But what if all of my students are hot?


They are, every single one of them. Even the ones I don’t see because they’re just a black rectangle on Zoom. Apparently, though, attractive girls’ grades suffered when we moved to online courses because they couldn’t appeal to professor’s biases.

It’s a garbage study, though, as Rebecca Watson explains. The paper claims that…

As education moved online following the onset of the pandemic, the grades of attractive female students deteriorated. This finding implies that the female beauty premium observed when education is in-person is likely to be chiefly a consequence of discrimination. On the contrary, for male students, there was still a significant beauty premium even after the introduction of online teaching. The latter finding suggests that for males in particular, beauty can be a productivity-enhancing attribute.

I don’t understand the mechanism behind that — so we have some kind of radar that senses hot men even over wi-fi, but that fails when we try to detect hot women? How is “beauty” a productivity-enhancing attribute?

Did the author consider the possibility that all of our students and professors have been experiencing great strains over the last few years? Deciding that the one decisive parameter was what they look like seems exceptionally reductive.

Then I had to wonder how the scored “beauty”, and it turns out the author just scavenged up photos on social media, had a couple of students look at them, and rate them. This seems rather arbitrary, and dependent on biases by the judges, as well as accidents of photography. I know I hate it when people take candid shots of my face before I’ve put my makeup on, and also, I don’t know about you, but I automatically deduct 2 points from any photo in which the subject is making pouty duck lips. Sorry.

Final gross error: he included enough detail about the subjects that they could in some cases tell what their score was…and their grades. Oh, and big problem, there was no informed consent, none of the students knew they had been incorporated into this “study”.

The author, Adrian Mehic, is an economist, so I’d already be suspicious of his psychological/sociological study, but the ethics violations and the ridiculous conclusion he draws (“attractive women get better grades because they’re being unfairly advantaged”) confirms that this is a dumpster fire of a paper, constructed out of a thoroughly p-hacked grab bag of fuzzy data.

Comments

  1. leerudolph says

    I automatically deduct 2 points from any photo in which the subject is making pouty duck lips.

    That accounts for the otherwise inexplicable lack of pouty duck lips in your arachnid pin-up galleries.

  2. drsteve says

    I always assumed that most if not all college students are somewhere in the high 30s Celsius, and as such are quite hot.

  3. says

    I’ve found that Rebecca Watson (skepchick) usually provides interesting information and sound reasoning in her articles. And, even if there is bias based on how pretty a student is, instructors are who should responsible for being factual and impartial.
    I’ve mentioned this before. Traditional music auditions use a screen between the performer and the judges. This prevents bias based on observed characteristics such as gender, ethnicity, dress, etc. Which should tell us that if the ‘hot girls’ can’t be seen that could help eliminate bias in some circumstances where an instructor might let ‘hormone driven’ bias sway him/her.

  4. raven says

    I automatically deduct 2 points from any photo in which the subject is making pouty duck lips.

    How many points does he deduct from photos in which the subject doesn’t have 8 legs?

  5. says

    Well, he’s an economist, right? They tend to believe in weird things–invisible hands and rational actors and shit. Stuff that requires years of training to take seriously. They are not to be trusted with designing scientific studies.
    What they’re good at is explaining why real-world results that defied their predictions still support their theories.

  6. UnknownEric the Apostate says

    so we have some kind of radar that senses hot men even over wi-fi

    That’s what those stud finders at Home Depot are for.

  7. robro says

    feralboy12 @ #6 — “Well, he’s an economist, right?” I was immediately reminded that the psychologist Daniel Kahneman won the 2002 Nobel in economics because his research with Amos Tversky demonstrated that one of the foundational principles of economics…that markets behave rationally…is flawed at best.

  8. lotharloo says

    Haha, this looks like a case “torture the data until it confesses”:

    For instance, let me quote from the study (emphasis mine): “The results suggest that the switch to online learning did not result in an overall deterioration of the grades of high-attractive students.” Huh, well that’s weird! In fact, the only way they could find a statistically significant result was to look specifically at high-attractive WOMEN, and even then they didn’t get it. To get a result, they had to look at women who were taking non-quantitative courses. Mehic says he decided to break up classes into two groups:

    BTW, I have an alternative hypothesis, plz help me publication: when classes go online, girls with hot sounding names should get better grades, because obviously the instructors can’t see their faces, only names, duh. You might be thinking this hypothesis will not go anywhere cuz one can’t determine if a name is hot. But you would be wrong! The solution is obviously to google the first names of the students with “safe search off”, then grab the first picture that includes a female human being and pass it to a “neutral” panel of judges consisting of my bros and their football coaches; if the first picture is too old or something, try the second picture until a satisfactory picture is found. Then, we collect some data, pass it through a combination of python scripts and codes I found online, add some machine learning stuff to make it sound cool and by then hopefully the paper is ready for publication at whatever premier venue that this study got published.

  9. nomdeplume says

    This is like creationism – conclusion reached before research. “Hey, attractive girls, OBVIOUSLY get better grades during in-person teaching. D’Oh.” So you just throw some stuff together which will “confirm” this.

  10. birgerjohansson says

    What if students get lower grades because they are unable to physically intimidate the staff online? That seems a perfectly valid question but he did not adress it. Also, students with a strong BO should get better grades online. There are probably more potential issues he should investigate.
    Alien shape-shifters with ESP will find it harder to control the minds of the professors if they are not in the same building, so they should get lower grades.

  11. Pierce R. Butler says

    But what if all of my students are hot?

    Then you get to grade on the curves. Ba-dum-tish!

  12. gijoel says

    It wouldn’t have done much good in my last degree as 90% of the lecturers were married (straight) women. I bet if we find Adrian’s social media accounts they’ll be filled with rants against “Stacys” and “Chads”.

  13. jrkrideau says

    Received 23 June 2022, Revised 25 July 2022, Accepted 1 August 2022, Available online 6 August 2022

    I don’t know about publishing times in economics but that seems fast to me.

    lotharloo @9 seems to have a point. That data appears to have been flogged.

    I also note he gets a Cronbach alpha of 0.94 but does not mention any coder training. It’s a weird paper. Maybe having each photo rated by 37 raters smooths out rater variability? Essentially, having two sets of raters seems strange too.

    I cannot see it passing peer review in a respectable psych journal.

  14. microraptor says

    If all your students are hot, it indicates that your classroom has insufficient air circulation.

  15. chrislawson says

    shermanj@3–

    Please don’t take this as a personal criticism — it’s more a jumping off point for a savaging of the paper — but blind orchestra auditions were introduced not because “hot” women were getting unfair advantages but because women weren’t getting selected regardless of ability (or “hotness” btw).

    This entire paper is sexist bullshit from a reactionary economist. The poor design is merely the self-deludory mechanism necessary to maintain his worldview, and while it is important to point out the design flaws, one should not do so while accepting the awful premises of the author (an argument straight out of the incel/angryman playbook) that attractive women have an institutional advantage when studying engineering (I mean, this alone is fucking hilarious given engineering is by far and away most male-skewed discipline at university level, and around 40% of female grads quit young or never start working in engineering because of the sexist and anti-progressive behaviours they encounter).

    I looked at the paper, and I can state unequivocally that it is utter crap, written by someone with no expertise in social research, teaching or psychology, published in an inappropriate journal that should have known better than to publish it.

    Rebecca Watson already covered lots of the problems and I thoroughly commend her analysis so will only briefly relist them: the trial was unethical, the trial was not pre-registered, the author deliberately skewed the attractiveness of the photos he chose, the author fudges his categories for the purpose of justifying his p-hacking, the author justifies female beauty bias as discriminatory, but male beauty bias is because attractive men are better engineers — seriously!

    But there are other egregious errors:

    As previously mentioned, why of all the university disciplines to study was engineering chosen? It is known to be the most discriminatory against women. The author is not an engineer, so it can’t be that he was interested in his own field. He chose it, of all the possible faculties, because he wanted to normalise sexist discrimination and bury it in his study baseline. I can’t think of any reasonable explanation for this decision although I’m happy to be corrected if someone else can provide a plausible defence. Certainly no explanation is given within the text of the paper.

    “…the switch to online teaching during the pandemic enables us to more credibly isolate the effect of appearance. This is because only the mode of instruction changed, and not the structure of the courses.”

    Every university had to make major adjustments to structure to accommodate online teaching. Ask any uni teacher anywhere who had to switch online if there was no change in structure and they will laugh in your face.

    “For all courses, passing grades are given by 3, 4, and 5, where 5 is the top grade. The grading scale is absolute, meaning that the cutoff level for each grade is determined before the start of the course, and is not affected by the relative performance of students.”

    Again, this shows that he knows absolutely fucking nothing about how students are graded and the role of bias within even supposedly “absolute” grading scales. Which is kind of important if you’re going to write a paper on how bias affects grading. The odd thing is…Mehic is a teacher at the university where this study was performed. He must know how this works, and the limitations. Which can only mean he has let his biases interfere with his own experiential knowledge just so that he can complain about how unfair it is competing with pretty women.

    And how unfair? Well, according to his own analysis, when choosing the biggest effect size via subgroup analysis, a 1-sigma difference in attractiveness led to a 0.08-sigma difference in grades. Yes, 0.08. This is fucking trivial, even if it were a reliable finding (which it almost certainly is not). And remember, sigma is an exponential function, so this is even more inconsequential than it looks. It is the equivalent of an IQ difference of 100 compared to 101 (not that IQ is a good test, but it also is based on the normal distribution function, so it’s a good way of demonstrating how pathetically small this effect size is). In epidemiology, anyone reviewing a paper with this effect size would describe it as not practically meaningful.

    And although I said I wasn’t going to repeat what Watson covered, I’m going to make one exception with direct quotation because it really needs to be hammered home:

    “…switching to full online teaching resulted in deteriorated grades in non-quantitative courses for attractive females. However, there was still a significant beauty premium for attractive males. Taken together, these findings suggest that the return to facial beauty is likely to be primarily due to discrimination for females, and the result of a productive trait for males.”

    So women’s beauty premium is a discriminatory bias, but men’s beauty premium is because attractive men are better at engineering. And no, this is not just a single line. He repeats it in his conclusion.

    “This finding implies that the female beauty premium observed when education is in-person is likely to be chiefly a consequence of discrimination. On the contrary, for male students, there was still a significant beauty premium even after the introduction of online teaching. The latter finding suggests that for males in particular, beauty can be a productivity-enhancing attribute.”

    There’s even more, but I think we have long reached the point of diminishing returns.

    Seriously, fuck the author. Fuck the editors who published the paper. And if Lund doesn’t sack him for unethical research on their own student base, fuck the administrators. (Crappy research is protected by academic freedom, unethical research isn’t.)

  16. whheydt says

    Re: chrislawson @ #16…
    When I was an engineering student, classes were graded on a curve. The exception (of those I took) wasn’t engineering. It was a two-quarter sequence on navigation taught by the Dept. of Naval Science and the US Navy lieutenant that taught them explained why.

  17. justanotherguy says

    We centrist liberals just cleaned up the mess you Progressives helped create in 2016 by voting for Bernie Sanders (in the primaries) and Jill Stein (in the general election).

    Any Progressives out there still believe the two major parties are the same?

    The far-left mistook decades of the two major parties cancelling each other out for the two major parties being the same.

    While it is true that the centrist politicians of both major parties are too beholden to corporate interests, the politicians in the wings of each party couldn’t be more different.

    And those are the politicians who’s followers tend to dominate in the primaries more and more as time goes on.

    Don’t do that to us centrists again, Progressives. We must all hang together, or we will surely hang separately.

  18. lochaber says

    @18

    fuck off out of here with your centrist bullshit.

    A lot of centrists voted against the Democratic party, because that’s what centrists do, they vote against the incumbent in midterms. And against the incumbent when the economy isn’t great, like, say, recent high inflation, or expensive gasoline.

    It was mostly fed-up Zoomers and Millenials, who have had to deal with the worst fallout of Boomer Reaganomics, as well as a lot of women (and some men) who place a high value on bodily autonomy.

    US “centrists” are considered “far right” by most metrics of almost any other wealthy nation. The U.S. is a ridiculously wealthy back-asswards abomination built of regression and foot-shooting “principles”.

    Only in the U.S., is state-funded healthcare considered an extremist, far-left-wing, “threat to freedom”

    Just to top it off, two words: Sinema, Manchin

  19. says

    Here’s another wrinkle to this thesis: even if it can be shown that prettier/”hotter” women get better grades, that’s not necessarily because their teachers just mindlessly gave better grades to whoever they got the hots for. Another possibility is that the prettier students were better able to engage with their teachers, and work more effectively with them to help themselves to understand the course material, and thereby earn better grades honestly by knowing the material better. That’s still an unfair advantage, if it’s happening at all; but it’s a different unfair advantage from the one this paper alleges. And those students who depend most on one-to-one interaction with their teachers would suffer most from a sudden shift to remote-meeting classes, whether or not they were good-looking.

    Also, this is more about charisma, which is not the same as visual beauty; but those two things go together a lot. In any case, this is just one more thing this paper failed to address.

  20. StevoR says

    @ 18. justanotherguy : The USA has an actual “far left” now? Inside the centrist to centre-left at best Democratic party especially? Since when?

    Any Progressives out there still believe the two major parties are the same?

    Quite probly so, yes.

    (Vicar? This could be your cue here? If Vicar actually counts as progressive?)

    How do you think they can best be persuaded otherwise?
    Is it possible the two parties may be – if not the same then at least too close together and similar in policies?
    Does it matter if the Progressives think the two parties are too similar if they vote Democratic despite thinking that and work to make the Democratic party more leftwingand progressive generally?

    The far-left mistook decades of the two major parties cancelling each other out for the two major parties being the same.

    Oh did they now? You citations & supporting evidence to show this to be the case would be? For that matter, your definition of the “far-left” and how it differs from the left and centre here would be _____?

    While it is true that the centrist politicians of both major parties are too beholden to corporate interests, the politicians in the wings of each party couldn’t be more different.

    Actually, they could be a lot more different.

    For example, more Democratic party politicians could adopt the policy platorms and rhetoric and asome of thearguments and cases put by say the Australian Greens party :

    https://greens.org.au/principles

    Or the German Greens :

    https://www.gruene.de/artikel/das-neue-grundsatzprogramm (may be initially in German, English translation should be available.)

    Or well, a whole lot of others not just limited to Europe of course but see for instances :

    https://en.wikipedia.org/wiki/List_of_left_and_far-left_parties_in_Europe

    And those are the politicians who’s followers tend to dominate in the primaries more and more as time goes on.

    Which are who now and what?

    Don’t do that to us centrists Preogressives again, Progressives. Centrists. We must all hang together, or we will surely hang separately.

    Fixed It for You?

    @19. chrislawson : “@18. OT and self-entitled in the one post!”

    Agreed. Could be a mistaken thread i.e. posting in the wrong thread error or trolling maybe. Guess we’ll find out soon enough. Missed a few days here due to power outage at my palce -amnd much of Adelaide with storm damage recently but don’t recall seeing that nymn around much. Could be my memory of course but anyone vouch for / against that commenter here posting on other threads and constructively or not?

  21. John Morales says

    StevoR:

    (Vicar? This could be your cue here? If Vicar actually counts as progressive?)

    <snicker>

    That’s your appeal? Really?

    I can answer for The Vicar: it’s Bill Clinton’s fault; it’s Obama’s; it’s Hillary’s; it’s Bidens. It’s the Democrats not doing their job, is what it is.

    Simple as that.

  22. macallan says

    As my professor used to say – economists are the best at math, they can differentiate non-contiguous functions.

  23. StevoR says

    I wonder if #18 intended /thought they were posting on this thread :

    https://freethoughtblogs.com/pharyngula/2022/11/14/wow-i-guess-that-last-election-was-more-devastating-than-i-thought/

    Or anotherpolitics /midterms related one rather than, well, this one?

    @23. John Morales : “That’s your appeal? Really?”

    No. That was my example in response tothe question “Any Progressives out there still believe the two major parties are the same?” Thanks for helping to prove my point albiet the uncertainty of whether Vicar really counts as a or a typical Progressive or not. Clearly there’s at least one person who claims both parties are the same or at least blames the Democratic party more than the Repub one putatively for being too much like it as the gist of the Vicar’s case seems to be.

    Note I am specifically NOT saying that Vicar is eirher representative or all or many Progressives nor that I agree. I don’t.

    @15. microraptor : “If all your students are hot, it indicates that your classroom has insufficient air circulation.”

    Or insuffiicient shade or cooling incases of outdoor classes in hot environments eg. Summerinthe Bush, hot deserts, svannah, Veldt etc..

  24. John Morales says

    StevoR:

    Thanks for helping to prove my point albiet the uncertainty of whether Vicar really counts as a or a typical Progressive or not.



    Um, you are welcome for my… um, help. Always happy to help.

    Not that I think you actually got what I intimated, but still, your gratitude is hereby acknowledged.

    (BTW, what exactly was this point of yours that I supposedly helped to prove?)

  25. StevoR says

    @ ^ John Morales : Thought I’d already explained that & that it was pretty clear – but, again, #18 justanotherguy asked :

    “Any Progressives out there still believe the two major parties are the same?”

    To which my response is : Yes. Quite probably.

    Example (exhibit A?) The Vicar who always blames the Democrats and accuses them of being equally as bad as Repugs. (Or nearly so?)

    Offering the Vicar (& like -minded Progressives / commenters here) their cue to confirm or deny that and you citing the Vicar as saying pretty much what I gather Vicar is saying.

    The only question here being whether we put the Vicar in the category of being a Progressive or another different category.

    As noted in #25 I do NOT agree with the Vicar here and nor do I believe the Vicar is representative or most or even very many Progressives however there area few misguided Progressive incl the Vicar and some old Berniebros who from what I understand DO hold the position that # 18 justanotherguy (JAG) accusingly & mistakenly appears to associate with all Progressives.

    Inside the metaphorical shell of a nut; I think JAG thinks all or most of us agree with what The Vicar thinks which we mostly don’t. Certainly JAG asked if any Progressives think as the Vicar seems to think* to which the answwr is yes because the Vicar thinks that way (& there area coupelof others who sometimes make the same or similar arguments it seems.)

    .* Thinking wrongly that both parties are the same or very much too similar..

  26. StevoR says

    To clarify my #27^ :

    ..nor do I believe the Vicar is representative or of most or even very many Progressives however there area few misguided Progressive incl the Vicar and some old Berniebros who from what I understand DO hold the position..

    Because I really suck at both typing and seeing my typos before hitting send. /Cap’n Obvs.

  27. chris61 says

    @16chrislawson
    He’s a graduate student. Personally I cut him a bit of slack for that. And he chose the first two years of engineering because all the courses being mandatory, all the students took the same courses.

  28. jimzy says

    @PZ Myers: But what if all of my students are hot?
    I didn’t know you ran the North Korean reactor operator’s training school.

  29. jrkrideau says

    @ 16 chrislawson
    I looked at the paper, and I can state unequivocally that it is utter crap, written by someone with no expertise in social research, teaching or psychology, published in an inappropriate journal that should have known better than to publish it.
    Elsevier seems to have some very dodgy journals recently, Could this be one?

    @ 18 justanotherguy
    You may have missed it but Sweden is not a US state.

  30. chrislawson says

    chris61@29–

    Bullshit#1. Being a doctoral student is no reason to cut him slack on grossly unethical, poorly designed research that is falsely presented as supportive of sexist beliefs. At best, it co-incriminates his academic supervisors.

    Bullshit#2. Engineering is far from the only discipline where the subjects are the same for all students. At Lund itself, undergrad Biomedicine is a 3-year course where every student has to complete every subject. And even for courses where there are some elective choices, he could have simply excluded those subjects if he wanted to avoid self-selection bias. For instance, in his own discipline of economics the Lund program only has one elective semester out of six. So why would he ignore courses like Biomedicine and Economics in favour of Engineering…could it be because they have roughly 1:1 ratio of male:female enrolments and are known to be much more supportive environments for female students?

    Really, chris61, I don’t understand why you keep turning up here whiteknighting regressive asses with piss-poor defences that fall apart like wet tissue on the slightest investigation.

  31. chrislawson says

    jkr@31–

    Elsevier has been running the predatory journal game for quite some time, it’s just been getting more heat for it in recent years. Having said that, a quick glance at this journal’s website shows a list of papers that look good (to a non-expert eye), a modest but reasonable impact factor, and a semi-open publishing policy. So I expect this is an editorial slip-up rather than a fundamentally untrustworthy journal. (And it’s worth reminding that even the most prestigious journals have published woefully bad papers on occasion.)

Leave a Reply