Q&A about GMAT

平常心

Is there a real correlation between GMAT scores and success on the programme?


There is a very high correlation across race, gender and age, between the GMAT and the performance on the core component. We did a substantial study over five years in Europe, we did the same kind of study in the United States. And that correlation is strong. There is no research on the correlation between GMAT and ultimate success as an executive. Perhaps there should be.


Given that correlation – how much reliance do you think schools should place on GMAT when they are admitting a student? And how much latitude should they afford students with lower scores?


That’s entirely the decision of the school and there are a lot factors that have to come into play: work experience; undergraduate curriculum study; interviews; GMAT. Of those, the only normed assessment is the GMAT. If you have been out of school a long time and have not taken any preparatory stuff, it’s a pretty safe bet you haven’t tried to [calculate] a first derivative or figured out what the cosine is or dealt with Pythagoras. And the reality is that [this will] drive your score down. But it doesn’t suggest for a minute anything other than the fact that you should probably brush up on that stuff before you go to school because you may have to do it in your first semester.


Why is there such a large calculus component to the GMAT, when it forms a relatively small part of the MBA curriculum?


GMAT [is] driven by conversations with the schools. And what the schools find most unpleasant and most uncomfortable is admitting a candidate who then simply can’t survive. And what they find is that areas that can be fatal in the first semester and the first year tend to be the quantitative courses – finance, accounting production, micro economics, macro economics to some extent - things that demand a quantitative facility, which you often have in your tool bag but have not used for a while. That’s not to suggest that 50% of your fist year is going to be driven by purely quantitative questions, but rather that mission critical skills you’ve got to bring to the first year, have got to be in place.


Students can learn techniques of how to perform well on the GMAT. Does this undermine the fundamental point of the exam?


I don’t think so. I think what the preparation programmes provide is what a good coach provides an athlete. No matter how much coaching I ever received it was highly unlikely that I would ever run a mile in under 4 minutes 30. On the other hand, without training conditioning I could be running a six-minute mile, so what the coaching will do is hone your skills, practice you, drill you. I was fascinated to talk to some young people from one of the investment banks, and I couldn’t understand why they were taking prep courses – I taught them myself and they didn’t need it, they were bright and capable – and their answer was: “Discipline. The prep schools gave me discipline. Otherwise I’d stay at the office and work.” So I think it just trains you and hones you but its still not going to make you a world class athlete if you’re not a world class athlete.


Because GMAT is a computer-adaptive-test [whereby the test adjusts itself to the respondent’s ability level], not all candidates complete exactly the same questions. Is this equitable?


When we converted from a paper-based, linear test to the Computer Adaptive Test we ran a comprehensive comparative analysis. We did it over 5,000 [tests] – half taking the test on paper and then on computer and half of them on computer first and then paper, so that we could adapt the test and create comparable scores. So without going into the psychometrics, it does work. (You don’t get an identical test if you take it April then take it June either.)


Why are GMAT scores going up and up?


What is interesting is that the mean is not going up. It is stabilised at around 519. Last year we tested almost 250,000 people. Two years before we saw 208,000. So that’s 42,000 more. Very few schools see applicants who score under 400. But under 400 is roughly equivalent to north of 600. And so it appears the scores are drifting up because we are seeing more and more scores north of 600. There are more and more. But there are also more and more under 500.


GMAT has been criticised for being culturally biased. It is said that it particularly works against Asian students. Does GMAC believe this to be an issue? If yes, what have you done to combat this?


We are very sensitive to that. And being a Canadian who tested miserably on the verbal, I am very sensitive to it myself. So I commissioned in 1997 a comprehensive study of eight European schools over five years, and 33 different nationalities to determine whether there was a cultural bias and one of the deans of a schools gave us his biases right of the bat. He predicted that there would be two different nationalities that would suffer. One European and one Asian. We found that it was absolutely consistent - even with that impact as a predictive measure it was absolutely consistent.


You’ve also added essay questions. Has this helped to alleviate even that small remaining bias?


Absolutely. And by-and-large, the students who come through it come through well. The essay question was important because it allowed the schools to evaluate how well a candidate performed in a tight time-constraint, as opposed to with the essay they submit with their application that could have taken weeks to write.


In a similar vein, the test is only available in English. Does this disadvantage candidates whose first language is not English?


We spoke with our schools about whether we should adapt the test to other languages. And their reaction was: “We want to teach in English and so whether a candidate is facile in his native tongue doesn’t help us understand whether a candidate will do well in our programme." And so it was the schools who said don’t change it.


How about where an MBA is not taught in English?


They don’t use us. They use their own assessments.


So there are no plans to offer it in other languages, which might make it a standard across all programmes?


We are certainly investigating it. We are investigating and have created a pilot for the verbal reasoning piece in Spanish. Just as a pilot test. And we’ll be testing it out in the next little while and we are investigating others. But the question is whether or not there is a market demand for it.


Finally, what advice would you give to someone who has a GMAT test coming up?


This sounds really trite, but I would prepare depending on my own psyche. I would certainly prepare by buying the official guide and practicing with every question you could possibly practice with. The software is free, you can download it on our    site. And I would practice with that. And if you need the discipline or if you need to brush up on some of your skills I would take advantage of [prep] programmes. And use the advice our parents always gave us: get a good night’s sleep. Because I don’t when was the last time you sat for four straight hours for an exam but it’s a long time. There are very few people who have been working in the work place for three or four years who accustomed to it – they go and get a coffee or look out the window or take a break.



Dave Wilson has been the president and CEO of the Graduate Management Admission Council (GMAC) since 1995. The Council operates in 70 countries worldwide.     It is best known for the Graduate Management Admission Test (GMAT), a standardised test required by over 1,700 graduate schools of business worldwide by candidates seeking admission to graduate studies in management.


Mr Wilson received his MBA from the University of California (Berkeley). He has served as a faculty member at Queen’s University and the University of Illinois, and at Harvard University’s Graduate School of Business.

braveMBA

Thanks a lot.