The GENDER GENIE! Analyzes Your Writing

[Brownian Motion]

It's probably not infallible, since it guessed Ann Coulter was a guy.

http://www.bookblog.net/gender/genie.html

[Jazzmoose]

Hmmm..put in one of my posts from the political forum and got this:

Words: 189

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

Female Score: 242

Male Score: 492

Okay, not bad, but it didn't realize I was a moose!!!

[Jazzmoose]

It's probably not infallible, since it guessed Ann Coulter was a guy.

Since this isn't the political forum, I'll show a little restraint...

[gdogus]

My results:

Words: 768

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

Female Score: 802

Male Score: 1563

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

It seems to have been correct with us guys (so far), but it's apparently not all that amazing:

The Gender Genie knows all.

According to Koppel and Argamon, the algorithm should predict the gender of the author approximately 80% of the time.

Accuracy Results

Am I right?

yes  159762 (59.38%) 

no  109309 (40.62%)

269071 total responses since September 13, 2003

Now, people could be reporting that the genie has guessed incorrectly just out of spite, but according to these numbers, the algorithm sucks.

Edited May 8, 2005 by gdogus

[Spontooneous]

In honor of the impending Bloomsday, I entered a piece of Molly Bloom's soliloquy...

It guessed male.

[gdogus]

In honor of the impending Bloomsday, I entered a piece of Molly Bloom's soliloquy...

It guessed male.

Beautiful.

[Alexander]

I submitted a piece of short fiction I wrote a few months ago. It guessed male.

Female Score: 3692

Male Score: 4335

[Kalo]

Ann Coulter's NOT a guy?

[Spontooneous]

I was having trouble deciding my gender. Turns out I'm male.

[Dr. Rat]

Words: 514

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

Female Score: 688

Male Score: 971

The words it chooses as "male" ae strange. For instance, my big masculine scorer was the word "some."

--eric

[maren]

Things I wrote in the Politics Forum here were classified as "male" --

But an abstract for a scientific paper I ghost-wrote turned out "female" -- of course, the topic was "estrogen replacement therapy"

[Brownian Motion]

But an abstract for a scientific paper I ghost-wrote turned out "female" -- of course, the topic was "estrogen replacement therapy"     

Is it common among medical researchers to farm out their writing tasks? My wife, who is an historian, has never heard of this being the case in her discipline.

[Dr. Rat]

But an abstract for a scientific paper I ghost-wrote turned out "female" -- of course, the topic was "estrogen replacement therapy"     

Is it common among medical researchers to farm out their writing tasks? My wife, who is an historian, has never heard of this being the case in her discipline.

I have a friend who makes a pretty good living doing this sort of work for doctors.

I suppose either historians are expected to be able to write, or there is less excess cash hanging around to pay ghostwriters in the history field than in the health field.

--eric

[maren]

But an abstract for a scientific paper I ghost-wrote turned out "female" -- of course, the topic was "estrogen replacement therapy"     

Is it common among medical researchers to farm out their writing tasks? My wife, who is an historian, has never heard of this being the case in her discipline.

I have a friend who makes a pretty good living doing this sort of work for doctors.

I suppose either historians are expected to be able to write, or there is less excess cash hanging around to pay ghostwriters in the history field than in the health field.

--eric

There's not a lot of excess cash in academic medical research (pharmaceutical/device industry, maybe... but even there the significant $$$ would go to the ad copy writers, not the scientific writers!).

It's more that academic history is so closely intertwined with writing -- whereas lab bench, operating room and bedside skills are quite a bit further afield -- and also, my "ghost-writing" has always been for people that I work with daily: answering phones, pipetting samples, doing literature searches (in the library in the old days, mostly online now) and writing grant proposals, reports, patient-consent-forms and a million other things... and a lot of the "writing" is re-writing, copy-editing. Or in the case of an abstract, distilling the gist of a 10 page paper into 500 words, 250 words.