511	1,795	$914,491	$45,725	3
3	Swarthmore	$1,508,483	1,524	$989,818	$49,491	2
4	Pomona	$1,700,454	1,560	$1,090,035	$54,502	1
5	Middlebury	$907,668	2,532	$358,479	$17,924	12
6	Bowdoin	$904,215	1,762	$513,175	$25,659	8
6	Carleton	$653,465	2,020	$323,498	$16,175	17
6	Wellesley	$1,499,872	2,411	$622,095	$31,105	6
9	Claremont McKenna	$543,236	1,278	$425,067	$21,253	10
10	Haverford	$402,730	1,177	$342,167	$17,108	14
11	Davidson	$509,583	1,742	$292,528	$14,626	20
12	Washington & Lee	$1,218,132	2,173	$560,576	$28,029	7
12	Wesleyan	$589,811	3,215	$183,456	$9,173	23
14	Vassar	$814,130	2,446	$332,841	$16,642	16
17	Hamilton	$657,529	1,861	$353,320	$17,666	13
18	Harvey Mudd	$243,125	773	$314,521	$15,726	19
19	Grinnell	$1,500,219	1,655	$906,477	$45,324	4
19	Smith	$1,429,527	3,113	$459,212	$22,961	9
21	Bates	$231,451	1,725	$134,174	$6,709	24
21	Colby	$611,441	1,825	$335,036	$16,752	15
21	Colgate	$693,436	2,903	$238,869	$11,943	21
24	Oberlin	$699,895	2,974	$235,338	$11,767	22
25	Bryn Mawr	$671,103	1,755	$382,395	$19,120	11
25	Macalester	$654,465	2,033	$321,921	$16,096	18

I should first note that I dropped the US Naval and Military Academies (which tied for 14th) from the sample -- no endowment data was available, and I'm not convinced they belong in this group anyway.

In no particular order, here are the things that stand out to me:

1.) As I said above, the wealthiest tend to be ranked the highest.  No surprise.  Five institutions in particular far out-distance the rest in terms of resources per student, four of which are ranked 1-4.

2.) Given that all of these schools are in the same general arena in terms of prestige, there's quite a bit of variability in terms of financial resources -- two schools have less than $10K per student to spend while five have over $40K.

3.) Notable underachievers (high resources, low ranking) include Grinnell, Smith, Washington & Lee, and (arguably) Wellesley.  Two of these are single-sex institutions, which may limit applications (raising acceptance rate) and/or increase alumni giving relative to other factors.  Washington & Lee is one of only two schools located in the South and has a reputation as being more conservative than the other schools.  I'm not sure how to explain Grinnell other than possibly the Iowa location.  As I mentioned above, exactly five schools have more than $40K to spend per student (the same five that have endowments exceeding $1.5 billion), and four of those rank 1-4.  Grinnell ranks 19th.

4.) Notable overachievers include Carleton, Davidson, Wesleyan, and Bates.  I'm not sure I have a great explanation for these other than the possibility that their reputation is larger than their endowment (an argument I don't find super compelling).  More on this below.

5.) An interesting factoid: Macalester's endowment is exactly $1 million more than Carleton's.  Macalester is 45 miles north of Carleton, enrolls 13 more students, and is 19 spots behind in the rankings.


Were I to do more than muse about summary statistics, I'd wonder why some schools earn much higher ratings relative to their financial resources than others.  To the extent that a given school's goal is to enroll the best students and earn the highest ranking, some are drastically more successful/efficient than others.

Of course, I'd expect that a large part of the explanation is that some of these schools place far more emphasis on the factors USNWR measures and/or the USNWR rankings themselves.  I'd also imagine that, among other things, allocation of these resources, admissions practices, location, reputation, campus offerings, and -- in various ways -- performance of campus leaders, faculty, and staff influence the success/efficiency of a school (as measured by USWNR relative to endowment resources).

Lastly, I'd wonder if the endowment, say, 20 years ago might be more strongly correlated with the ranking than the current endowment.  Or at least if some the low endowment/high rank schools had higher endowments relative to their peers 20 years ago and some of the high endowment/low rank schools had lower endowments relative to their peers 20 years ago.  This certainly would've influenced the faculty they hired 20 years ago, the perception of the college by students' parents, and probably the perception of the school by people in academia, at foundations, working as college counselors, etc.  In other words, schools that have dramatically expanded recently may be thought of as more resource-rich than they currently are and schools that have seen recent dramatic growth in endowment may be thought of as more resource-poor than they current are.

I'm no expert on higher ed or philanthropy, so I can really only guess about these numbers.  All I know is that I find them interesting and that there are probably some good explanations for the patterns we see.  If anybody knows more about these topics than I, I'd love to hear your explanations.

Update: I neglected to note that along with allocation of resources, I'm sure the allocation of the endowment matters as well -- if  large chunk of the endowment is restricted to, say, maintaining the hockey rink or seminary that it will affect the school differently than if a large chunk is restricted to, say, financial aid or even unrestricted uses.

I also neglected to mention that not only do some schools emphasize the ratings and try to game the system, but some might even just flat out lie.

More on The Wrong Question About Teacher Pay

In my last post I suggested that the NY Times was asking the wrong question about teacher pay.  They start the discussion by asserting that teachers make more than other professionals with similar SAT and GRE scores and ask if that means if teachers are overpaid or that teaching needs to pay more to attract higher achieving applicants.

To me, it raises a different question: why people would work in another field if they can make more by teaching? I can think of two broad sets of possibilities.

1.) As an anonymous commenter mentioned on the last post, it's possible that people underestimate how much teachers make and simply don't realize they could make more by teaching.  If this is the case, then we should spend our time shouting from the rooftops about how much teachers actually get paid.

2.) Or, people would rather do something else than teach despite the fact they earn less money.  This could be true for any number of reasons (working conditions, barriers to entry, advancement possibilities, salary structure, etc.).  If this is the problem, we should our time figuring out how to make teaching a more attractive field for both prospective and current teachers.

In either case, the two solutions offered by the Times don't make much sense.  Arguing that teachers are overpaid at the same time that we've reached a consensus that we need more high-quality teachers seems counterproductive; if we're currently short on great teachers, I doubt that reducing pay will do much to help.  And if people are already willing to sacrifice higher salaries in order to enter a field other than teaching, I'm not sure why we think throwing more money at them would solve the problem.

Given the absurdly high turnover rates in our most troubled schools, and the research on why such turnover occurs, it seems likely that working conditions in some schools might be a large part of this problem.

In previous research, I found that 40% of teachers at high-poverty NYC middle schools are in their first or second year in the school and that fewer than half of all teachers have 5 or more years of teaching experience.  The starting salary for somebody fresh out of college in NYC is now over $45K, which won't make anybody a millionaire but is more than a lot of recent college grads earn.

Would NYC have better teachers who stuck around longer if they paid starting teachers $50K or $70K or $100K?  That's certainly a possibility, but besides potentially costing billions of dollars each year, it's also possible that there wouldn't be a huge difference . . . if teachers are leaving to take lower paying jobs elsewhere, then it seems dangerous to assume that paying them more will keep them in place.

Apologies for arguing by anecdote, but in my own case I've sacrificed somewhere near a quarter million dollars of pay to attend grad school for six years (compared to teaching in NYC) . . . and there's a really good chance that I would've made more next year as a 9th year teacher in NYC than I will as a newly minted PhD.  And not for a second does that make me regret leaving teaching.  Well, ok, maybe for one second . . . but that's about it.

Anyway, my general point is that we have plenty of evidence that things other than money are driving the current trends we see in the teacher labor market -- low numbers of high-achieving people entering the field and very high numbers of teachers leaving the field (though this attrition is concentrated mostly in high-poverty schools) -- so we should look at levers other than money when creating policy designed to attract and retain the best and brightest teachers.  The fact that people are passing up money to do things other than teach only serves to bolster this point.

The Wrong Question About Teacher Pay

The NY Times posted an online debate today that asked whether teachers are overpaid.  The following is their prompt, verbatim.

In the private sector, people with SAT and GRE scores comparable to those of education majors earn less than teachers do. Does that mean teachers are overpaid? Or that public schools should pay more to attract top applicants who tend to go into higher-paying professions?

I'd argue that this is the wrong question.  If current teachers earn more than similarly achieving (as measured by the SAT and GRE) private sector employees, that means they must earn the same as private sector employees who are higher achieving.

Given the recent focus on teacher quality, shouldn't the question then be why these higher-achieving graduates choose to do something other than teach despite earning no more than current teachers earn?

Moving on from the "Search for Universals"

As I've delved deeper into the field of education research, I've grown increasingly frustrated by two large problems I see in that research.  While listening to Malcom Gladwell's TED talk on consumer choice, something he said helped crystallize the second issue (I'll explore the first in another post).

In the talk, Gladwell discusses the shift of food science (starting with Prego pasta sauce) from trying to create the "one best" product for all to creating multiple products that fit specific groups.  He compares that trend to cancer research, arguing that "the great revolution in science of the last 10-15 years" is "the movement from the search for universals to the understanding of variability".

If he's right, then it seems ed policy researchers missed a memo somewhere (to be fair, it's not just ed policy -- I see the same problem in lots of policy literature).

The vast majority of the policy research I've read, seen presented, or heard discussed focuses exclusively, or at least mainly, on the average effect of a particular variable, intervention, or policy.  Over the years, we've developed increasingly sophisticated methods to more accurately estimate these average effects.  But I rarely hear people discuss the differential effects of the variable, intervention, or policy of interest.

In other words, we keep trying to figure out if different policies "work," but we define "work," as making a statistically significant difference for the average student, teacher, principal, school, district, or state.  If we instead asked for whom a given policy works, we'd likely find that it works very well for some while harming others.

In the talk, Gladwell mentions a food scientist hired to create the "one best Pepsi" who conducts taste tests of the product with varying amounts of sugar.  To his surprise, no one, clear winner emerges.  Some people prefer only a little sugar, some like a medium amount, and some like a lot.  He then has an epiphany and realizes that there's no such thing as the "one best Pepsi" (which eventually leads to the creation of a wide variety of Prego sauces that target different audiences).

I'd argue that education policy is almost exactly the same.  For example, imagine a new math curriculum.  How do we decide if it works?  The gold standard of research would dictate that we randomly assign, say, 100 classrooms to use the new curriculum and 100 to use the old one.  We'd then compare the average scores at the beginning and end of the year of the treatment and control groups.  If kids in the treatment group score significantly higher, on average, than the control group then that curriculum earns a gold star.

We spend increasing amounts of time trying to figure out which math curricula will yield the largest achievement gains across the students who use them.  But it seems exceedingly unlikely that "one best math curriculum" truly exists.  Some states, districts, schools, teachers, and/or students will do best with curriculum A, some with curriculum B, and some with curriculum C.

Wouldn't our time be better spent figuring out for which students curriculum A would be best and for which students curriculum B would be best (and why)?  And we could say the same about all of the largest issues in education policy -- teacher training, teacher pay, charter schools, and so on.  In later posts, I'll explore how our research and policy would differ if we aimed to understand and account for variability rather than simply finding the one best policy.

I've long thought it odd that we spend most of our lives being told not to stereotype and make generalizations while the most educated people in the country strive to make the largest generalization possible in their research.  Most (though not all) researchers focus incessantly on "generalizability" (one of those words we use in the field but the spellchecker won't recognize): if we can generalize your findings to 10 million students across the country, you're likely to publish the paper in a better journal than if we can only generalize your results to students in one school.

As it turns out, I was right to think that odd.  Greater generalizability matters in many ways.  But, at some point, we start missing the point -- and for exactly the same reasons teachers and parents across the country are telling our kids not to generalize.  Everybody is different.

Every student is different.  Every teacher is different.  Every principal is different.  Every school is different.  Every district is different.  Every state is different.  The best policy, on average, will help one and hurt another.  Until we understand this variability, we're doomed to fail.