The Data:
(Edit: corrected counts for game weighting filter)
550 women ... 23,600 women's ratings
7,300 men ... 318,000 men's ratings
Ratings only counted/used for games where at least five people have given a game weight.

The Notes:
When viewing the "percentage of ratings" charts below and noticing any differences, please bear in mind that in each chart the area below both lines down to the x axis is the area in common between the two.


Other than that... a few pictures says almost everything I might have wanted to say.

The Charts (ie Results):

Weight (1 to 5)


Duration (Minutes)
NB: Anything over 480 minutes garners ratings from approximately 0% of M/F alike


Weight (1 to 5)
NB: There are so few ratings at weight 5 (see 1st chart) as to make the above 5 result volatile.


Duration (Minutes)



The Conclusions:
So. If there's any gender bias for long/heavy/short/light games, the bias is so small as to make no sense as a basis for making assumptions about what someone likes. Specifically...

a) "What games might person X like?" => does not materially change based on X's gender.

b) "Do women like game G?" => is not materially different from men, basing on G's weight and duration

c) "What kinds of games is person Y likely to rate?" => does not materially change based on Y's gender.

d) "What percentage of gamers who rate game Z are women?" => has a trend depending on weight/duration of Z. This isn't the same question as (b) or (c) above. About 4% of ratings in my sample at the heavy/long end of gaming are from women, 8% of ratings at the light/short end of gaming are from women. Per long/heavy gamer guy there's fewer women taking/making the time to play/rate long heavy games. (But only by a factor of two vs short/light games.) But for men and women both it's the minority end of ratings... which is why it comes out to be not a material difference between the "likely" behaviour of a woman gamer vs the "likely" behaviour of a man gamer.


The Footnote:
I based rater gender on first name. For example, half the raters I've tagged as women have one of the following 40 first names:
Anna, Anne, Barbara, Beth, Catherine, Christina, Christine, Danielle, Debra, Elizabeth, Ellen, Emily, Emma, Heather, Jen, Jennifer, Jenny, Jessica, Julia, Kathleen, Kristen, Laura, Lauren, Laurie, Lisa, Liz, Margaret, Maria, Melissa, Michelle, Natalie, Nicole, Samantha, Sara, Sarah, Sharon, Stephanie, Sue, Susan, Suzanne

Interesting statistics!
After seeing the curves are more or less the same in all cases I do think that the first lines are the most interesting ones.


That's more than 13 times more men than women being active on BGG (which I assume is your data source).
If one also assumes that this represents the general world-wide men/women-ratio then what I would like to see is statistics comparing BGG:ers to non-BGG:ers (or at least boardgamers compared to non-boardgamers) All that data might be difficult to gather though...

I see the expression "my wife likes this game" or "my girlfriend likes this game", and I also saw a thread where it seemed to me like someone expressed dislikes towards such expressions (I could have misunderstood it). Well. Here are the facts:

There are more than 13 times as many men as women playing.
So statistically, most of the time, the woman is a non-gamer, which usually is a problem for the gamer (usually a man) since he/she wants to play more. Therefore it is important to have games that the non-gamers (most often "the wife") will enjoy.

Just my spontaneous reaction about this Not trying to attack anyone, just asking people to see the bigger picture. Carry on!

While it's clearly the case that there are vastly more men than women on BGG, it's quite likely the ratio can't be accurately drawn from my data above. Ballpark round numbers guestimate, sure. But not accurate.

eg I've actually tagged about 6,500 women and about 31,000 men. But most don't have ratings for this exercise. (ie the few that had any ratings at all haven't rated any games that have enough weight raters.)

Nice job! definately an interesting finding for me as if I were asked to guess, would have said there will be a gender difference in ratings.

Nice to extrapolate the 13 to 1 ratio of male to female gamers too. Sounds right..

Woo hoo! I like me them odds.

The bigger picture? Well. Here are the facts:

There are (far) more than 1,000 times as many nongamers as gamers. So statistically, most of the time, a man is a nongamer. Yet (many) folks assume that the default position is man=gamer, woman=nongamer (and therefore take the attitude that women on this site are abberations but men are the norm - we're all abberations! )

(I imagine that) None of the women on this site object to men discussing "my wife/girlfriend likes". The comments you're referring to arise when a couple of anecdotes are extrapolated to "all women like" or "women only like". This is particularly unrealistic when you consider that there is a short list of games that gets continually recommended for "the wives and girlfriends". So, these games get recommended as female-friendly, they get played with wives and then re-recommended. Which is fine, but people need to recognize that it's a (very) incomplete list and that women's taste in games varies as much as men's do. I play regularly with two women who, without my influence, would very much be nongamers - and our list of games we play bears little ressemblance to the "wives like" list because, instead of going for the old standby games, I pick games specifically tailored to their interests and mine (which is what I'd recommend husbands do too).

So, thanks to Joe for assembling these graphs that support what I've been saying all along - women's taste in games is as diverse as men's. (And it's not as if we're jumping into the "what should I play with my wife?" threads to complain, so I'm not really seeing what the problem is anyway.)

I wonder if there also is a tendency for bias in these ratings toward the inclusion of English-language gamers of both sexes. (I mean over and above the bias already introduced by this being an English-language site.) By that I mean people who have first names that are more recognizable to a native English speaker as being male or female. Of course I don't know much about Joe's background, and perhaps he is far more knowledgeable than I am. But in the course of my work I encounter a lot of people of non-Western descent, and I often have the experience of not being able to guess someone's gender from their name, or of finding that my instinctive assumption is wrong. I know Linda and Susan will be women, and Paul and David will be men (though I can't be sure about Pat and Chris, or, these days, Taylor and Kendall), but I wouldn't be sure how to sort Yue and Miraz.

Joe, what do you think? In going through the names, did you find a significant number you had to put aside into the "can't tell" category?

And all of the statistics don't count that I'm the one that uses this site. My wife plays games, but she isn't the user of the site. She isn't the one rating the games.

It isn't that she can't register, she doesn't want to. She would rather play a game than talk about them online. Is she a gamer? Sort of. As much as me? No.

So for all of us that talk about our significant other (whether it be wife or husband, boyfriend or girlfriend), you can bet that there's one (at least) other gamer.

My daughter plays Carc, TTR, Settlers, and Puerto Rico. She isn't registered either, but she's a gamer, and more so each day.

Mike

Oh my... I was hesitating to post that previous message
Just a reply to everyone.

First, when I'm talking about "gamer" I mean the person who likes to play a game, possibly goes and buys it and probably suggests that it's played. The person doesn't have to be talked into it.

I know you can't say "The stats here say 13:1 so that's what the world is like". I'm saying it's probably a reliable indication as to what the "global ratio" is.
Even if the ratio is not correct, even if it is 10:1, 10:5 or (unlikely) even 2:1, there are more male gamers than female. Just as for some reason there are more male computer gamers than female ones. I don't know why it's like this and I can't even prove it, but I would bet my own mother on it.

I can come up with a counterexample to gvchief's counterexample, but I won't, because individual counter-examples don't matter.
If you have a bowl of 1 million black tokens and 100000 white tokens, and you would pick out 10000 of these, you would probably get around 10 times as many black ones than white ones.

You can always come up with a counter-example to any statistical claim. Still, more men than women are gamer-types. And I never in my post claimed there are differences in tastes, or anything different between male and female tastes in games.

What I'm saying is, there are more male than females, and to quote myself:


I personally believe that the expression "games that women like" is a generalization based on the most likely situation, which is that the man is a gamer and the wife isn't. Meaning, when you say "games that women like" (at least when I say it, there are exceptions), you mean games that would be liked by the majority of women, that are the non-gamers, but also male non-gamers of course. However, for the non-singles (me included), the livingpartner of most men (the majority of gamers) is a woman so that's the most natural way to express it. Is it 100% correct? Probably not, but there are often generalizations when you talk about events or statistics and what effects they bring, and most of the time you understand what is being said anyway.

I didn't want to make such a big thing out of all this, but I can see I've stepped on some sore toes.

Lori that's an excellent question.

Yes the approach is biased toward using the ratings of people of European descent (including English descent).

I started with a large sample of names irrespective of gender or country, and used frequency counts to bias my searches, so it's not purely a matter of what names I can think of to begin with. There's quite a few names that are inherently ambiguous. I did make some use of various websites of names origins and usage for some foreign names or indeed any name where I was remotely uncertain. Some languages, there's generally no resolution of gender from name anyway. Some names in English do not indicate gender either. I believe some European languages are pretty rigorous about gender representation in names.

I tagged about a third of the users who give any ratings at all. Of those, I've explicitly tagged about 10% as ambiguous. There's a diminishing return per distinct first name as the names get rarer, and not remotely everyone even enters their actual name in the first place.

btw if you check people who have declared their country of residence, you get this many pages of users (sample):
United States 2636
United Kingdom 295
Canada 277
Australia 106
New Zealand 27
Germany 95
France 78
Italy 74
India 25
Singapore 19
Malaysia 10
China 7

Peter thank you for a polite response.

The point then is we can easily be more factually correct with the phrases we use. Phrases like "games non-gamers might like" or "gateway games" or "games to play with your non-gaming partner" are not more difficult to say and are more accurate than "games to play with your wife" or "games women might like".

Since men gamers are also the vast minority of men, it's not correct to make gamer-assumptions about men and non-gamer assumptions about women.

Also the point is to recognise that when someone is asking for recommendations, gender plays no part in the likely outcomes. If the focus person for recommendations, self/partner/friend/co-worker, is male or female isn't important. Find other attributes to narrow down the recommendations.


My wife has moved from "non gaming partner" to "gaming partner" over the past couple of years. All the "games to play with your wife" lists are useless to us, or anyone else (male or female) whose partner plays cheerfully. "Games to play with your wife" lists are likely to be useful to a woman with a non-gaming husband. "Wife" in the title of such a list is factually misdirecting if the list is supposed to be about "non gaming partners".


btw this entire article is in response to being asked to produce it by someone else.

(edits because I kept mangling my intent with poor phrasing even worse than usual.)