GMAT Scores, Yield, and Selectivity (for Kellogg)

Interesting analysis.

Rhyme, you're a freak.

Thanks.

Rhyme, I read through your analysis and I liked it a lot.

I did have one question. I believe for Businessweek's numbers, the total full-time enrollment reflects all current MBA students, in other words 2 classes of students.

My calculations are as follows (based on Businessweek's figures):
4299 applied
989 accepted (based on 23% admit rate)
564 enrolled (based on 57% yield)

I assume both admit rate and yield are rounded off, so the actualy total should be 560-575 I guess.

Now, Businessweek gives total enrollment at 1250, which actually means that 625 students should enroll each year, but I guess varying annual rates could account for the differences in total numbers every two years. What do you think?

Actually, I'm revising the numbers - several different links at Businessweek have different numbers. The most current appears to be this:

https://www.businessweek.com/bschools/05 ... ellogg.htm

3918 applied
940 admitted (24%)
526 enrolled (56% yield)
Total full-time 1310

I believe that the difference between Rhyme's Businessweek link and Kellogg PDF are that the link was for '03, while the PDF was for '05.

Freaks make up 50+% of the HBS student body. The rest are the sons and daughters of royalty, influential plutocrats/bureaucrats, nobel prize winners, world class athletes, extremely under-represented ethnicities, and, of course, the ultra rich. Sad, but true. Rhyme is just really doing his homework in order to maximize his chances of admissions at a few big name b-schools.

I know a Japanese at HBS right now, and yes, he is definately what most of us would consider a freak^3!

I would argue that your yield spread is not nearly wide enough. My guess is that yield rate for applicants under 640 would be well over 80%, quite possibly over 90%. For people with scores in that range Kellogg is obviously a "reach" school, and such an applicant probably applies to just 1 or at most 2 reach schools. It would stand to reason that this type of applicant is not often admitted to their "reach" school, and I would venture to guess that very nearly all of those admitted accept and enroll.

On the other hand, Kellogg has a reputation of being a back-up for people applying to Harvard and Wharton. I read (I can't remember where, maybe clear-admit?) that Kellogg has actually been wary of accepting superstar candidates who they know are targeting Harvard/Stanford/Wharton. If I recall, the blog/forum/board I was reading stated that it was very important to show Kellogg that they are your number one choice because they wanted to increase their yield and hated being thought of as a back-up.

Anyhow, people that score in the 750-800 range probably apply to several schools in the ultra-elite bracket, say 5, and if they are admitted to 2 or 3 or 4, that means that the yield could not possibly be greater than 50% at any of the schools admitting them, and could be as low as 25% or even lower. Kellogg is certainly a top-choice school for many people, but in the broad landscape of business schools, I think that Kellogg trails (ever so slightly) Harvard, Stanford & Wharton, and quite possible Chicago, Columbia and MIT as well. I would be shocked if their yield on 750-800 scorers approached 40%.

Harvard has an overall yield of 90%, Stanford 80%, Columbia 75%, Wharton 70%, MIT 70%, Chicago close to 70%. Kellogg has an overall yield under 60% (which actually trails, for example, Michigan at 63%). It would make no sense that ultra-high scorers would prefer Kellogg at a disproportionate rate - I think common sense tells us that ultra-high scorers, just like most others, would prefer Harvard or Stanford if they are admitted.

(Please note that I think Kellogg is a great school, and if I were admitted at this moment I would drop everything and celebrate...now back to the issue at hand).

Again, pure speculation but I believe yields would look more like this:

up to 640: 90-95%
650-690: 75%
700-740: 50%
750-800: 40%

Remember, their average GMAT is 700, which necessarily means that the yeild rate of admitted students over 700 (including the 700-740 range) will be a chunk lower than their overall yield rate of 57%.

Looking once again at the figures provided in the official Kellogg document, I would guess things look something like this:

(For 2 year MBA & MMM applicants because their numbers are grouped together, all numbers rounded off)
up to 640:
total applicants 785
total enrolled 50
total offers based on estimated yield of 90% 55
admit rate 7%

650-690
total applicants 1100
total enrolled 139
total offers based on 75%: 185
admit rate 16.8%

700-740
total applicants 1610
total enrolled 272
total offers based on 50%: 540
admit rate 33.5%

750-800
total applicants 432
total enrolled 89
total offers based on 40%: 225
admit rate 52%

I think these numbers make sense - they "smell" right to me. For example, the 650-690 score range is below that average for admitted students. That necessarily means that the acceptance rate is below their overall average of 23%. I do not think that Rhyme is correct with his calculated estimate of 22% admit rate for this score range; that would put the average GMAT for Kellogg at somewhere around 670 when it is actually 30 points higher. Remeber, ALL of the applicants in the score range scored below the average GMAT for Kellogg. Their acceptance rate must be substantially lower than the 23% overall average.

The acceptance rate for 700-740 "smells" pretty right as well. Again, their overall acceptance rate is 23%, and 700 is their average score for accepted students. So, basically every student in the 700-740 range was above their average, so a 33.5% admit rate for these students as compared to 23% overall feels about right. As with all the other numbers, the is likely variation within the range. For example, 700 scorers should be right at the overall average of 23%, while 740 scorers should be quite a bit higher. In the same way, 690 scorers are probably around a 20-21% rate, while 650 scorers are probably just slightly higher than the group "under 640".

For applicants scoring under 640, I was actually surprised by the number of applicants in this range. 20% of all applicants are essentially taking a long-shot. As argued above, I believe that well north of 90% of those admitted with this score will enroll. To think of it another way, getting accepted with a score in this range is a long-shot, winning the lottery, and the only people that will turn down an offer such as this is someone that won the lottery twice in the same year by gaining admission to another school of equal caliber even with the under 640 GMAT. I just don't believe that happens very often.

For applicants with scores of 750-800, my take is the yield must be substantially lower than the overall average of 57%. These are the people that are able to gain admission to several top schools and will necessarily turn down 1 or 2 or 3 or more. An overall acceptance rate of 52% feels about right.

So Rhyme, I like your analysis. It was an interesting break from studying for the GMAT. I do, however, disagree with your estimated yield numbers and the results that they produced. It's just not possible for the admit rate for 650-690 scorers to be essentially equal to the overall average admit rate - these scores are entirely below the average GMAT for this school. Also, I think that those scoring under 640, and to a large degree those scoring 650-690, are extremely likely to enroll if admitted. Sliding between 90+% and 75% for score from 640-690. It just makes sense that they won't have as many other top offers, and therefore no reason to decline. Alternatively, those with higher scores are likely to have many more tantalizing offers, and by necessity must decline a much larger number of ultra-elite admission offers.

GMATT, good point. I would be interested to find out how many of the 50 or so students admitted to Kellogg with GMAT scores below 650 fall into the categories that you mentioned of "sons and daughters of royalty, influential plutocrats/bureaucrats, nobel prize winners, world class athletes, extremely under-represented ethnicities, and, of course, the ultra rich".

Or alternatively, what chances someone not in one of those categories has of gaining admission with a GMAT below 650.

Sorry for the confusion Pelihu, I was referring to HBS, not Kellogg. I have no relevant info on Kellogg`s stats, and therefore shouldn't have even posted on this thread. Essentially, I was just defending the comment of Rhyme being called a "freak" because he has mentioned some of the other schools he is applying to.

The 'freak' comment was all in good fun. Rhyme is a friend - on and off the board. I actually know his real name!!!

And I'm probably freak-ish, too - I got a 760, which, when I got it, definitely made me freak out.

Good points. I would agree the spread is actually much larger. Interesting stuff isnt it?

In any case, my odds of getting to Kellogg are similar to my odds at Chicago and Harvard, minimal

Interestingly enough however, and contrary to my long held beliefs, I got some curious feedback to my GMAT score. In talking with several (four) Chicago graduates over dinner the other night, they all seemed to think my GMAT score secured me a spot. I can't imagine its that simple, but its encouraging to see consensus in this regard.

Hi,

Yes very interesting. Regarding your input from the 4 Chicago grads, I think that is good news. I remember reading on the official Chicago Business School admissions forum about GMAT scores. An applicant with a very impressive overall application asked if he should re-take the GMAT. I believe he was around 660, but everything else was spectacular.

The response, from someone with the Chicago Business School admissions committee was that it seemed like this person would be a good candidate for admission, but that he/she should still consider re-taking the GMAT. The reason was that many employers do in fact look at GMAT scores, and that even if this person were admitted, their GMAT score would follow them throughout their career (I think they might have said it might haunt him/her).

Sure, this was just a single response for a single person, but I think there is ample evidence from official school forums, and admissions consultant blogs, and so forth along these same lines. I believe that schools like to say that they consider the overall package, and that a low GMAT shouldn't discourage anyone from applying, and that it's only one factor and so forth. It's in their interest to get as many applications as possible - it increases their fees they collect, but more importantly it makes their admit & yield percentages look better. But when it comes right down to it, GMAT matters a lot more than they let on.

Were the 4 Chicago grads you spoke with recent grads? I believe that there has been a trend in recent years of a larger pool of competitive GMAT scores - especially on the Q side.

Rhyme, Pelihu - dudes what have you been smoking?

Anyway, it was very interesting to read all that (though some of it went over my head because I wasn't concentrating).

I would assume that depending on the position of the school w.r.t rankings and student preferences, the yield would sort of be a bell curve, with the max yields around the median range of the school's GMAT. Those who are low would be at the lower ends, and those who have very high scores probably will fall outside the school's "acceptable reputation" for such achivers.

Having said all that, I found the chicago graduates observations based on GMAT scores quite interesting. Anybody got ideas on how a 740 would do for a INSEAD applications :D :D would love to know!

(I'm not applying to any US school - damn all of them are 2 year courses and I'm too old for that)

Have you read "Managers not MBAs"? Its a great book. Highly recommended.
Written by an ex HBS professor.

There's an extremely telling portion in the book, where he recounts a conversation with a dean of, as he puts it, a "competitive school" looking to "increase its rankings."

In discussion with the Dean, he asks him how he intends to increase his rank, and the dean cooly replies with four of five points - I don't remember them all but I do remember these.

"1) Increase the cost of admission so as to appear more competitive.
2) Encourage as many unqualified applicants as possible to apply so as to artificially appear more selective.
3) Reject more candidates with lower GMAT scores for the sole purpose of
increasing the overall score.
"

The book goes on to say, without naming the school, so I dont know which one it was, that the Dean did exactly these things, and the schools ranked increased substantially, despite not changing any fundamentals in the curriculum.



I think they were probably all about two years out, so graduates of say, 2004 or so. They didnt specifically mention the trend of higher scores, but you are right that this is the case.

I forget where I read it, but in 1999, a competitive score for Kellogg was something like 650, by 2000 it had gone to 680, by 2001, 700. Something like that (my dates are probably off) but the gist of the article was just that: GMAT scores for the top schools keep increasing year over year.

Rhyme,

I have not read that book, but it sounds interesting. When I have some downtime (maybe in the spring when I trying to decide between several top schools ) I would love to give it a look.

Still, I have seen plenty of annecdotal evidence that supports those exact comments. For example, right here at the GMATclub there is an article that talks about some schools "buying GMAT scores".

https://www.gmatclub.com/content/courses/scores.php

I have read elsewhere that several years ago, Columbia was concerned about its ranking and placed increased emphasis on the GMAT scores for its admits. Some years ago, Columbia was towards the bottom of the top 10, but with the increased emphasis on GMAT scores and yields they are solidly among the ultra-elites. I look at their 12-15% admit rate, as well as their 75% yield and it sure looks to me like they are putting upward pressure on Wharton.

I believe that many people believed Duke did the same sort of thing for a while to move up and solidify its rankings, and if I were a betting man, I would be betting that Yale is in the process of doing the same thing to solidy their place among the elites, with the thought of moving up to the ultra-elites eventually (they are Yale after all).

I also think that Harvard must certainly maintain its admit & yield percentages by encouraging EVERYONE to apply. I personally know a whole bunch of people that were applying to schools like USC & Santa Clara, that said, screw it I'm taking a chance and applying to Harvard. Of course, for many people Harvard is not a stretch, it's a pipe dream, but I wouldn't be surprised if 25% of the people that apply to Harvard each year have ZERO chance of being admitted. I'm sure that Harvard gladly accepts the fees, and more importantly welcomes these numbers that help push their admit percentage from 15% down to 10% (I think their yield might well stay the same).

It makes perfect sense for every school to say, yes come and apply. Yes, we look at your entire application. Yes, the GMAT is only one factor, so don't let a 550 or 500 or even 450 stop you. I also makes perfect sense for schools to downplay the influence of an 'in-your-face' hard number like the GMAT score. After all, if schools came out and admitted that people that scores less than XXX have less than 2% chance of being admitted, they would probably lose quite a few applications, while hurting their own admit % and yield rates.

But if you look hard enough, I think it's clear that the GMAT has substantial importance for each school. Just think about how you would react if an elite or ultra-elite school reported an average GMAT 15 points lower than all of its peers. They simply cannot let that happen if they can help it at all.


Necromonger,

I have not read anything about business schools outside of the US, but I am happy to hear that you are taking your high GMAT score elsewhere . Good luck. A few US schools do offer 1 year programs, among them Cornell & Northwestern I think.

haha, this was great. Good job rhyme!

pelihu, if employers are that interested in GMAT scores, do students put them on their resumes when interviewing? Do you think that investment banking and private equity employers are the ones who really want to see the GMAT?

Yeah student (especially those with GMAT above 700) put them on their resume. IB and MC's are interested about the GMAT