?

Log in

No account? Create an account

Mah voice, let me show you it. - Tactical Ninja

Nov. 26th, 2010

10:40 am - Mah voice, let me show you it.

Previous Entry Share Next Entry

This is fairly old, but interesting reading - Why James Chartrand Wears Women's Underpants. It's about the difference in opportunities on the faceless internet for a writer, before and after she started using a male pseudonym.

And it's been a while since we had one of these - The Gender Genie, which claims to be able to tell you whether a piece was written by a man or a woman based on the words used. I fail to see how 'said' is a masculine word and 'with' is feminine, but hey, I tried it. I used my five latest blog posts.


Or at least, sometimes I'm female and sometimes male. Two of my posts were written by a woman according to the Genie, and three by a man. I know five isn't a representative sample, but I've been doing this journal thing for seven years on a pretty much daily basis and I think I've probably developed a 'voice' in that time, you know? I also realise that there's been a lot of talk of sheep shearing in the last week, but it doesn't seem that shearing sheep is among the things this algorithm analyses.

Never mind that as I said to someone the other day, shearing sheep is not a manly thing, it's just a thing.

So anyway, after that I found all the academic essays I have on my computer and put them through. Apparently when I'm writing nonfiction, I'm a man. This might be why I spend so much time scoffing nuts and scratching myself while writing essays. It might also be why I get As. *cough*

Actually I think I get As because I'm damn good at writing essays, but it's interesting to see that these A essays are considered almost overwhelmingly masculine by the algorithm above.

Breakdown by topic, in chronological order, with the Genie's findings alongside my mark:

2008
“Won’t anybody think of the children?” - Neoliberalism in education policy in New Zealand. - M A+
“Please remove your invisible hand from my backside!” Women, work and welfare in the last 35 years: Feminist perspectives. - M A+
“You Have Made Our Lands Important”* – New Zealand’s development as a settler society and beyond. - F A-
Class only affects other people: Stratification in the ‘classless’ New Zealand society. M A-
Measuring crime – are police statistics reliable? M A

2009
Which came first, the chicken or the egg? Exploring the link between drug use and crime. M A
Useful Tool or Dangerous Problem: The social construction of knowledge about LSD 1954-1971. M A+

2010
The aim of critical analysis: an evaluation with examples. M A
Internet use and study style: How easily distracted are SACS301 students? (a data analysis). M A
Analysis of qualitative vs quantitative approaches to research: two examples. M A
Scarlet Women in the Golden Era – State intervention in the private lives of women in New Zealand. M A+
Getting an Early Start: An investigation into a New Zealand home visitation programme for child welfare. FA+

You'll note that both of the essays written from a feminist perspective are written by Tats The Man - in fact, only two of the essays were written by Tats The Woman. This is obviously because anything of quality written as nonfiction must be overwhelmingly written by a man, right? *cough* Wrong.

Actually I don't think that's quite the case, but I do think it's likely that the 'rules' for academic writing were originally made by men and that the 'rules' that were used to develop the nonfiction algorithm for the Genie were probably based on those rules. And I know that I have learned those rules really well, so that when I'm writing facelessly and academically, to all intents and purposes based on this idea, I'm a man. But of course this requires everyone to believe there's a difference and that this difference is measureable. If enough people believe and apply these 'rules' I guess it's possible for there to be a difference between the way women and men write.

But what I actually think is that this idea of there being a difference between men and women's writing is utter bollocks and that if the reader doesn't know your gender before reading, they probably won't know it after. They might think they know, but they actually won't. I base this on the fact that a sample of my blogging comes out about 50/50 female to male on the 'objective' measurement tool, and a 100% sample of my academic writing only produces the correct gender 16% of the time.

It just so happens that I'm doing an experiment with non-gendered identity at at the moment - I'll let you know how that pans out.

In other news, I had a haircut!



Cass is awesome at doing the kind of cut that doesn't look like you had a haircut but it just looks better. And the last one she gave me lasted 6 months.


Today is Fidels and somehow I got roped into watering plants on some roof over the weekend, and I'm betting Michelle has her Polwarths in so there'll be shearing on Saturday. On Sunday I'm doing nothing, damnit!

(well, maybe some groaning. and spinning. and cleaning. and the groceries)

*sigh*

Comments:

[User Picture]
From:pythia
Date:November 25th, 2010 09:50 pm (UTC)
(Link)
Apprently my journal entries are slightly more female than male, but any academic writing is WAY more masculine. Makes me think that they've put 'emotive' words as feminine, and 'technical' writing/terms as masculine.
(Reply) (Thread)
[User Picture]
From:tatjna
Date:November 25th, 2010 09:53 pm (UTC)
(Link)
Down the bottom of the analysis page there's a chart showing the words they are using to analyse it, and they seem to be mostly conjunctions and *argh fails at grammar* words that aren't nouns or adjectives.

And there's some weird weighting too. I tested it out with a couple of sentences in which I used more 'male' words than 'female' (picked randomly), but it still weighted the female ones more heavily and decided I was a woman that time.

(Reply) (Parent) (Thread)
[User Picture]
From:pythia
Date:November 25th, 2010 10:01 pm (UTC)
(Link)
Oh, two of the four essays I tried were about feminist issues and gender as well, and it decided I was ManMel for those, too. Hah.

Also, it's interesting that they seem to note present tense terms (is, are) as male, but past tense (was) as female.

i think the gender genie is full of shite.

Also, loving the bollocks tag.
(Reply) (Parent) (Thread)
(Deleted comment)
[User Picture]
From:tatjna
Date:November 25th, 2010 10:53 pm (UTC)
(Link)
I figured it was probably something like that. Although chances are all those works are playing by similar rules as well - which is kind of what I was getting at.
(Reply) (Parent) (Thread)
[User Picture]
From:richdrich
Date:November 25th, 2010 10:16 pm (UTC)
(Link)
Has anyone fed it the work of Jan Morris or George Eliot?
(Reply) (Thread)
[User Picture]
From:tatjna
Date:November 25th, 2010 10:25 pm (UTC)
(Link)
Or Vanessa Royall, or Leigh Greenwood.
(Reply) (Parent) (Thread)
[User Picture]
From:ferrouswheel
Date:November 26th, 2010 03:02 am (UTC)
(Link)
Y'know, the same statistical theory as the sentiment analysis in NetEmpathy could probably be applied to gender prediction of tweets. hmmm.
(Reply) (Thread)
[User Picture]
From:tatjna
Date:November 26th, 2010 07:04 am (UTC)
(Link)
If I say 'with' a lot in my tweets, will NetEmpathy think I'm a girl?
(Reply) (Parent) (Thread)