Hacker News Reader

Comment by utternerd | Hacker News Reader

I question the method of this "study". I don't think it's a leap to say 40 undergrads writing in their journals is anecdote.

View on HN

It's a very common methodology and it's hard to do sociological studies any other way really. They can't follow subjects around with cameras or something without creating a fake environment.

That doesn't seem like an argument that the technique is valid

I agree with you. The grandparent of this post is commuting an appeal to nature fallacy.

Just because sociologists do it, doesn't necessarily mean it's the right thing to do.

Not every study is a double-blind random sample with statistical modeling and error bars.

Which also means not every study should be used to project its results over an entire population.

You're maybe non familiar with qualitative research methods? A study like this can be used to discover the kinds of challenges that women face in engineering work environments. It would be a mistake to try to make quantitative claims about these results, but no one is doing that here.

I agree with the first part of your comment, but the authors are making specific suggestions about changes to college programs.

Yes, specific suggestion about qualitative changes to college curricula, based on qualitative results of a study. That seems reasonable to me. Once these changes are enacted, of course, they can be subject to quantitative study.

I'm familiar, I just don't find the study itself to be that compelling. We have two cherry-picked journal entries. The first one we have absolutely not context, and to be honest, sounds overly dramatic: 'the guys in their group came in and within minutes had sentenced them to doing menial tasks while the guys went and had all the fun in the machine shop'. In the second instance, we have someone complaining about being an intern and not getting interesting opportunities-- how original. To be fair, I'm sure her co-workers were probably creepy and that quite frankly is unacceptable.

I guess I just don't find any of the insights in this article particular compelling or new. We do have a gender problem, but I don't think bs studies like this lend any credibility to the issue.

Which population? It's at least four engineering schools and a defense contractor. It would be extremely odd if the problem only covered those specific "populations" and didn't extend any further.

Not an excuse a study being garbage. Many fields with studies based on people are flooded with absolute crap based on researching undergrads. This is no different.

If you can't spend the effort to get any kind of reasonable sample, your work doesn't carry any weight.

Reply on news.ycombinator.com