The Ashley Madison Test of College Faculty Cluelessness

Science marches on. Recent developments have made it possible to answer the age-old question: What percentage of your college faculty used their work email addresses to establish accounts at Ashley Madison?

In a sample of 33 highly ranked colleges, the answer ranges from a low of 1.6% at Oberlin to a pretty much unbelievable 22.6% at the University of Minnesota. (The full rankings appear at the bottom of this post.)

I got these percentages by counting the number of unique email addresses ending with, say, “harvard.edu” in the leaked Ashley Madison database, and dividing by the number of Harvard faculty, as reported by the college either on its website or in its Common Data Set filings.

The methodology, is, of course, fraught with peril. First, the majority of academic email addresses belong not to faculty, but to students. But it seems like a good guess that faculty (by virtue of their average age) are both far more likely than students to be trolling Ashley Madison, and far more likely than students to be clueless about acquiring anonymous email addresses. Besides, a quick spotcheck of the email addresses in the Ashley Madison database does indeed confirm that most of them (at least in one small but random sample) belong to faculty members.

There are also, of course, staff, and the staff-to-faculty ratio probably varies a lot from school to school, so weeding out the staff could change the relative rankings quite a bit. But again, my (still small but still random) sample continues to indicate that these are mostly faculty members.

A far more important issue might arise from the fact that some universities have multiple campuses. It’s possible, for example, that the 657 umn.edu email addresses in Minnesota’s numerator came from many campuses, while the 2913 facuty in their denominator represents only the main campus. This will tend to inflate the rankings of the big state schools, and might account for the appearance of Minnesota, Virginia, Michigan and Cornell at the top — suggesting that if the numbers were crunched more carefully, the prize might go to Liberty University. If I were going to use these rankings for anything important, I’d give this issue a harder look.

One might also note that anybody can type anybody else’s email address into Ashley Madison, but I’m inclined to discount the importance of that, because it’s hard for me to see what the motive would be (except, perhaps, as part of a campaign to flood someone’s email box with unwanted replies).

With those caveats, feel free to use these rankings as a measure of your college faculty’s average cluelessness, at least when it comes to maintaining anonymity over the Internet.

For each college, this chart presents first the numerator (number of unique addresses appearing in the Ashley Madison database), then the denominator (number of faculty), and then the cluelessness score (numerator over denominator, expressed as a percentage):

ashley

Share/Save

32 Responses to “The Ashley Madison Test of College Faculty Cluelessness”


  1. 1 1 Rob Rawlings

    Wouldn’t the number of email addresses for a faculty be a function of both cluelessness and adulterous intent ?

  2. 2 2 Alan Wexelblat

    I suspect the data are further contaminated by local variations. Without any actual examination I can construct a narrative that says the rank of a school may vary with the size of its graduate student population. Another confounding factor is that many schools and universities provide lifetime/perpetual email access to graduates. I have an “alum.mit.edu” address myself that I use for almost nothing so I might be tempted to put it into a database on a questionable site. Schools that provide this lifetime email are likely to be overrepresented in this database, and thus distort the numbers.

    Also, I think it’s considerably more likely that students at certain institutions (e.g. MIT) would think it hilarious to put their professors’ email into such a site whereas students from other schools might favor less electronic forms of pranking.

  3. 3 3 Bob G

    Could the accounts also include alumni using University.edu addresses?

  4. 4 4 Steve Landsburg

    Bob G: I suppose they could. I have no idea how widespread this practice is. I imagine it varies a lot from one college to another, which is one more reason to discount these rankings.

  5. 5 5 Ross

    I rofl over seeing the big two Christian unis in the top ten.

  6. 6 6 David Wallin

    No, really. I just used the website to further my research…..

  7. 7 7 Sam

    22% of nearly /any/ population belong to Ashley Madison simply doesn’t pass the snicker test. Barring extraordinary evidence I’m going to say that result likely invalidates your methodology.

    Here’s a major flaw… I’m going to bet that the AM database contains all registrants since 2001. And it sounds like you’re comparing this against the /current/ number of faculty. For schools with high faculty turnover, you would need to apply a multiplier to your denominator based on the turnover rate. I’d wager that will change your results significantly.

    I’m a U Michigan grad. I find it likely that in any given year there are 100K active umich.edu email addresses and in the 14 years of data in AM’s database that’s probably closer to 500K. You found 3200 addresses in their database. There are plenty of adult students at Michigan. I wouldn’t find it surprising that they might use what they consider a temporary, throwaway account (their university email) for this purpose.

  8. 8 8 Phil

    Aren’t there more staff (janitors, coaches, accountants) than faculty?

  9. 9 9 Tom
  10. 10 10 John Faben

    @Sam (7): “22% of nearly /any/ population belong to Ashley Madison simply doesn’t pass the snicker test. Barring extraordinary evidence I’m going to say that result likely invalidates your methodology.”

    I think it’s quite plausible that at least 22% of the people who are signed up to Ashley Madison from a given university are signed up to Ashley Madison…

  11. 11 11 Roger

    These people are not clueless. They were just overly trusting. Having an extramarital affair always involves some risks.

  12. 12 12 nobody.really

    When picking a person to have an affair with, to what extent does social class matter? This question comes to mind as I compare U. Minnesota and Oberlin College.

    The number of potential mating partners for people at U. Minnesota is vast, especially at the Twin Cities campus, so the benefits of having a mechanism to flag likely candidates would be great.

    In contrast, Oberlin College is in a depressed little town in a cornfield “35 miles from Cleveland and 100 miles from civilization.” Faculty members discover that they cannot find any social equals to date besides each other. And there just aren’t that many of them. So the schools strive to attract “mated pairs” of faculty if they hope to get them to nest. In short, Ashley-Madison may not provide the same benefit to Oberlin faculty as to U. Minnesota faculty, since Oberlin faculty get introduced to all their likely mating partners at each faculty meeting.

    In sum, the difference in the behavior of faculty at Oberlin and U. Minnesota may not reflect the relative cluelessness about e-mail accounts, but rather relative expectations of utility for Ashley-Madison.

  13. 13 13 Blair

    Filtering out staff probably raises faculty numbers rather than lowering. Support staff know they’ll get the sack.

  14. 14 14 khodge

    How much of the low cluelessness factor at the University of Rochester can be attributed to the quality and influence of the Economics professors?

  15. 15 15 khodge

    #13 Blair, given the context, your second sentence is fraught with possible interpretations.

  16. 16 16 Ryan

    Most schools have as many staff that have the same email domain as faculty. Further, it’s likely not going to be current students or graduates in the last 8 years. The median age was 46 and hardly anyone under the age of 30.

  17. 17 17 Eli

    Cornell lets all grads keep their net-IDs forever at no charge. Thus, I’d guess that a large percentage of the Cornell e-mail addresses involved are alums rather than faculty/staff. It would probably be a good way to hide your “real” e-mail while still having assured forwarding. Liberty’s large number, likewise, probably comes from the fact that, measured by total enrollment, it’s actually the largest school on this list: it has truly massive online programs which are heavy with non-traditional students.

  18. 18 18 Anna

    What about your samples indicates these are staff/faculty rather than alumni or a combination of both? Liberty, for example, has a huge online program in addition to resident students who all receive .edu e-mail addresses.

  19. 19 19 Kirk

    On a different note, I wonder how many people on Ashley Madison are not actually cheating on someone, but just looking for a good time.

  20. 20 20 R.J.Smith

    This has to be one of the most wreckless examples of the disregard for truth that I have seen in recent times. In your race to shame educators, you didn’t even bother to pick up the phone to ask even ONE of these institutions who has access to an .edu address? Had you gone to that trouble, you would have immediately found that most schools extend that benefit to all students and alumni whether residential or online, yet you have attributed all Ashley Madison account with the .edu to faculty. How uncalled for it is to libel such an accomplished group when the facts were so easy to check!

  21. 21 21 Charlie

    Considering the difficulties with your methodology, you may want to consider removing this post. Do you really think that 22% of all faculty at Minnesota have signed up for an account? The Ashley Madison website isn’t even that well known; I bet less than 5% of the faculty had even heard of it prior to the data breach. The same could be said of all other schools on this list. Clearly the alumni addresses at these schools are diluting your numerator to such an extent as to make this analysis meaningless (or worse). An alternative to removing this post would be to take the time and look up all of the email addresses for a given school. For example, UVA has a high percentage and only 216 email addresses. A safe assumption is that all faculty members have their email addresses available through the university web pages at least once (probably on a list of faculty landing page). If you google each address, you will know which addresses are faculty. This would give you a better estimate of true faculty compared to your small random sampling method. Overall, please reconsider your analysis; it is probably unfairly representing the faculty of most of these institutions.

  22. 22 22 Steve Landsburg

    Charlie:

    Do you really think that 22% of all faculty at Minnesota have signed up for an account?

    If I really thought this, then I probably wouldn’t have called 22% “pretty much unbelievable”…. but let’s go on.

    Your suggestion that I (or someone) look up all 216 UVA addresses is harder to implement than you might think. In addition to the time cost, many universities offer only limited directory access to outsiders. But fortunately, the University of Rochester had only 51 addresses, and I have full directory access there, so I went ahead and looked them up.

    Result: Of 51 addresses, 4 are unambigously current faculty. Another 4 are students. 24 are not in the directory, suggesting that they’ve left the University. Most of the rest have ambiguous listings as something like Faculty/Staff. (I’m sure you could resolve some of these with a bit of Googling).

    As far as the rankings are concerned, the key question is whether you’d observe similar percentages at other univesities. This would take some effort. Clearly we need a grant.

  23. 23 23 nobody.really

    R.J. Smith and Charlie are exactly right; Landsburg’s allegations are slanderous. I know quite a few U. Minnesota profs and I can tell you without reservation: The suggestion that these intelligent people would spend hard-earned money in the vain attempt to induce someone to have sex with them is absurd. It just ain’t gonna happen.

    Dudes – IT’S A JOKE. Or what passes for a joke on this site, anyway. Landsburg has identified an apparent (and topical) anomaly – and from where I sit, it’s an amusing anomaly.

    If you google each address, you will know which addresses are faculty. This would give you a better estimate of true faculty compared to your small random sampling method.

    It would. And it would also be CREEPY AS HELL.

    Guys, if I say that Yalies have bigger dicks than Harvardites, people will simply smirk at my tastelessness. But if I pull out a tape measure, I’m gonna find myself on the business end of a lecture on Inappropriate Touching. Landburg’s post is funny BECAUSE the analysis is so slight, no one’s interests are threatened. It hints at something scurrilous, though we know it’s likely spurious.

  24. 24 24 Steve Landsburg

    nobody.really: Thanks. Your post makes me (at least slightly) regret that I went as far as I did in response to Charlie.

  25. 25 25 Charlie

    nobody.really – Thanks for insight. I didn’t realize it was a joke because the author didn’t write anything funny. There were many opportunities to add the wink/nod that this was all in jest which I think would have made this piece much more entertaining and reduced many of the critical comments related to this post. Re-reading in that light makes the piece fun, and I laughed trying to come up with alternate university mottos and mascots. Setting that tone early on would have helped.

  26. 26 26 Steve Landsburg

    Charlie:

    I didn’t realize it was a joke because the author didn’t write anything funny.

    I’m sorry I failed to amuse you. For what it’s worth, the above comment made me laugh out loud.

  27. 27 27 lauren

    how did you pick the universities used? why didn’t you use many of the top schools with AM accounts?

  28. 28 28 Greg

    Is there a way to find out other colleges lists? I am curious about checking the colleges/universities in my state.

  29. 29 29 Steve Landsburg

    lauren:

    why didn’t you use many of the top schools with AM accounts?

    Because there was absolutely no serious purpose to this exercise, I picked the first universities that came to mind.

    As far as “the top schools with AM accounts” — figuring out the identities of those schools would take more effort than I cared to invest. But the data set is quite widely available, so if this matters to you, have at it.

  30. 30 30 Steve Landsburg

    Greg: The data set is widely available.

  31. 31 31 RandomWords

    The moment I heard of Ashley Madison I asked myself how a site like this is going to work in light of the huge differences in male and female sexual attitudes and desires, especially concerning casual sex (http://psr.sagepub.com/content/5/3/242.short, http://www.tandfonline.com/doi/abs/10.1300/J056v02n01_04#.Vd8YryXtlBc).
    The answer is: It doesn’t work, at least not without massive cheating on the side of the website: http://gizmodo.com/almost-none-of-the-women-in-the-ashley-madison-database-1725558944

  32. 32 32 Dave

    Cornell alumni and employees all have Cornell email addresses and they keep them for life. Odds are very few of those 292 Cornell email addresses at Ashley Madison are current Cornell faculty members.

  1. 1 AshleyMadison Leak: A First Peek at the Data | M.L. Data Solutions
  2. 2 Apparently, His Faith Failed Him - Page 3 - PeachParts Mercedes ShopForum
  3. 3 Biglaw According to Ashley Madison - Windypundit
Comments are currently closed.