posted
I have the raw data to figure out who did best, but don't have the time. If you'd like to do it, let me know and I'll send you the spreadsheet.
Yes, as we figured it was a bit of a crapshoot. However, there were 2 female writers (#8 and #17) whose submittals were clearly considered to be male by a significant majority of the voters, and 1 female writer (#12)whose submittal was considered to be clearly female by a significant majority of the voters. None of the male writers were clearly considered to be either male or female by a signficantly large majority of the voters.
My submittal was #6 and it was considered to be written by a female by the majority of the voters, but not at what I would considfer to be a significant majority.
posted
Looking over the 1st 8: there was one in which I couldn't tell the gender of the POV character (1st-person). Of the remaining 7, there was 1 (#1) in which the POV character was male but most people (correctly) thought female. In that one, the issue was a woman being annoyed at not having a child named after her, which I'd call a woman's perspective.
Of the remaining, unless I made a mistake, all 6 had the votes match the gender of the POV character, not necessarily the writer. One was mine (#5): the POV character was a "she" and I'm a "he." I think this is natural. If we're immersed in a viewpoint, we naturally associate the gender of that viewpoint with the piece.
#8 was STRONGLY voted as male (incorrectly), and in it, we had an encounter between a male POV character and a woman who turned out to be a technological re-creation of a woman. Good job, #8!
I guess I should do some work today, so I won't analyze the other 8, but this is my conclusion so far.
posted
I was only 40% right on my guesses. It was really hard for me to tell which was which. I only felt really strongly on two of them. #4 I thought was female. Right! #16 I thought was female. Wrong.
My sample had close to a 50-50 split. Does that mean I write gender neutral? or Does it mean that I failed to write in a genuinely male POV?
Very interesting. Thanks for coordinating it mike.
[This message has been edited by NewsBys (edited March 08, 2005).]
posted
Mine (#9) was a relatively close split, which leads me to think a lot of folks were reverse-psychologizing. I mean, it's a first person female voice, and it's a love scene. But I guess the polygamy element might have made more people think "guy". I knew it was a pretty girly clip, but I had gotten it ready to email to someone else and though "heck, I'll just send this to mike."
Dude, I just realized that I failed to read/rate the last two clips.
I think this study shows that women are more interested in studying this sort of thing than guys. Except mike. Of the 15 I rated, I got 7 wrong, so I stink pretty much.
posted
Mine was the only DEAD EVEN split (#3). Hopefully that means my writing will appeal to both genders. Either that or it means it appeals to neither!
I was also the LONE and ONLY CORRECT vote for #8! Thank you. Thank you.
And I scored 61% correct guesses (because that's what they were for the most part.)
posted
Thank you, mike. This has really been enlightening for me and has helped me more than I expected.
As for judging the gender, I looked at style and subject matter. I also looked at how believable the POV seemed to be.
My guesses turned out to be so wrong I don't even want to mention them.
One of the pluses is that I realized I'm doing better with POV than I thought.
Oh, and I agree with Mary. There were some great snippets in there. The ones that particularly appealed to me at this moment in my life were #7, #9 (I'm in the mood for something romantic, I guess), #10, #15 and #16.
Mike, thanks for this. It has been fun to see the results. I only got about 40% correct. I suck at 'guessing'.
I also looked at subject matter and style. I think it might have been more difficult with just 13 lines. Kind of hard to get a feel without really seeing the flow of the story. I still enjoyed it thought.
posted
My sample was #7 and was male POV. 9/14 or Roughly 64% were correct. (10/15 or 67% if you count my vote, but I cheated)
I would like to hear more about how people chose their guesses. My guesses were all intuitive right-brain sort, without analysis at all, except on basis of emotional response to the writing. I got 4 correct out of the 17 I guessed, roughly 24%. I thought about doing it in a more analytic manner but decided against it as a real story reader reads by intuition and not by reason.
I do have a question though about the samples. How many people did what I did and opened a random file and took out a random sample? How many people chose to pick a sample which was either contra-gender or gender neutral? How many analysed the samples and what method was used? How many guessed on a fuzzier basis?
posted
I made a table to explore the gender of the POV in these pieces. #6 did not identify the POV character's gender.
GUESSED is about what gender got the most guesses. #3 had a tie.
Of the others, 12 of 16 had a match between guessed gender and POV gender.
POV GUESSED MATCH? 1 M F N 2 M M Y 3 M tie 4 F F Y 5 F F Y 6 ? F 7 M M Y 8 M M Y 9 F F Y 10 M F N 11 F F Y 12 F F Y 13 M M Y 14 F F Y 15 M M Y 16 M F N 17 M M Y 18 M F N
I didn't pick my writing sample at random. I specifically picked one with the opposite POV gender from me (which is hard, since I don't usually write from female POV).
[This message has been edited by wbriggs (edited March 08, 2005).]
posted
I came up with my guesses based on imagining the author reading the story out loud. I guessed correctly for all of the male authors. I haven't yet gone back and looked at the ones I got wrong.
I had two possibilities to choose from for my sample. The one I didn't choose was a (probably YA) fantasy, with a young girl as POV. I figured that would be too obviously female. So I went with the other which happened to have a male POV (it was #15). And slightly more people thought I was male than female (8 out of 15 votes).
Thanks Mike for running the experiment. It was fun
posted
I got only 6 correct. I have no idea what percentage that is. I'm not a math person.
I was #2 and I chose a selection that could stand alone and make sense. I avoided choosing dialogue because I felt that it would skew the vote toward whatever gender was most prominent in the conversation. For the same reason, I did not choose anything that was deep penetration into a character's thoughts. So that was about as "neutral" as I could find.
I actually put quite a bit of thought into each vote. Unfortunately, that did not seem to help me choose correctly. I based my decision on how authentically I thought the gender of the POV character was portrayed and then on the subject matter. Anything romantic, I attributed to a female writer, which probably was not fair. The two that were the most difficult for me were #3 and #7.
posted
I think the main thing that indicated "male" for me (incorrectly, much of the time) was if there was a lot of description of spatial relations.
If there was a lot of dialogue, to tended to guess female.
I made notes on the first 4. 1 I guessed female because "edon" is a weird name, and it didn't otherwise seem to be a fantasy story. 2 had a spatial description, though as I recall it also contained the word "copse" which should have been a tipoff. 3 had a lot of dialogue. 4 had "Oh..." and "darling" which I thought sounded girly. But it was not written by a girl.
posted
At mike's suggestion I posted the same fragment (#12) to gender genie. Just to see, like. Most voters thought it was written by a female. Last time I submitted something to gender genie it thought male and this time...
(Drumroll.......)
quote: Female Score: 121 Male Score: 181
The Gender Genie thinks the author of this passage is: male!
So I guess Hatrackers are better judges. (Quite right too!)
posted
Well, Gender Genie must be using a logical algorithm. So, our readings must be more intuitive than otherwise, and we inuited you better than others. Posts: 2710 | Registered: Jul 2004
|
posted
Two points to consider here: first, the samples are too short for good intuitive analysis. I almost always guess correctly the gender of the people here at Hatrack, even if their username is unclear -- but I generally have to read several posts before I pick up on it. I wish I'd had the time to participate in this; I would have liked to see how I did statistically when faced with such a brief selection (but after a quick glance at a few of those pieces, I probably wouldn't have done better than anybody else). And this brings up the second point -- you guys are good enough writers to carry a POV other than your own (at least for a few paragraphs). It's a lot easier to guess username/gender relationships, because most of what's written on these forums is done from our own POV, not deliberately from someone else's. I suspect that with a long enough sample of fiction, patterns would still develop revealing author tendencies (not just gender, but a little psychology and philosophy). Even if the author is deliberately trying to mask their identity, that masking should reveal a pattern. But I don't know of any way to test this without posting/analysing really long pieces -- too time consuming for anyone not working on a doctoral thesis.
Posts: 491 | Registered: Oct 2004
|
posted
Jeraliey, Good job fooling us! For me a couple things that made me think male. One was the POV's name (Vallacher) sounds like something a guy would make up. Also the POV being male. You also had sparks flying and a scene that was kind of industrial sounding.
It was a great example of how we really can't tell and writers do have the ability to 'be' anyone.
Printer-friendly view of this topic
