Gender of the author: guess with the genie

[Razor]

Evidently I write like a girl. Most of what I put in was female, and what wasn't was really, really close.

 

So... should I like... make my copy of Word do some bicep curls and chinups so it's more manly? O.o

[Demetz]

[Liberty Island Prologue] Female: 1311 Male: 788

[Liberty Island Chapter 1] Female: 9817 Male: 9950

[Nathan and David 1] Female: 3625 Male: 5831

[Nathan and David 2] Female: 3928 Male: 5185

[Nathan and David 3] Female: 4649 Male: 6156

[Questions Confusion and Confessions] Female: 5366 Male: 4182

[Together] Female: 2248 Male: 2186

[Webs Chapter 1] Female: 6734 Male: 5153

[Webs Chapter 2] Female: 8262* Male: 4891

[Webs Chapter 3] Female: 6085 Male: 6425

Total rated Female: 5

Total rated Male: 5

 

Total Female points: 52025

Total Male points: 50747

 

Percentage Female: 50.62

Percentage Male: 49.38

 

Notes: If you look at webs chapter 2, you'll see a ridiculously high ratio of female points to male points. This is because the words with and not are severely overvalued in the female category. Those two words alone were responsible for 5622 out of the 8262 female points. If you discount this outlier, the overall points shift back toward male, but it is still very close.

 

Taking ten samples of my writing I got five rated male and five rated female. The point margin is ridiculously close even discounting that outlier. This leads me to the conclusion that this gender genie's lamp was purchased in a knock off shop.

 

Additional Note: It amused me greatly to see Nathan and David consistently rated as written by a male author despite the fact that some would consider it to be gay erotica written for women.

[Bondwriter]

Ok, I did it with Quite a Boarding School, which is on our site and not on eFiction.

Words: 26143

(NOTE: The genie works best on texts of more than 500 words.)

 

Female Score: 26041

Male Score: 31698

 

The Gender Genie thinks the author of this passage is: male!

 

And the Genie is quite right. I still haven't gotten around to read the long PDF explaining the gender issue in writing.

 

And my word processor is not on steroids.

And I didn't rub the genie for hours so that he'd spit out this result...

[jfalkon]

I went throught 20 files from my computer. (Most of them are stories on this site.) All were fiction. 90% of the stories were identified as male. So I guess I'm a guy most of the time. Some of the scores were close for example, 2354 female and 2453 male. That was only a 4% difference. Another close one was 1742 Female and 1419 Male. (19%) Most, however, were quite significant ie a difference of at least 50%.

 

The two stories that came back as female both centered around female characters. Considering some of the feminine words are her, she, and hers that makes some sense. They only count 33 gender biased words, after all. I think the test may tell us more about the characters in the story than the authors.

 

The other thing I noticed is that "said" is a male word. Does that mean that men write more dialogue?

[AFriendlyFace]

The other thing I noticed is that "said" is a male word. Does that mean that men write more dialogue?

My take was actually that men write more purely functional, basic dialogue, whereas women might get more creative.

 

For example rather than 'said' an author could use the words 'growled, laughed, giggled, snorted, asked, inquired, responded, answered, enthused, gasped, etc."

[Bondwriter]

The two stories that came back as female both centered around female characters. Considering some of the feminine words are her, she, and hers that makes some sense. They only count 33 gender biased words, after all. I think the test may tell us more about the characters in the story than the authors.

I tried it with yet another story, Imprisoned in Silk (dealing with forced feminization, yes, this is the kind of stuff I write sometimes...)

 

Words: 20939

(NOTE: The genie works best on texts of more than 500 words.)

 

Female Score: 26920

Male Score: 23868

 

The Gender Genie thinks the author of this passage is: female!

So yes, JFalkon, you must be right. This story has mainly female characters. And a happy ending, but this genie doesn't yet compute this type of content. The next step would be to have a program rate the cliffhangers in a story. Maybe CJ would eventually have to get out of his denial.

[Phantom]

This is a fun toy, but it insults me!

 

According to it my writing style is female by a wide margin!

 

Just goes to show you can't always trust a computer

[Bondwriter]

According to it my writing style is female by a wide margin!

 

Just goes to show you can't always trust a computer

I must agree with JFalkon on this one, a story involving male characters is written by a male, one about female/not sure characters turns out to be female, so, no big worry. Unless your story was about Yukon lumberjacks trying to build a dam and make themselves a hut for the winter...

[TalonRider]

Just goes to show you can't always trust a computer

Actually, you can trust the computer, it's the programmer you can't trust.

[Tiger]

Actually, you can trust the computer, it's the programmer you can't trust.

Exactly Jan! The software is only as good as the programmer. The first rule is WYSIWYG (What you see is what you get). Any mistake in the code will result in the accuracy of the application being hindered.

[Duncan Ryder]

I tried a few chapters of EW and a few of Light. EW came out more female than male, though chapter by chapter the margins were always really close. Light came out male, the newest chapter hugely male, like by a 2 to 1 margin. No idea what that means... I did wonder if a story being in first person would make a difference, since EW is in first person....

[Cynical Romantic]

This is just a silly toy, relying on stereotypes. Virtually everything I've ever written that I plug in comes up "male", though I'm obviously a woman writing it. Ironically, one of the few pieces of writing I tested in it that came up "female" was one of my few attempts at writing a story from a male point of view. Of course, it was a first-person point of view, so I have to wonder if Duncan is maybe onto something.

 

Anyway, I think this site is more for entertainment than about anything serious.

[AFriendlyFace]

Hmmm

 

I'm going to make a comment/observation, and I do hope it doesn't rain on anyone's parade or offend anyone, but I will say that I'm a tad disheartened by how many male authors have expressed disappointment that their writing is 'female' and by how many female authors have expressed disappointment that their writing is 'male'.

 

...I'm assuming you all are mostly just joking around and don't really care, but speaking as someone who loves being male and being around males, I have to say that I love defying gender stereotypes even more. Does any of this sort of thing really matter? I mean it's interesting, but it's interesting to me because it (supposedly) highlights one of my 'female' areas. I'm fine with that, happy about it even. I would never want to walk too far on either side of the gender lines. I'm a guy, and I like doing 'guy stuff', but that doesn't mean that - except for a few physically impossible things - I can't and won't do anything and everything that any 'typical' female might do and enjoy it and be proud of it. Just as while I'm 'gay' it doesn't mean that I can't and won't do anything and everything any 'typical' male might do and enjoy it and be proud of it.

 

For me personally being gay is about ignoring gender roles and expectations and doing what I want when I want it without shame, guilt, or hesitancy.

 

*shrug*

 

That's just me though

-Kevin (who's perfectly delighted to 'write like a girl')

Edited August 22, 2008 by AFriendlyFace

[Graeme]

I'm not disappointed. I just found it amusing

 

And seriously, given that it seems to be getting the gender wrong more than 50% of the time, maybe they should just invert the gender selection in the program

 

You can't blame a "gay writing like a female" bias, either, because I noticed that it was also wrong most of the time for the straight writers.

 

In short -- it was a bit of fun, but not to be taken seriously.

[jfalkon]

Hmmm

 

I'm going to make a comment/observation, and I do hope it doesn't rain on anyone's parade or offend anyone, but I will say that I'm a tad disheartened by how many male authors have expressed disappointment that their writing is 'female' and by how many female authors have expressed disappointment that their writing is 'male'.

 

...I'm assuming you all are mostly just joking around and don't really care, but speaking as someone who loves being male and being around males, I have to say that I love defying gender stereotypes even more. Does any of this sort of thing really matter? I mean it's interesting, but it's interesting to me because it (supposedly) highlights one of my 'female' areas. I'm fine with that, happy about it even. I would never want to walk too far on either side of the gender lines. I'm a guy, and I like doing 'guy stuff', but that doesn't mean that - except for a few physically impossible things - I can't and won't do anything and everything that any 'typical' female might do and enjoy it and be proud of it. Just as while I'm 'gay' it doesn't mean that I can't and won't do anything and everything any 'typical' male might do and enjoy it and be proud of it.

 

For me personally being gay is about ignoring gender roles and expectations and doing what I want when I want it without shame, guilt, or hesitancy.

 

*shrug*

 

That's just me though

-Kevin (who's perfectly delighted to 'write like a girl')

 

I think no one takes it to seriously here but you make a good point. There could potentialy be a sort of feed back loop if people try to be either "male" or "female" by some random deffenition. People want to be more like their gender so they act out the steriotype. As a result the steriotype becomes more extreme. Then people try to live up to the new standard.

 

I agrea with the rest of you that it is just a toy. What I find interesting is the numbers that the creators of the algorythm claim. It is suposed to be 80% acurate. I'm not sure if this refers to the original algorythm or the simplified version that we have been looking at. There is also an issue of what is an acceptible margin of error. By that I mean if the female score is 1207 and the male score is 1210 should that really count as male and if not where should the cut off be? Also why does it work better on works of a particular length? Why should that matter? Did the original 600 writing samples have anything in common? How were they selected? There are plenty of unanswered statistical question here and that is the part that is interesting.

[AFriendlyFace]

I agrea with the rest of you that it is just a toy. What I find interesting is the numbers that the creators of the algorythm claim. It is suposed to be 80% acurate. I'm not sure if this refers to the original algorythm or the simplified version that we have been looking at. There is also an issue of what is an acceptible margin of error. By that I mean if the female score is 1207 and the male score is 1210 should that really count as male and if not where should the cut off be? Also why does it work better on works of a particular length? Why should that matter? Did the original 600 writing samples have anything in common? How were they selected? There are plenty of unanswered statistical question here and that is the part that is interesting.

Good point about about whether extremely similar scores really 'count' anyway.

 

I really do think we're going to have a bias here in our community though. The program probably wasn't created for gay literature, gay writers, or gay friendly writers. I'm sure that does make a difference.

 

I think the length thing is a rather valid point. The more the program has to test and analyze the less likely that it'll be a random, freak occurrence.

 

For example, let's say I don't know what the probability of flipping a coin and getting 'heads' is. If I flip it twice, three times, or even four times it's really not that unlikely that I'll get something like 75%-100% probability of heads (or tails) when obviously we all know that the real probability is 50%.

 

If I flip my coin 1,000 times then I'm probably going to have something really close to 50% (maybe 48 - 52%) if I do it 10,000 times I would expect it to be something like 49.XXX% - 50.XXX% and so on. Small 'anomalies' work themselves out in the long run over time.

 

So a female author could easily submit two or three lines of text that used the words 'above' 'said' etc. and come up with a male score, because most authors are likely to use those words at least occasionally, but I'm sure the theory goes that if she submits longer work such random chance will be reduced or eliminated.

[Dion]

I never took it as anything other than a fun way to pass the time. It's not going to be very accurate, but it is entertaining.

 

Interestingly enough - I tried breaking down the parts I put into the Genie by how I felt when I wrote them. I picked one story only (I don't separate them into chapters until the story is complete) and divided it, as best as I could remember, into sections according to when I wrote them.

 

The more emotional sections rated 'female' while the rest rated 'male'. The overall result was 'female'. In a way, this is pleasing to me. I tend to lean more toward evoking emotion when I write as opposed to provoking thought. I may have come up with a winner! LOL!

[Conner]

Evidently I write like a girl. Most of what I put in was female, and what wasn't was really, really close.

 

So... should I like... make my copy of Word do some bicep curls and chinups so it's more manly? O.o

 

[b]Say it ain't so, Jamie![/b]

 

 

How about you do the bicep curls and chinups .....and post pics?

 

Conner