0
What is the purpose of producing flawed research like this?

You do understand that the premise of this is completely false? That an evaluation of how many pull requests are accepted/rejected is of no value at all of it doesn't consider why the request was accepted/rejected. ie was it rejected because of bad coding, or because it didn't fit with the project aims and/ethos. For example anyone of those rejected pulls (from either gender (identifiable or not) might have been because the particular code modifictaion was a duplicate of something someone else had already done, or because the project owner doesn't want his/her code to contain that capability. etc.

This looks like another complete misuse of data to represent something which the data doesn't prove in anyway, shape or form. Surely if you wanted this study to be of any value you would have gone back to the project owners to understand why rejects were rejected. Not assumed it was always because of gender bias and/or bad coding.

waiting for moderation
1 Answer
0
Accepted answer

In the paper, we present several alternative theories that might explain the data that we gathered. If you'd like to give a theory of your own that explain the data, we're listening.

waiting for moderation
0

I'm afraid this looks like so many other studies which reach similar conclusions. Performed with an already decided agenda. Hence the starting assumption that rejects/accepts would be because of gender bias, which led to the incorrect conclusion that this somehow gave some indication of the coding abilities of the identifiable genders, and bias against identified genders. All of which was reached without ever examining the code itself, or the actual reason for the reject/accept as given by the project owner. It's like you don't understand that correlation doesn't equal causation, and went onto produce a report to prove it.

-
waiting for moderation
0

What is the purpose of producing flawed research like this?

Let's see, this paper has now been mentioned by multiple newspapers and magazines with headlines like:

"GitHub Users Find Women's Code Better Than Men's — Until They Know Who Wrote It"

"Women considered better coders – but only if they hide their gender"

"Women coders do better than men in gender-blind study" etc. etc.

While Murphy-Hill's "In the paper, we present several alternative theories that might explain the data that we gathered" sounds very academic and reasonable, however the reaction of the media certainly certainly supports Dowell's "Performed with an already decided agenda". Also the authors received a grant (money) from the NSF.

The authors definitely have received attention, and if they claim that attention is not what they were seeking, they need to prove otherwise.

Honestly, the very first thing that should jump out to a researcher from the data is that there were 150,000 men and 8,000 women (as reported by Villemaire). The stark difference here is in the participation rates of men and women, not some difference between 74% and 78%.

I mean, how much intelligence does it take to understand that a self-selected sample of 8,000 vs. a self-selected sample of 150,000 from populations of approximately equal size cannot be used to draw inferences about the general populations. The media headlines about this article such as "Women coders do better than men in gender-blind study" certainly show no such discernment.

- edited
waiting for moderation
0

I have finally worked out what it is that bothers me so much about this paper, and what led me almost immediately to see it as a piece of research which was conducted with a preset agenda.

The one field of human endevour which stands the best chance of not being burderned with human prejudices is computing. Primarily because the systems don't care about race, gender, sexual orientation, belief systems or cultures. They only care about numbers.

What you've done here is attempt to impose your belief in all humans practicing bias and therefore discrimination onto that. You've picked the worst possible community to attempt to do that to though, because again the very concept of 'open source' is one which by it's very nature negates bias and discrimination. If a project lead/owner is being an arsehole, the open source community is the one community where he/she is going to be called out as being and arsehole. If he/she is a real arsehole anyone can fork the project and create a non-arsehole burdened version.

So I come back to my original question. Why create a paper which was only every about creating a case to support your own beliefs (that bias existed), when it's clear such bias doesn't and can't survive in the field or community you are talking about, and then present it as having some real meaning? Couldn't you find some kind of real field of research which was about discovering something real? Instead of producing something which was only ever about reinforcing your own preset ideas about how there must be data to support something which doesn't make any sense at all.

-
waiting for moderation
0

Dave, I'd suggest you read up the research of bias, especially implicit bias. You claim that "What [I]'ve done here is attempt to impose your belief in all humans practicing bias..." That's not a belief, that's just prior work. For example, "Implicit biases are pervasive. Everyone possesses them, even people with avowed commitments to impartiality such as judges." (http://kirwaninstitute.osu.edu/research/understanding-implicit-bias/). See also: "Pervasiveness and correlates of implicit attitudes and stereotypes".

-
waiting for moderation
0

I had a feeling that would be your response. You're more than entitled to your views, your ideals, and your biases, which is what I believe your paper demonstrates. I'm not the only one who seesn what you've produced in this particular report as being on dodgy ground, and for a number of different reasons. You're not the first to produce such questionable material, and I doubt you'll be the last. I just worry about the damage such material does to the whole equality debate, and how it affectes human interactions. It does have an effect, that much is undeniable, and I suspect when it is slanted quite as badly as I feel this is, I think that effect is negative. How many git project owners are now thinking "you know, it'd be easier to just reject every pull request from any woman, just because it means I won't have to live with researchers calling me sexist when I've never cared about the gender of submitters"? And I can assure you that exactly the effect you've had upon me. I don't use github, but I can't see why my reaction to the constant stream of anti-male bias which is constantly being poured forth shouldn't affect everywhere on the net. I might as well be hung for a sheep as a lamb, if you're going to go around accusing me of practicing gender bias, irrelevant as to whether it bears any resemblence to reality, why wouldn't I do so? I can't understand how any of you can believe that you and your output exists in a vacuum and that you don't contribute to the positions.

-
waiting for moderation