<p>In collegehelp fashion, I took a look at the Bloomberg profiles to see what percent of LAC grads who obtained an MBA did so at a top 10 program. Only schools that had over 30 entries were included in the outcome, which stuck near USNWR rankings but not exactly. Any school that is called ‘university’ instead of college that has an MBA program or a common name that corresponds to more than one school, i.e Wesleyan, Trinity or Union is too involved to isolate out the data to ensure that it is specific to the school. Here are the results:</p>
<p>For purposes of the study, the top 10 MBA programs were:
Stanford
Harvard
Penn (Wharton)
Northwestern (Kellogg)
UChicago
Dartmouth (Tuck)
MIT (Sloan)
Columbia
UMichigan
NYU (Stern)</p>
<p>I don’t know how valuable this data is. I mean, is a Holy Cross student better prepared for business school than a student from Swarthmore (which is not on the list)?</p>
<p>I don’t believe that Swat has less than 30 entries.</p>
<p>One might argue that Swatties are more intellectually-oriented than Jeffs and Ephs, but even so, one would expect there to be a respectable number of MBAs from their population, and for that number to have a great success in admission to the top programs.</p>
<p>Swarthmore was not on the list because it had less than 30 entries; however, of those listed from Swat, 80% attended a top 10 MBA program. I get the impression that LACs like Swarthmore, Haverford, Vassar, Bates, Carleton, Grinnell, Oberlin do not attract as many students that would be disposed towards getting an MBA, which is at least partially why none had 30 entries on Bloomberg. Also, those schools don’t seem to have as much school spirit as Amherst, Williams, Middlebury, Colgate so that grads may have been less inclined to include their degrees on the profile even if they had obtained an MBA and were from one of those schools.</p>
<p>
</p>
<p>There’s no such thing as a link. It’s a manual process that you can only conduct if you have access to Bloomberg.</p>
<p>Here are a few other theory/explaination to consider. First, your cutoff of 30 entries is arbitrary and too high. While you’ve looked at a percentage of graduates from each school so you don’t need to do a per student adjustmet on your percentages, the size of the schools and the number of graduates they have directly impacts the number of total profiles a school might have.</p>
<p>For instance, Haverford, Swarthmore, and Grinnell are much smaller than all but 1 of the schools on the list. In fact, Williams, Hamilton, and Washington & Lee are 50% larger than Haverford. For other schools, like Middlebury, Cogate, Bucknell, and Holy Cross, the differences are even more dramatic because the aforementioned are between 2-3x larger than the likes of Haverford, Swarthmore, Grinnell. So, I don’t accept either theory (less likely to pursue business careers or school spirit) that gellino offers up. (The smaller student body is less true for Vassar and Oberlin.)</p>
<p>Second, you could have data selction bias. I don’t know enough about the Bloomberg profiles: what’s in them, who they cover, how people get in, what field they focus on, but many such databases are known to have a coastal bias (favoring inclsion of those on the East and West coast) which would surely hurt the likes of Carleton, Grinnell, and Oberlin.</p>
<p>I’m all for having fun with data and statistics, but I hope this one was meant to be taken as fun and not too seriously.</p>
<p>I chose 30 because that is the threshold that is generally considered acceptable to make data statistically significant. </p>
<p>I would grant you that geography bias could have some effect on the results and had thought of that (although this wouldn’t negatively affect Swarthmore, Bates and Vassar) as well as noticing the lack of entries from larger schools like Oberlin and Vassar. However, in considering the different Wall St positions I have held and my MBA class, I have met way more than 2-3x people from Middlebury, Colgate, Bucknell, Holy Cross than I have met from Swarthmore, Haverford, Grinnell. As a matter of fact, the number of entries on Bloomberg profiles that held an MBA that were from Amherst, Williams, Middlebury, Colgate were each 3-5x the number that were from Swarthmore. Bloomberg profiles are for those who use Bloomberg as part of their job, so this would certainly be “business job” related.</p>
<p>Okay, so the data is based on people who use Bloomberg as part of their job. So, we’re not just talking MBAs we are talking MBAs primarly working in finance/Wall St and probably more likely those engaged in trading. To me, that’s a signifcant fact. Certainly a far narrower group and starts to get into self-selction which may also be reflected in college choices. (In other words, your self-selection argument goes more to choices about finance/trading careers than just MBAs, generally.) Thanks for the clarification.</p>
<p>While you’ve justified why you’ve chosen the number 30, that doesn’t change the fact tht there is a logical alternative explaination to explain why certain schools (i.e., small ones) have less than 30 entries–other the the expainations you offered.</p>
<p>Also, FWIW, I don’t agree that you need 30 to do the analysis you’re seeking to do. An n=30 is only relevent to address the level of confidence that some sample group you pick of a larger group is represntative from within the population from which the samples are drawn. But, that’s not the type of analysis being done here. You are actually looking at 100% of the graduates in the DB from each school. </p>
<p>Those with Bloomberg profiles are a far wider group than just mostly traders; not to mention that most traders probably don’t have an MBA. There are probably much more research analysts and portfolio managers as well as those with C-level experience in corporate America. There are also a fair amount of law firm partners who have an MBA in addition to a JD on there. </p>
<p>
</p>
<p>I would think this is precisely why one would want to use a sample size of at least 30 in this particular example. Otherwise, certainly every top 50 LAC is going to have some absolute number of its grads who have who have a top 10 MBA. Requiring at least 30 entries to me would ensure that a representative sample for a specific school is presented as opposed to a cherry-picked list. However, this exercise wasn’t really meant to be a stickler to statistics or be biased against Swarthmore, which did show a high percentage going to top MBA programs, albeit much lower absolute numbers even after adjusting for class size. </p>
<p>I like the WSJ survey and wish they did it more than one time, which would make up for one-time randomness that the study allows for. I think schools like Colgate and Hamilton that show up as top 10 LACs for top MBAs but not top 50 for WSJ is mainly because they are disproportionately poor in having their grads get into top med schools, which brings down their overall score. However, that still doesn’t really explain how they are bested by what I would otherwise have thought are somewhat lesser schools like Macalaster, Trinity, Bates, Case Western, at least on an avg SAT, acceptance rate basis.</p>
<p>I am a Trinity College student. I think if Trinity didn’t do that much to isolate the data, it would be ranked a lot higher, which is also a good way to attract more talented students…</p>