So a rigorous statistical analysis of the Gold Medal Winners' stamp images reveals...

[joanofarchitrave]

3/12 stamps involving female athletes show images of them in sporting action.

13/20 stamps involving male athletes show images of them in sporting action.

I know they had to pick the images fast and under pressure to produce them in time. I think, personally, it was predictable that in those conditions, the female images shown were more likely to be smiling non-threatening images, rather than photos of women in athletic extremis.

Or is this quite positive, showing women in positions of triumph rather than still in competition?

It's not a huge issue, acelt, obviously. And you are absolutely right that there was absolutely no conscious sexism and that they will have selected the images for their "iconic value" - I doubt that you'd find anyone much to disagree with you. It's just interesting that in images of women "iconic" tends to involve "face visible and preferably smiling" more than in images of men, while in images of men "iconic" tends to involve "doing something active, even if it obscures the face" more than in images of women. It's a reflection of the fact that, even in a celebration of what women have achieved , what they look like is more important than for an equivalent group of men - unconsciously, if we're looking through a bunch of shots of a woman, we're likely to pick one out thinking "oh yes, she looks really good here", more so than for a man.

The images of Laura Trott with her medal and of Etienne Stott and Tim Baillie in their canoe are both fantastic photographs, both wonderfully iconic images. But they are iconic in very different ways. And when one type of image is overwhelmingly chosen to represent women and the other to represent men, it's a reflection of how we think about them.

It's not a "look at this outrageous sexism" issue. It's not a "complain to the Royal Mail" issue. It's just a quick snapshot of how women and men are still thought of in subtly different ways.

"Most people (in my opinion) are overthinking this to death, if you have to work this hard to find sexism then it probably isn't there. There are enough real instances of sexism in our country/world that people do not need to imagine more where there isn't any."

I don't reckon we're overthinking it acelt. Obviously it's nowhere near the top of the list of things to worry about, and may not even be a "bad" thing as such, but I find it fascinating to look at. Sexism, and other -isms, don't just pop up through isolated incidents by nasty people - views about gender are ingrained in the subconcious way we see the world. Men are often portrayed as active and women as passive in society - it's interesting to see if these ideas show up in the stamp choices.

"For the record, I believe the choice of image of each athlete was selected for their iconic value, whether the athlete in question was mid-event or not, whether the athlete was male or female, a fact which I have no problem with."

Out of interest, why do you think that selecting for iconic value led to more images of men in sporting action and women not in sporting action? Or do you think it was a coincidence?

x-post with Casey - you put it way better than I could :)

i don't know how bad this is... i understand the arguments about showing female sport as less 'action-oriented' but on the other hand i think it's good to show the faces of the female athletes to help young girls to relate to them, i think this is more necessary for female athletes and young girls than it is for male athletes and young boys.

imo there's a negative in relation to how men feel about women's sport, but there's possibly a positive in how girls feel about women's sport and female sport champions...

there's no doubt that the contorsions of facial expression during sport can look a bit daft when captured in still photography (missing the beauty of the motion) - i would hate for that to put teenage girls off in any way.

"Out of interest, why do you think that selecting for iconic value led to more images of men in sporting action and women not in sporting action? Or do you think it was a coincidence?"

In statistical samples of 10 and 17, yes I believe it is more likely to be coincidence than a sexist choice (even if it is an unconscious one).

Oh yes, it's a symptom rather than a cause. Or mostly so, at least. But why is it "more necessary for female athletes and young girls than it is for male athletes and young boys"? Why are teenage girls more likely than teenage boys to be put off by facial expressions? Because society has relatively fixed ideas of "femininity" and "masculinity".

If all other imagery of men and women was noted for its even-handedness then, yes, it would be more likely to be coincidence. But almost any other set of imagery you care to name shows exactly the same pattern, to a greater or lesser extent.

Obviously it could be that the Royal Mail image selection person/team is both one of very few people in Western society to be unaffected by unconscious cultural bias and at the same time has coincidentally arrived at exactly the distribution of images that previous studies of that bias would have predicted, yes.

What I am mightily het (sic?) up about is being insulted by you (the menz nonsense). I do not like people that have to resort to insults to deflect away from the fact that they are losing an argument, it shows such ignorance, unintelligence and disrespect.

As you now seemingly agree with all my points, I am not sure I have lost any argument.

As for your strawman arguments

The only straw men were the ones you put up, where you inferred I'd said things I hadn't.

and for the record I can guarantee you I watch a lot more female sports than you do

Guarantee, eh? Is that the 1 year, 3 year or the full lifetime?

i think that sometimes the choice of shot has depended on the quality of the picture and other factors. For example, I read an article on how the Post Office chose which pic and the speed with which the whole thing had to be done.

The Alaistair Brownlee one is an action shot. But it was not their original choice. They wanted the one of him with the union flag round his shoulders smiling as he walked over the finish line. But when they 'blew it up" they saw he had loads of spittle round his mouth and, in their words, looked as if he had just been let out of an asylum! So they went for the action shot instead....

CaseyShraeger - I wasn't actually saying that teenage girls are more likely than teenage boys to be put off by facial expressions. I was saying that I am more worried about teenage girls being put off because less of them do sport in the first place and they are suffering for it in terms of health, strength and self-confidence.

I am sure teenage boys don't want to be photographed with a massive facial contortion either , but due to social conditioning, more boys know the positives of sport and so are less easily put off.

I totally agree that society has relatively fixed ideas of "femininity" and "masculinity" as you say, but I work with 10-14yr old girls and I personally would go for gently easing girls into sport gently by assuring them they can do sport and look 'good'.

I personally would love to say 'fuck you' to the patriarchy and show women at their most raw and agressive (wouldnt' that be an amazing photography exhibition?) but i don't think that at this point in society that will help teenage girls operating within our societal constraints to do more sport.

Yes, that's why I said it was a symptom rather than a cause.

These 'results' are statistical junk as there is no randomized control group for them to be compared against.

That aside, please continue grinding your axes.

You did get that "rigorous statistical analysis" was a joke, didn't you?

What kind of randomized control group would you have liked to have seen, by the way? If you're comparing surgical with drug treatment options then you can have a control group who receive no treatment, but if you are comparing "men" and "women" then finding a control group who are neither men nor women is probably challenging -- so I'm sure that's not what you're suggesting.

I assume this week's release of official statistics showing that girls were more likely than boys to be named Amelia is also statistical junk because it didn't include a randomised control group?

Both, like the census, are whole-population studies - in one case looking at all the names given to all the babies born in England and Wales in 2011 and in the other looking at all the stamps featuring British gold medal winners issued in the UK in 2012. Now obviously one is much (much, much, much, much, much) more reliable statistically as the basis fir drawing wider conclusions from than the other, but that's to do with the size of the population studied and the objectivity of the results (a baby is either named Amelia or it's not, no "well, if you look at it from this angle it might be named Amelia, from a certain point of view"). Randomized control groups have nothing to do with it.

Ah, the 'it's-only-a-joke' defense. When a serious question is wrapped in a 'joke' (which, as is the joker's intention, is still taken seriously by others: "This is precisely the kind of small thing which exemplifies how insidious sexism is." - SGM ) it's often because the question, which despite 'it's only a joke' protests, is genuine, isn't worth considering based on the evidence.

A randomized sample would be drawn, randomly, from an equal set of male and female photo's each containing an equal number of 'action' shots.

The subject being explored would be gender-based selection bias in humans, making your first counter-example irrelevant. Drug-substrate interactions and surgical procedures are not assumed to be gender biase-susceptible decisions in terms of efficacy, it is a simple 'does A or B cause C more effectively' question, hence the use of negative controls. As far as is practical those control groups should still be randomized in order to minimze the effect of selection, conformation, etc... biases. As for your other response...

"I assume this week's release of official statistics showing that girls were more likely than boys to be named Amelia is also statistical junk because it didn't include a randomised control group?"

Data sets are not statistical tests.

No, fair enough. It was an observation of a sort which might in another life field lead to a hypothesis, not in any way a statistically measurable set, so I shouldn't have presented it, even jokingly, as a piece of evidence of anything.

But my axe is in fantastic shape.

"But my axe is in fantastic shape."