It’d be neat to run the SDMB post history through a gender guesser algorithm and see how our posters write.
Thanks for creating the poll. It seems like my guess was pretty close as after 185 votes slightly over 70% of those who took the poll said they were male.
On the other hand, there’s another poll active right now that has it at 57% male, 30% female. At least one of those polls is clearly unreliable, and there’s little enough difference in methodology that I think we can say that neither is reliable.
As another option, many years ago, back in the days of IRC chat, I wrote a script that could determine gender from username with about 95% accuracy. If I could dig up that old thing, and find some way to get it to run, and look through an appropriate sampling of the SDMB, I could get a number from that.
We’d also need to decide what the appropriate sampling is. You could go through a fair sampling of the usernames in the member database, or of all of the posts in currently-active threads, or of all of the usernames in currently-active threads, and you could get different numbers for all three of those.
You’d also want to account for spammers in some way, who often have completely random usernames and may not have any gender (and certainly not any relevant gender). The simplest pass would be to exclude all banned users from the count, but then you’re also excluding folks who were banned for reasons other than spam, and I wouldn’t be surprised if there’s a gender bias on who ends up getting banned.
As an aside, I just ran that previous post through the guesser that Reply linked to. It’s a little shorter than the page wants (another complicating factor, since many posts will be even shorter than that), but it thinks that if that’s informal writing, it’s highly confident (correctly) that I’m male, but if it’s formal, then it’s a toss-up, leaning ever-so-slightly towards female, and possibly (incorrectly) European.
I’m going to go dig up one of my longer posts, and see what it says about that.
It’s likely that on a board which is overwhelmingly male, females would find it more relevant to announce their gender, in order to contradict the presumption that they were male.
OK, two conclusions: First, remarkably few of my posts are over 300 words. Second, for those that are, that analyzer shows a fairly broad range of results, from strong male to weak female. It may be that the number of words needed for good results is actually more than 300.
That other active poll that Chronos mentions also has just over 10% identifying as pizza.
But won’t a poll just find the ratio of male/female among members who take part in polls? That won’t necessarily correlate with the ratio of male/female Dopers. Only admin can really answer this one.
And how would an admin answer it? I think there’s a spot for gender in the profile, but it’s optional, and a lot of folks leave it blank. And even at that, anyone can read profiles, not just admins.
EDIT: Correction, there is not a spot for gender in the profile, unless you shoehorn it into “biography” or something. So admins wouldn’t even know that.
But how many as sausage pizza?
Cut, or uncut?
I am not now, nor have I ever been, a pizza.
This business about females announcing their gender just because they are in the minority is a bad idea-it’s like putting a target on them when it comes to misogynists.
No idea what you’re talking about.
What I meant was that sometimes it’s relevant to the conversation whether the speaker is male or female, e.g. when they want to share some incident or particular experience-based perspective. In such cases, when the group is mostly male, a male is more likely to just assume that people will assume he’s male and thus not bother to mention that fact, while a female looking to make a comparable point would be less likely to make a similar assumption and more likely to announce her gender.
The reverse would be true of a female dominated group. For example, I would speculate that on a nursing message board, you might find more examples of people saying “I’m a man” or “I’m male” than the reverse, and for similar reasons.
Whatever the distribution:
“Your odds are good, but the goods are odd”
*Stolen from any local engineering school
The global average which is roundabout 50/50. Localized polls could move it as far as 20/80 but stat of large numbers will bring it close to even. I thought there were enough dopers.
I take your point. The question then is one of those imponderables like what song the Sirens sang or did Lady Macbeth have children.
But, you are what you eat, right?
And there are also some cases of women in a male-dominated group deliberately remaining ambiguous, or possibly even outright lying, so as to avoid sexism. IIRC, that’s the reason for Una Persson’s original username of Anthracite, before she determined that this board was sufficiently friendly to women, and also why Joanne Rowling used her initials instead of her first name on her books. Of note, this could also apply to participation in polls.
I write like a chick?:eek:
Not surprising. Pretty much all the people who taught me to write and affected my development were women.:D;)
Wow. I ran a 2,000 word speech through it and it not only guessed “male” for informal writing and “weak male” for formal, but it also got this part:
I am European, or at least I was. But what does “weak emphasis” mean in the context of written text with no formatting?
ETA: Review of a prior thread indicates the program thinks about half of the content submitted is from Europeans.