Skip to content
 

Teeth are the only bones that show

“I lived in the country where the dead wood aches, in a house made of stone and a thousand mistakes”The Drones

Sometimes it’s cold and grey and Canadian outside and the procrastination hits hard. Sometimes, in those dark moments, one is tempted to fire up the social media and see what’s happening in other places where it’s probably also cold and grey, but possibly not Canadian. As if change were possible.

And sometimes you see something on the social media that annoys you.

In this case, it was a Pink News article titled “Men with muscles and money are more attractive to gay men, new study finds”.

Now, I curate my social media fairly heavily: I did not get into this liberal bubble by accident and I have no interest in an uncurated life. So I didn’t see this from some sort of Pink News feed, but rather from some stranger saying (and I paraphrase) “this is crap”.

A quick sidebar

Really? A study shows men with muscles and money are more attractive to gay men? Really?

For a community of people who are nominally dedicated to performing masculinity in all its varied, rainbow forms, the gay media is pretty much exclusively devoted to good looking, often straight, rich, often white men who are so dehydrated they have fabulous abs.

To paraphrase Ms Justin Elizabeth Sayer, “I know you find it hard to believe, with all the things you could be as a gay person, boring would still be an option. But it is.”.

We don’t need new research for this, we could just look at the Pink News website. (Please don’t. It’s not worth the pain.)

A journey in three parts: Pink News

The best thing that I can say about the Pink News article is that it didn’t regurgitate the press release.  It instead used it as a base to improvise a gay spin around. Pink News: the jazz of news.

It’s immediately clear from this article that the findings are based around the content of only one (creepy) website called TubeCrush, where you can upload photos taken of unsuspecting men on public transport. So a really classy establishment. (I mean, there are not enough words to get into just how ugly this is as a concept, so let’s just take that as read.)

The idea was that because the photos on this site focus on ripped white men who often have visual signifiers of wealth, gay men are attracted to ripped, rich, white men.

Of course, this excited me immediately. The researchers are making broad-ranging conclusions from a single, niche website that fetishizes a particular lack of consent.

Why is this exciting? Well I’m interested in the way that the quantitative training that scientists (both of the social and antisocial bent) have received doesn’t appear to have made the point that an inference can only ever be as sharp as the data used to make it.  Problems come when you, to paraphrase the Mother Superior in Sister Act 2: Back in the Habit, try to make a silk purse out of a sow’s ear.

That terrible gayface study from a few months ago was a great example of how quantitative social science can fall apart due to biases in the data collection, and this looked like a great example of how qualitative social science can do the same thing.

A journey in three parts: The press release

Coventry University, for reasons best known only to themselves, decided that the revelation that pervy website TubeCrush had a tendency to feature ripped, rich white men was worth a press release.  Here it is.

The key sentence, which Pink News ran with, was

The researchers say their study of the entries on the site quickly revealed that both straight women’s and gay men’s desired particular types of men.

It goes on to talk about how, despite London being multicultural, most of the photos were of white men. Most of the men were good looking and muscled. Most of the commentary on TubeCrush mentioned the muscles, as well as the expensive suits, watches, and phones the men were wearing/using.

Why is all this important? Well the press release tell us:

The academics said that public transport – in this case the Tube – has now become the space where gender politics is decided.

How fascinating. Not in schools, or universities. Not in families or friendship circles. Not at work or in social spaces. On. The. Tube.

The press release ends with a few quotes from the lead researcher Adrienne Evans, which is a useful thing for journalists who need to pretend they’ve talked with the authors. I’m just going to cherry pick the last one, but you should get the point (or follow the link above):

“It’s a problem as because although it appears as though we have moved forward, our desires are still mostly about money and strength.”

That is a heavy extrapolation of the available data.

A journey in three parts: The paper

Finally, I decided to read the paper. If you’re going to do a deep dive methodological criticism instead of doing the pile of things that you are actually supposed to be spending the afternoon on, then you should probably read the actual paper.

So here is the paper, entitled “He’s a total TubeCrush”: post-feminist sensibility as intimate publics by Adrienne Evans and Sarah Riley, published in Feminist Media Studies.

Now I have no expert knowledge of feminist studies (media or otherwise).  I’ve done the basic readings, but I am essentially ignorant. But I’m going to attempt to parse the title anyway.

An intimate public appears to be the idea that you can leverage a private, shared community identity into a market. The example in the paper is “chick lit”, which takes shared, performed aspects of femininity and basically markets products towards them. Arguably another example would be Pink News and the gay media industry. The key feature is that “Intimate publics do not just orient desire toward traditional gender roles, but evoke a nostalgic stance toward them, so that one element of the intimate public is feelings of nostalgia.”. So an intimate public in this context will always throw back to some supposed halcyon days of peak masculinity and celebrate this by marketing it, as TubeCrush does. So far so good.

The post-feminist sensibility is basically the idea that post-feminist media simultaneously sell the idea of equality while also expelling that you can achieve equality through consumption.  (I mean, they use a lot more words than that, so I’m probably missing nuance, but that seems to be the thrust).

So the main argument (but less than half the paper) is devoted to the idea that TubeCrush is an example of post-feminist intimate publics. (a post-feminist intimate publics? The grammar of academic feminism is confusing)

All well and good. (And probably quite interesting if you’re into that sort of thing).

Also, as far as I’m concerned, completely kosher.  They are studying an object (TubeCrush), they are categorizing it, and they are unpicking the consequences of that categorization. That is essentially what academics do. We write paragraphs like:

TubeCrush makes masculinity a bodily property in much the same way as femininity is within post-feminist sensibility (Gill 2007). The aesthetic idealization of strength in the posts can be tied both to the heightened visibility of masculinity more generally, and to its location within an “attraction to-” culture. The intersection of heterosexual women’s and gay men’s desire arguably heightens the emphasis on strength as a key component of post-feminist masculinity, where gay male culture holds up heterosexual male strength as part of its own visual landscape (Duane Duncan 2007, 2010). In a culture that only very recently was not visible at all, effeminacy is disparaged and what is celebrated are “visible public identities that [have] more in common with traditional images of masculinity” (Duncan 2007, 334). In this way, representations of masculinity on TubeCrush demonstrate the maintenance of hegemonic masculinity, tied into notions of strength and phallic power.

Ah but there it is. They extrapolated from limited available data (a deep read of TubeCrush) to population dynamics.

On the maintenance of hegemonic data practices in social sciences

(Sorry)

And this is the crux. Moving from this to the press release to that awful Pink News article is a straight ride into hell on a sleigh made of good intentions.

Statements like “TubeCrush is a reestablishment of traditional gender roles within the context of post-feminism” are perfectly reasonable outcomes from this study. They take the data and summarize it in a way and sheds light on the underlying process.  But to move the conclusions beyond the seedy TubeCrush universe is problematic.

And the researchers are aware of this (a smarter blogger could probably make links here with the post-feminist sensibility in that they also generalize with their data while admitting that you need to buy the generalization through more data collection. But the analogy doesn’t completely hold up.)

While there was no rigid research design prior to funding, we believe this extended engagement with a website (whose materials only go back as far as 2011) has provided an in-depth understanding of the patterns and content of TubeCrush.

The first challenge when generalizing these results off the TubeCrush platform is that the study is not designed. I would term this a “pilot study”. I guess you can write press releases about pilot studies, but you probably shouldn’t write press releases that don’t mention the limitations of the data.

My major criticism of this article is that it treats TubeCrush, the photographs and the comments as an entity that just exists outside of any context. This is a website that is put together by a person (or a team of people) who select photographs and caption photographs. So these photographs and comments will always be a weak instrument for understanding society.  They are not irrelevant: the website is apparently quite popular. But the data has been filtered through a very limited sensibility. It reflects both the ideals of attractiveness and the sense of humour (or “humour”, I’ve not actually visited the site because I don’t need that in my life) of the curator(s).

TubeCrush makes it a weak tool for understanding society and all inferences built using only this data will necessarily be weak.

And I guess this is my broad point. Quantitative thinking  around the data gathering, the design of the experiment, and the limitations of the measurements can inform the reliability of a qualitative study.

Is this paper bad? No. Was the article? Yes. Was the press release? Definitely.

But what sort of cold-hearted person doesn’t love a paper with the sentence

The wordplay on the financial language of the double-dip recession to again signify performing oral sex on financially secure (if not wealthy) masculinities demonstrates the juxtapolitics at the heart of TubeCrush.

31 Comments

  1. Jonathan says:

    I was once pointed to a study that said there are no gay black men, that some black men have had their gender identity shaped by racism to make them gay. It was by a woman. I don’t know if she is gay, because that would be a hoot. Seems like nearly all these things are attempts to shoehorn an idea on to a version of fact, which reflects the way that social science has, to a material degree, been transformed into literature. So much of it is now the creative fitting of facts to tell the story you want to tell. I guess that would make it truly personal ‘science’!

  2. Morgan says:

    This is just standard grievance-mongering.

    “Look! I’ve found a way to interpret some obscure website’s content as grievance-worthy!”

    Good for you. Glad you found a way to feel aggrieved.

    But not really.

    • Dan Simpson says:

      I think their paper is a lot more nuanced than that.

      And the website isn’t all that obscure.

      These are important questions that could be answered better. But it’s good that people are answering them.

    • Andrew says:

      Morgan:

      On one hand, I respect your general point that we shouldn’t be spending all our time looking for examples of “someone is wrong on the internet.” But I’m also wary of reflexive anti-critical attitudes. If someone wants to post an article on a hot topic on a public forum such as Pink News (or, for that matter, the Proceedings of the National Academy of Sciences), they should accept that people might evaluate their work critically. Indeed, I think it shows a deep form of respect to people to point out problems with what they’ve written. “Aggrieved” is kind of a loaded word, as it has this implication of resentment.

      Why not just accept that Dan Simpson has a combination of statistical expertise, thoughtfulness, and interest in the topic, which allows him to contribute usefully to such discussions? The alternative, it seems to me, is people trying to enforce a sort of incumbency rule whereby the first people to get publicity writing about a particular topic get to determine the agenda and decide whether criticism is allowed—or whether it will be disallowed by calling it “bullying” or “aggrieved” or whatever. I’ve never heard of Pink News before reading the above post, and I know nothing about the author of that linked news article—but I’d hope that if he happens to come across some criticism, however phrased, he’ll make good use of it. And that others who are interested in these topics will get something out of Dan’s discussion.

      • Dan Simpson says:

        Ah I totally misunderstood this comment! I don’t think Pink News is obscure. Nor do I feel aggrieved. But I want to reflect on what the quant side of social science can bring to the quals. I’m sorry if that’s not how it came across. Blogs are what they are, I guess. And prone to poor expression of important points is one of the things that blogs are.

      • Dan Simpson says:

        A quick check shows Pink News is British. But the social media account I learnt of this from was very American. So I guess if you’re straight and/or not English then this would be obscure. But I think I’ve made clear that I’m not straight, and as it turns out I spent the last few years in the UK, so I don’t think it’s obscure.

  3. Dan Simpson says:

    I didn’t really expect the comments here to become a to and fro about feminist theory and qualitative social science. So let me just say that I have infinite respect for these fields. I have learnt more from intersectional feminists, transfeminists, and queer theorists than I ever have from statisticians (which is just maths I can get to on my own if I’ve got time). The critique here is of generalisation bias and not of their place as useful scholars adding to our understanding of society.

    • Anonymous says:

      From what you wrote here, i assume you may know a bit about what i am about to ask. I hope you don’t mind me asking this, and most of all i hope i will not offend anyone by saying the following:

      Ever since i learned about possible different “types” in gay culture/relationships/whatever a few years ago, i wondered if there has ever been research done about these possible types, and if they can be related to other certain possible types these types are (mostly) attracted to, sexual roles, and things like life histories.

      I can totally see there might be possible connections between these types of variables. Do you happen to know if such research has ever been performed? And if not, would that be ethical, and possibly interesting?

      • Dan Simpson says:

        Unfortunately I don’t know anything about this. I’d be hugely surprised if this wasn’t studied (social scientists love a taxonomy).

        There is a wealth of survey data on, at least, sexual practices, but the stuff I’m aware of mainly focusses on public health aspects.

        Sorry I can’t be more help!

  4. A.G.McDowell says:

    “How fascinating. Not in schools, or universities. Not in families or friendship circles. Not at work or in social spaces. On. The. Tube.” Schools, Universities, and especially work have rules and consequences about what constitutes acceptable interactions between attracted and attractor – and I expect the consequences to become more severe in future. I am not surprised to find people looking at others in an environment where there is little else to do and the rules less stringently enforced than in a white collar workplace, school, or university.

    • Dan Simpson says:

      It’s a bit of a grandiose claim, though. I totally agree that gender politics is enacted in an untethered way on public transport (I mean, get on a night bus in London), but it’s a gigantic stretch to say that gender politics is decided there!

  5. Anonymous says:

    I am a fan of the tv show “Portlandia”, so whenever i hear or read about feminists all i can think of this:

    https://www.youtube.com/watch?v=IrcGeSmEdC4

    00.00-03.41 is my favorite scene.

  6. Anonymous says:

    The Tubecrush paper was also mentioned on real peer review twitter: https://twitter.com/realpeerreview?lang=en

    This is another one that was mentioned there, which might be my favorite:

    http://www.diva-portal.org/smash/record.jsf?pid=diva2%3A565774&dswid=-7523

    “My main argument is that, since what we normally (can) expect from women and men differs, what tends to evoke gratitude, appreciation and love is also gendered. Hence, even if an individual man practices reciprocity in love, he will still be structurally advantaged to the extent that his behaviour implies a positive break with what can be expected from men in general. This tends to make the woman more grateful than the man – despite the actual symmetry – and thereby the asymmetry is paradoxically reinstated.”

  7. Anonymous says:

    “The researchers are making broad-ranging conclusions from a single, niche website that fetishizes a particular lack of consent.”

    &

    “Is this paper bad? No. Was the article? Yes. Was the press release? Definitely.”

    As all viewers of (straight or gay or whatever) porn websites know, there are lots of “categories” to choose from (which i find to be very helpful!)

    Considering the gay content, here is a quick overview of some possibilities of different “types” people may find attractive/arousing/whatever from the tv show “modern family”:

    https://www.yahoo.com/entertainment/modern-family-sneak-peek-jay-plays-gay-for-a-117042634710.html

    If the researchers would have focused on a different site which focuses on other “types”/”categories”/whatever, they could have come to a totally different description and/or conclusion i would reason.

    If this makes any sense, i would reason that this paper *is* bad, as it makes sweeping statements from a very narrow focus/perspective/sample/whatever. As such, it is precisely what i hate about this type of social “science”, and precisely what another commenter above so clearly described when he stated:

    “Seems like nearly all these things are attempts to shoehorn an idea on to a version of fact, which reflects the way that social science has, to a material degree, been transformed into literature. So much of it is now the creative fitting of facts to tell the story you want to tell.”

    • Anonymous says:

      “Seems like nearly all these things are attempts to shoehorn an idea on to a version of fact, which reflects the way that social science has, to a material degree, been transformed into literature. So much of it is now the creative fitting of facts to tell the story you want to tell.”

      This actually makes for a fun exercise.

      Given my previous post about different “types”/”categories”, i will try to give an example using the type “chub”: fat guys of whom you could say have “feminine” characteristics like moobs, big backsides, being “soft” overall, and without much body hair.

      I will assume Tubecrush could just as well mostly depict these types of men, and i will use the paragraph of the paper in question (depicted above in the original post) but adapt it to come to just about the opposite description/conclusion:

      “TubeCrush makes femininity a bodily property in much the same way as masculinity is within post-feminist sensibility (Gill 2007). The aesthetic idealization of “softness” in the posts can be tied both to the heightened visibility of femininity more generally, and to its location within an “attraction to-” culture. The intersection of heterosexual women’s and gay men’s desire arguably heightens the emphasis on “softness” as a key component of post-feminist masculinity, where gay male culture holds up heterosexual male “softness” as part of its own visual landscape (Duane Duncan 2007, 2010). In a culture that only very recently was not visible at all, effeminacy is encouraged and what is celebrated are “visible public identities that [have] more in common with traditional images of femininity” (Duncan 2007, 334). In this way, representations of femininity on TubeCrush demonstrate the maintenance of hegemonic femininity, tied into notions of “softness” and “bosom” power.”

    • Dan Simpson says:

      The paper isn’t too caught up in broad sweeping facts. You through the baby out with the bathwater if you just dismiss it. But their ain’t any baby in the press release (if we’re going to stretch that metaphor)

      • Anonymous says:

        “The paper isn’t too caught up in broad sweeping facts”

        From the paper:

        “In so doing, we challenge current orthodoxy by arguing that TubeCrush represents a shoring up of white male privilege.”

        “We read TubeCrush as offering a phallic-oriented visual economy that turns desire toward
        hegemonic masculinity by combining a celebration of mostly white men who represent traditional masculine values (muscles and money).”

        “As such, TubeCrush is a reestablishment of traditional gender roles within the context of post-feminism: it gives the impression of liberally making visible gay men and straight women’s desires, but its visual economy celebrates hegemonic masculinity in the form of money and muscle, while offering a pacifier for the neoliberal worker in an urban, alienating bustle of the big city.”

        “TubeCrush, we argue, thus alerts us to the ways gender power is reasserted, and that, despite its perceived subversive potential, locates desire back into normalcy.”

        Science!

        • Adam says:

          This comment might read as sassy but I hope that doesn’t put Dan off from replying — all of these do seem like broad, sweeping facts that are important to the paper.

        • Anonymous says:

          And just to correct something: i quoted Dan who talked about “sweeping facts”, but my original post talked about “sweeping statements” (“If this makes any sense, i would reason that this paper *is* bad, as it makes sweeping statements from a very narrow focus/perspective/sample/whatever.)

        • Dan Simpson says:

          Bless, Adam. I have never in my life been put off by sassy anonymous interactions.

          My view is that I spent a lot of words making the point that the claims aren’t generalisable. But I’m mostly not concerned that things are not supported by data. Firstly, feminist media studies isn’t “science” it’s a deeply qualitative discipline. There is a lot of room in the tent for that.

          Why do I not just want to blow this out in a giant pile of scorn? Well basically because qualitative theory deeply informs both hypothesis building and model checking, so I think it’s hugely important. This study tries to sketch out some boundaries of the world that we can come in and clarify later. So the article is an object that has some use, so let’s not actually destroy it. As I said above, there’s a baby in the bath water.

          • Keith O'Rourke says:

            > qualitative theory deeply informs both hypothesis building and model checking, so I think it’s hugely important.
            Agree but also adding salience to what happened in say in an experiment and what should be made of it – getting a profitably credible story out of it.

            I gave a talk along these lines to faculty of the http://www.sgs.utoronto.ca/prospectivestudents/Pages/Programs/Public-Health-Sciences.aspx – as part of an academic exchange between quant and qual faculty that were being merged together (1996?) into one department. I likely did not make enough of model checking though.

            I was trying to give an opinion where qual work would fit in with and could enhance quant work. I was promptly rebuked with one qual faculty telling me I should enroll in their course because it would change my life. I was so stunned I did not know how to respond.

            Afterwards, when I tried to discuss any theory of qual methods with one of them, they would respond “I am not a theoretical qual person but an applied qual person. So can’t discuss the theory – I just know its right”.

            Maybe things have changed…

  8. Keith O'Rourke says:

    > Quantitative thinking around the data gathering, the design of the experiment, and the limitations of the measurements can inform the reliability of a qualitative study.

    When I was in Toronto, this was the group that was most open to those lines of inquiry – http://www.ryerson.ca/tsc/about.html

    Hey, winters has just barely started ;-)

  9. Brian says:

    I’m not sure this is really “quantitative thinking.” Although I guess suggesting that humanities scholars would benefit from a grasp of basic logic sounds less generous.

    • Dan Simpson says:

      The sort of thinking that you’d have to do if this was a quant rather than a qual study. For example, issues around experimental design, sampling, generalization, data quality etc.

      And let’s not all pile on humanities scholars. They do really good work. And this study was done in the UK where a big part of a funding call involves demonstrating “impact” or “pathways to impact”, where impact is taken to mean impact on society outside your field. That sort of system strongly incentivizes this sort of “story time” rubbish.

      • Keith O'Rourke says:

        Most of what I learned about experimental design, sampling, generalization was from qual scholars (before my first course in statistics) but it was subset of qual scholars. For instance, learning about anthropologists taking sexual advantage of their study subjects and animal communications researchers taking the same advantage of their chimpanzee subjects (disclosed in a public lecture). The latter was suspected of leading to an exaggeration of language abilities.

        I only learned about poor data quality when doing actual statistical work.

Leave a Reply