Gaming the USNWR rankings

Parents Forum
1

Replies #12 on of http://talk.collegeconfidential.com/discussion/comment/21650912/#Comment_21650912 became a sub-discussion about colleges gaming the USNWR rankings. For reference, the ranking criteria are given at https://www.usnews.com/education/best-colleges/articles/ranking-criteria-and-weights .

Some examples of how to game the USNWR rankings:

  • Have smaller classes capped at 19, 29, 39, or 49 students, the thresholds to be counted more favorably in the "small class size" part of the rankings.
  • Emphasize test scores in admissions and scholarships, since test scores are the biggest component of the selectivity part of the rankings.
  • Offer test-optional, but avoid having more than 25% of the students take the test-optional route.
  • Offer spring admission to weaker admits. While sometimes used as a load-balancing method (since fall semester is more enrolled than spring semester otherwise), it can also move weaker admits' stats out of the view of rankings criteria.

Of course, these are not necessarily without value to the students. For example, any reduction in class size may be desirable to students who want smaller classes (but perhaps less so to those who cannot take the class because it is full at 19 but would have space if is allowed 25). Spring admission may be better for some than no admission. But sometimes the ranking effect is the intended effect with the value to the students a by-product, rather than the other way around.

2

The classic way that they game the rankings seems to be by increasing the application pool through over-the-top marketing campaigns, blanket application fee waivers, and lowering other barriers to applying.

3

The guaranteed sophomore transfer option is another way. It is used by many colleges including Cornell.

4

A discussion from a few years ago about a college that was very successful in the game.

http://talk.collegeconfidential.com/parents-forum/1681335-how-northeastern-gamed-the-college-rankings-p1.html

5

Early decision, SCEA and REA – all increase yield which then allows for a lower admit rate when RD is factored in.

Heavy use of the wait list to fill out the freshman class – also increases yield, decreases admit rate.

Spring admissions and large amounts of transfer students. Both keep the reported frosh fall incoming class smaller and higher stat.

Merit scholarships to incoming frosh to boost the stats at the 75th percentile and also increase yield on offers to high stat applicants. The best class money can buy.

Generous use of super-scoring which raises everyone’s reportable test scores a bit.

6

Colleges really can’t win, can they?

If they engage in “heavy use of the wait list”, they get accused of gaming. But if they don’t, and they “over-admit”, parents are in an uproar about three kids housed in a double, or four kids in a triple. If they don’t accept transfers, there is much grinding of the teeth about how elitist their admissions practices are. If they don’t offer merit aid, they get accused of keeping out the donut hole/upper middle class families who don’t qualify for much need aid AND they get accused of social engineering by admitting a class comprised of full pays and Pell recipients. etc.

7

@northwesty

Those admitted from the waitlist must be included in the acceptance rate.

8

“Colleges really can’t win, can they?”

Nope, they can’t. Another example is the criticism of colleges’ extended marketing efforts. A few years ago a joint study of a Stanford and Harvard team showed that most high achieving, low income students didn’t even apply to top selective schools and instead ended up at state schools; the study recommended that colleges increase their marketing efforts - specifically mailings - to attract these top students (often first gen and low income) to apply. Many colleges have done just that. But if you read the posts of the distraught parents on this board, the increased marketing is a nefarious effort to squeeze another $75 app fee out of applicants, get more applications just to reject and boost rankings, and surely just a way to tease their precious child with the hope of a college s/he has no chance of attending.

9

If everyone, or at least the top ~50 schools, superscore, does it matter? (Yes, I know that the UC’s do not superstore, and a couple of them are in top 50.)

Transfers are an interesting question. Are the top publics, which accept lotsa transfers, really gaming the system, or just fulfilling the state’s mission and giving first gen and low ec kids a shot at the state flagship?

10

@bluebayou The schools mentioned by CCers offering guaranteed sophomore transfer are all private.

11

If I were trying to rise in the USNWR rankings here’s what I’d do. Some things would hurt my school in other rankings but if my goal were climbing the USNWR list…

As already mentioned,
Make use of Jan admissions for weaker candidates (outside the top 10/25%, weaker scores)
Admit a small percentage (under the threshold where a penalty applies) of the class test-optional.
Limit class sizes to one under the thresholds.
Lure strong scoring kids with merit scholarships.
Heavy use of ED.

In addition I’d add,

Admit more local middle and upper class kids. Forget about economic diversity, as kids with a lot of need tend to graduate at lower rates. Admit more local kids because they are slightly less likely to transfer out due to weather and culture shock. Kids with $ are also more likely to take the SAT or ACT many times, resulting in better superstores than kids who can’t afford to do so.

Ignore the SAT essay and any lopsidedness in ACT scores. Only the SAT CR+Math and ACT composite count. Consider reporting your admittees’ superscored results.

Fewer faculty better paid. Student-faculty ratio counts much less than compensation/faculty member and % faculty with a terminal degree. Make liberal use of TA’s and administrators as part-time teachers since they don’t count in the compensation statistics.

Pack classes to the threshold limits. No seminar classes with 8 students since 20 is the lowest limit for which USNWR gives credit.

Get graduated seniors to donate a dollar. Ups the participation % incrementally. Some schools practically require the years graduating class to donate.

Pay more attention to the kids at the top 10% of the class. No sob stories (I had cancer! I just immigrated 5 years ago!) need apply. If your school ranks and you’re outside the top 10%, or for a less competitive school, the 25% you’re out of luck.

Make it tough to flunk out. Give a lot of kids extra chances and readmit quickly. Ups the graduation rate.

Make sure to fill out the peer assessment every year. Trash those above you in the rankings. Yours may be thrown out, but hey it might just be the 3rd or 4th low ranking and thus count.

Adjust your first year retention reporting based on which years look best. “If a school submits fewer than four years of first-year retention rate data, then the average is based on the number of years that a school submits to U.S. News.” If your strongest retention #'s are from 4 years ago submit all 4. If your strongest is a year ago submit only that year.

Special skills? Who cares. They don’t factor into the equation.

Sadly, I’ve seen schools do this kind of thing and have it work. The power of USNWR is such that as schools rise in the rankings they attract better students and the rankings become a self-fulfilling prophesy.

12

In some of these cases, it can be that any “ranking gaming” effect would be the by-product, rather than the primary intent. If the university had similarly high transfer intake for decades, including before there were USNWR rankings, it is likely that any “ranking gaming” effect from that is a by-product.

Sometimes, it is not obvious whether a given college policy was primarily intended to game the ranking, or was intended for other reasons. Spring admission is an example, since it can be used for reasons other than ranking gaming (mainly enrollment-balancing the fall and spring semesters, since the fall semester tends to be more heavily enrolled if everyone starts in the fall).

But then if there is concern about ranking gaming of this nature, they can alter the selectivity-based part of the rankings to include:

  • Spring admits who originally applied for fall admission in the frosh selectivity stats.
  • Transfer students, with selectivity measured by prior college GPA.
13

Spring admission started way back when as a way to compensate for the fact that more juniors spent second semester abroad than first semester- hence, more vacant beds Jan-May than August-December.

People criticize the colleges for not 'acting more like businesses" and worrying about capacity, revenue smoothing, etc. and then criticize the colleges for things like Spring admissions which were designed to do exactly that.

Why not get the holy “rankings industry” to stop penalizing colleges for stupid things?

14

Why would study-abroad (or other off-campus activities like co-ops) be more likely to be done in spring than fall? Is it because the abroad universities are “full” in the fall (for the same reasons that universities here may be)?

However, spring ordinarily has lighter enrollment for other reasons:

  • Most of those who graduate a semester early or a semester late usually has one extra fall semester than spring semester.
  • Probably about half of those who drop out or flunk out end up with one extra fall semester than spring semester.

Obviously, spring admission would help with enrollment-balancing. So would encouraging students who want to do study abroad or co-op to do so in the fall.

15

Study abroad (way back when) skewed Spring semester because students then spent the summer traveling before returning to their home campus for senior year.

16

Schools with strong football teams have more study abroad students in the spring.

17

yeah, I get that Tom, but I was raising the broader question.

btw: besides Cornell, which other schools offer a GTO (beside publics)? And, how many actually accept the offer? Are there really enough transfers to really sway the SAT/ACT medians should they have been accepted for fall?

full disclosure: I’m a fan of the GTO. It’s a great way to: 1) reward ED applicants; 2) handle legacies; 3) address instate students (for those GTO’s into contract colleges). Since Cornell is already transfer-friendly, why not offer kids that you like early, so that they can spend three years on campus, instead of two after earning an Associate’s degree?

And, as someone noted upthread: an acceptance, albeit delayed start, can still be better than a rejection.

18

I believe uc schools have a cc to uc programs in place. I don’t have any problem with that related to gaming the rankings.

But if you admit students for fall and ship them overseas like NU.in. All of the scores and admissions percentages etc should be reflected.

Let a bunch of great kids in the spring. That’s great and a good tool for schools to manage their classes. And it’s which is used heavily at some schools. NU gtown and Bu and some Ivy’s I know They should all be in the numbers as well.

Offer a satellite campus like Emory oxford or NYU shanghai or wherever just add them to the pile. Showing the stats separately via their websites is helpful but just add them all up and report to rankings agencies. As a totality.

Biggest scam is prestige ratings. Many which get filled out by guidance counselors who have an idea of prestige hard wired years ago.

And frankly are not that active in the college process. Check out the threads complaining about understaffed or uninspired public school gcs .

And it becomes a circular fee back loop. Top schools in the rankings have a brand that is generational and they are asking whether 2018 is the same.

And why is a subjective opinion of prestige weighted so heavily? No outcomes no student feedback.

And if a current student on cc asks for us to rank in prestige or focuses on this - there is a a backlash on the thread against that. But it’s 22 percent of what adults think should determine rank.

Sorry rant over.

19

@TomSrOfBoston “Those admitted from the waitlist must be included in the acceptance rate.”

Yes, but the yield on those WL offers is very very high. The schools that do this work the WL very hard so they know that most/all of the kids getting the WL offers will accept. Vandy, for example, fills about 10% of its class each year off the WL. Basically, it is the May/June version of ED – if we accept you, you will enroll.

“@bluebayou The schools mentioned by CCers offering guaranteed sophomore transfer are all private.”

For example, Notre Dame has 75 kids each year in its “Gateway” program. The kids attend frosh year across the street at Holy Cross College. They can go to football games, take some ND classes and participate in many aspects of ND student life. If they get reasonable grades at HCC (not too hard since HCC is less competitive than ND) you get a guaranteed spot at ND for sophomore year. Gateway is often used legacy applicants who we just a little bit low to admit for frosh year.

Helps ND’s stats and also brings some tuition revenue to HCC (which is run by the same order that runs ND).

20

“Biggest scam is prestige ratings. Many which get filled out by guidance counselors who have an idea of prestige hard wired years ago.”

Definitely. I’m disappointed that none of the rankings has a meaningful component that is based on actual student feedback. Yes, it would be expensive and a PITA to conduct student surveys, but that data would be hugely meaningful for potential applicants. If the survey was well-done, I’d give that type of result and ranking a lot of consideration in a college selection process. Students who have just graduated from a school would have very valuable insight on actual quality and student experience at each college.