I think QS is also adopting the new ranking methodology this year.
So no more Student Selectivity…
And only three public universities (i.e., UC Berkeley, UCLA, and U of Michigan) on T20 list.
As others have noted, US News creates it’s rankings primarily by surveys. I think I read this yr they only received approximately 50% of their surveys. So this ranking is simply opinion of half of the survey takers. Although not without value, I prefer rankings like Poets and Quants that use data (admissions percentage, test scores, income - broken down by some majors). Blending the two gives even more perspective…
From US News website –
“U.S. News collected the most recent data by administering peer assessment surveys in spring and summer 2020 directly to schools. Of the 4,816 academics who were sent questionnaires in 2020, 36.4% responded – a decrease from the 43% response rate in 2019”
So roughly 1/3 of colleges responded this year. I wonder if this is indicative of academia’s opinion of the value of this component. And as discussed in an article in the WSJ, schools they interviewed admitted that often the people completing this survey are not presidents/deans/provosts, but staff who get delegated the job and have acknowledged they could not possibly know all the information that would be required to properly evaluate other schools. Yet it makes up 20% of the rank. I personally will never understand that.
I would like this post 1,000 times if I could.
For all those who are posting about the various ranking lists and methodologies, doing analyses on changes from year and such…I am certain there are underprivileged HS students who could use your pro-bono help in navigating the college admissions process.
“Harvard <=$65k Income – $0k cost to parents
SUNY $65k Income – $23k cost to parents”
I’m going to put this in numbers so maybe people understand better this flawed and misleading argument about Harvard being an affordable choice for middle class families:
Harvard acceptance rate - 4%
SUNY acceptance rate - 60%
"I think I read this yr they only received approximately 50% of their surveys. So this ranking is simply opinion of half of the survey takers. "
I’ve worked with marketing groups and that’s a pretty good response for surveys. The features in many products we use are based on surveys with less respondent rate.
The discussion was about whether Harvard was affordable for a government worker family that was not “super rich”, not whether it was easy to get accepted.
Colleges that are very low cost to typical lower and middle class families tend to have very low acceptance rates. This goes beyond just HYPSM… type highly selective private colleges. For example:
Colburn – 5% acceptance rate
Navy – 8% acceptance rate (who they count as applicants is questionable)
Ozarks – 11% acceptance rate
Already mentioned in an earlier post regarding most UC’s moving up in the US News rankings, but note that UCSC was the only UC to decrease despite being admitted to the prestigious Association of American Universities (AAU) in 2019. Membership is by invitation only. Excluding UCR and UCM, all UC’s are AAU members, which includes all Ivy League schools, Stanford, MIT, other elite private schools, elite state flagship public schools like Michigan, Minnesota, Wisconsin, Washington, Oregon, etc., totaling 63 US universities members.
Here is a link for AAU members including the 7 UC members with the US News 2020 rankings of all 63 members.
https://en.wikipedia.org/wiki/Association_of_American_Universities
US News ranking of UC’s from 2020 to 2021:
UCLA #20, no change from 2020 to 2021
UCB #22, no change from 2020 to 2021
UCSB #34 in 2020 to #30 in 2021;
UCI #36 to 35;
UCSD #37 to 35;
UCD #39 to 39, no change;
UCR #91 to 88;
UCM #104 to 97; and
UCSC #84 to 97 (decreased 13 spots)
As mentioned in an earlier post UCM moved up to the top 100 and the school is only about 15 years old!
I doubt that the UC improvements in USNWR ranking have much to do with being admitted to the AAU in 2019 since that is not a direct factor in USNWR ranking. In your comparing to 2020 list 3 of the UCs you had no change, 2 change by 1-2 spots, 2 changed by 3-4 spots, and 2 changed by larger amounts. I listed the 2 that changed by 5+ spots below.
Comparing to 2020
UCSC – Decreased from 84 to 97
UCM – Increased from 104 to 98
Comparing to before the AAU in 2019 shows UCM had huge improvement in rankings, and UCSC had a huge decline. Both UCM and UCSC continue their improvement/decline in 2019 and 2020. The change from 2020 to 2021 above is continuing a trend that began in earlier years. UCM improved 32 spots in 2020, then another 6 spots in 2021. UCSC declined by 14 spots in 2020, then another 13 spots in 2021. I am not familiar enough with the schools to guess at why the change in USNWR ranking is occurring at these 2 schools, but I doubt it has much to do with AAU admission.
The others 7 had relatively little change in both 2019 and 2020. Overall 6 had a decline in rankings since 2019, and 2 had an improvement. I don’t see much pattern. It’s certainly nothing suggestive of AAU admission improving ranking. It probably has more to do with the USNWR methodology changes in this period.
Comparing to 2019
UCM #136 to 97 (large improvement)
UCSC #70 to 97 (large decline)
UCSD #41 to 35 (small improvement)
UCSB #30 to 34 (small decline)
UCR #85 to 88 (small decline)
UCI #33 to 35 (small decline)
UCD #37 to 39 (small decline)
UCLA #19 to #20 (small decline)
UCB #22 (no change)
Another reason for the UC rankings increase is that USNews is now factoring in student debt load and the percentage of students on pell grants, where the larger UC campuses excel since the student mix is generally more diverse in terms of socio economics.
UCs would do extremely well compared to similarly ranked colleges, if USNWR captured some good metric about SES diversity. For example in the NYT mobility study at https://www.nytimes.com/interactive/projects/college-mobility/university-of-california-los-angeles , 4 of the 7 highly selective publics with the most economic mobility were UCs – UCI, UCLA, UCB, and UCSD. They all had more economic mobility than any private college that is ranked well in USNWR, and they all had more than double the economic mobility of any Ivy.
However, USNWR rankings doesn’t use a good metric about SES diversity or SES mobility. They also do not directly rank based on % on Pell kids. Instead they rank the Pell kids’ graduation rate, which largely follows non-Pell graduation rate. I expect that highly selective Ivy-type private colleges do quite well in this category, including ones with a low percentage of Pell kids.
Nearly half of the new debt category is percent of students claiming federal loans, so I expect UCs being more SES diverse hurts them in this category overall. A ranking of the “national universities” that I’d expect to do the the best in the new debt ranking overall is below (combination of average debt of federal FA recipients and portion claiming federal loans). Unlike Caltech, Harvard, and similar; UCs have most loans as federal rather than institutional. For example, Caltech’s mostly institutional loans do not incur a USNWR penalty, but UCs mostly federal loans do incur USNWR penalty. I skipped University of Puerto Rico and some other lower ranked colleges, so actual rankings are worse than listed.
As summarized below, I think this category might help UCB and UCD slightly, but would probably slightly hurt the other 7.
Colleges Best in New USNWR Federal Debt Category
- Princeton – 4% and average of $10k
- Caltech – 5% and average of $9k
- Harvard – 2% and average of $13k
- Stanford – 8% and average of $11k
- Yale – 8% and average of $13k
- MIT – 11% and average of $13k
- Rice – 12% and average of $12k
- CUNY: City – 13% and average of $12k
- Vanderbilt – 12% and average of $15k
- Chicago – 11% and average of $17k
- BYU – 13% and average of $15k
- Duke – 20% and average of $11k
- Penn – 14% and average of $19k
- WUSTL – 18% and average of $16k
15. UC Berkeley – 22% and average of $14k - Northwestern – 21% and average of $15k
- Dartmouth – 26% and average of $12k
- Brown – 23% and average of $15k
- Tufts – 23% and average of $16k
- U Florida – 24% and average of $16k
- U Washington – 26% and average of $15k
- Georgetown – 27% and average of $15k
- Utah State – 29% and average of $14k
- SDSU – 28% and average of $15k
- Cornell – 31% and average of $14k
…
~30. UCLA – 32% and average of $15k
~30. UC Davis – 34% and average of 13k
…
~40. UCSB-- 35% and average of $15k
…
~45. UCI – 35% and average of $16k
…
~55. UCSD – 36% and average of $17k
,
High - UCM / UCR / UCSC – 42 to 48% and average of $19k
Interesting @Data10, can you also post this analysis for the national liberal arts colleges? Would be interesting to see how that helps the military academies.
US military academies do not participate in federal FA programs, so no information is not available. The federal database reports either all zeros or “data not available.” I didn’t include them due to lack of information, but they’d probably be ranked best in this category had stats been available.
- Berea -- 15% and $6k
- Pomona -- 12% and $10k
- Amherst -- 17% and $13k
- Haverford -- 18% and $15k
- Bowdoin -- 18% and $16k
- Bates -- 22% and $13k
- Wellesley -- 24% and $11k
- CMC -- 22% and $14k
- Trinity -- 13% and $21k
- Williams -- 24% and $13k
The only US service academy that has non-zero monetary costs to the student is the US Merchant Marine Academy (costs were around $9k in prior years, dropped to around $5k recently).
https://nces.ed.gov/collegenavigator/?s=all&id=197027#finaid says that 6% of beginning undergraduates received Pell grants, and 7% used federal loans.
Thank you for posting @Data10, this is very helpful. Great to see Berea on top since they have done so much to create a new model focusing on first gen students.
Or is it just possible that the East Coast universities are better based on what is measured?
USNWR doesn’t directly consider location. The referenced west coast universities are certainly not ranked poorly. I do think public colleges are underranked due to USNWR methodology, but the highest ranked public colleges were ULCA and UCB, rather than east coast schools.
I don’t think Pomona in #4 and CMC in #6 is a big anomaly, suggestive of a bias against west coast . However, Pomona never ranks as high as Williams. I believe Williams has been ranked 1st on the LAC list during each of the past 17 years. It’s more that the methodology favors Williams than a coastal bias. There are plenty of good east coast LACs that are consistently ranked below both Williams and Pomona.
Stanford does usually rank below HYP, and I explained the reasons for that earlier in the thread. I think a big part of it is Stanford’s popular 5/6 year co-terminal masters program, which hurts their graduation rate. If HYP had a similar number of co-terminal masters students, they would probably also rank lower.
Right, as the methodology favors HYP over S.
But, that begs the question: is/was the methodology tweaked to ensure HYP rules the roost?
When the rankings first came out in 1983, Stanford was #1 for three straight years, but quickly dropped to #6. It did get back to #2 in '91, but has been struggling to get back there ever since.
Also back in '83, Michigan and Illinois were top 10 and ranked higher than MIT.
Cal Berkeley was #5, and then dropped into the 20’s three years later.
There was some interesting methodology tweaking going on in Year 4 of the report: Yale, Princeton, Caltech, Harvard were the top 4. The above publics were no where to be seen. But, tweak again, so Caltech drops to 5 in Year 5 and HYP assumes its rightly place in the world with young whipper-snapper Stanford at #6.
(A cynic would say such a ranking is much better to sell magazines to the population centers of the east coast.) 
It was more a complete overhaul than a tweak. The first few rankings were entirely based on a survey USNWR sent to college presidents, asking them to name the top undergraduate colleges. The top 5 USNWR ranked colleges based on this survey were something like below.
- Stanford
- Harvard
- Yale
- Princeton
- Berkeley
After the college rankings articles started to gain popularity, USNWR hired statisticians to come up a more scientific looking formula for ranking colleges. The first guy they hired created a formula, which ranked Hellenic College of Theology #1. . Hellenic College of Theology had a high expenditure per student, driving up its ranking.
USNWR didn’t think this ranking would go over well, so they hired another statistician to create a different best college formula. The 2nd formula used 5 metrics to rank colleges and added up the scores of those metrics – SAT scores, acceptance rate, yield rate, expenditure per student, and % of faculty with a PhD. I believe SAT scores had the highest weighting. The new rankings had HYPSMC as #1 to #6, which USNWR apparently approved of, although some might consider the order unexpected with Yale #1 and Harvard below Caltech. The new ranking formula hurt publics like Berkeley, which dropped from #5 to #24. And helped selective privates with large endowment per student, particularly smaller ones like Caltech that didn’t do as well on the college presidents survey, which increased from #21 to #3.
In the following year USNWR had another major change and switched to the following weightings, which return an academic survey and add in a small weighting for5-year graduation rate. The new weightings yielded more changes in the rankings. With the peer survey returned, Berkeley had another large change – jumping from #24 to #13, and Caltech dropped below Harvard. The top 3 were Yale (#1) , Princeton (#2), and Harvard (#3).
25% – Academics Rank Colleges on Marginal=1, Distinguished=5 Survey
25% – SAT Scores, Acceptance Rate, Yield (“Selectivity”)
25% – Faculty Ratio, % of Faculty with PhD, Faculty Salary, … (“Faculty”)
20% – Expenditure per Student
5% – Graduation Rate
Throughout the 90s, USNWR made smaller changes to the rankings, which might be considered tweaks – things like adding more weight to graduation rate and adding in an alumni giving weighting. By 1996, the “best college” weightings had become more stabilized, with very few changes in the next few years. These more stable weightings produced the following “best colleges.” At this point, the college rankings edition of USNWR were huge sellers, by some estimates earning millions each year.
- Harvard
- Yale/Princeton
- Stanford
- MIT
- Duke
- Caltech/Dartmouth
- Columbia
- Chicago ...
- Berkeley
It should be obvious that the weightings are completely arbitrary and are not a meaningful measure of the “best colleges.” The best colleges for any particular student is likely to be completely different from the ranking on such lists.