Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
More women in a STEM field leads people to label it as a ‘soft science' (theconversation.com)
18 points by pseudolus on Jan 24, 2022 | hide | past | favorite | 10 comments


I’ll save everyone some time: the article doesn’t contain any stats, nor mentions of how the survey was done or which scientific disciplines were undervalued. It also doesn’t link to any peer-reviewed article where the author could have filled those gaps. It however links to articles from reputable scientific journals such as Business insiders or the New York Times and myths such as the "wage gap".

It’s ironic considering the article is whining about some scientific disciplines being considered less rigorous, yet makes no attempt to be rigorous itself.


It's a research brief, and it links to the DOI. You can find the actual paper: https://psyarxiv.com/xyast/download?format=pdf


The only linked paper is from Journal of Experimental Social Psychology, so basically soft science whining about hard science with "muh men bad" argument.

So typical divisive BS gender study. I'll continue to work with coworkers from both sexes without taking offense in a disparity that isn't even proved to have influence on papers' quality and science output.


The article doesn't discuss what the fields actually are that got labeled. In my opinion things such as sociology, psychology, economics, and political science are all "soft"

Simply because the complexity of the systems makes them insanely difficult to analyze and we don't have a firm grip on the mechanics, not because there are more women.

Correlation != causation

Edited to add: from my perspective, the label of something as a "hard" science comes from the solidity of analysis and models. How well and how often, how reproducible are the models and results when compared to the real system? The closer the model is to reality, the more solid or "hard" the science


It also doesn't reveal the size of the effect, or their sample size. It keeps its cards very close to the chest.


It’s part of the published article. Behind a paywall of course, but here is a preprint: https://psyarxiv.com/xyast/

A few parts to the study. One had 167 participants and had them label fields as hard or soft given a (sometimes incorrect) number of men vs. women.

Another had 220 and went the opposite way (given a description of a field as hard vs. soft, estimate men vs. women).

All participants recruited via mechanical turk.

Theres a lot more discussion and a few other subparts. Glancing at it it seems like a correlation, but not super strong.

Also, not my field at all. So reading it in detail is probably not helpful.


> All participants recruited via mechanical turk.

That skews the demographic so much the study is basically unusable.


While I kind of agree, I am also sympathetic to the fact that finding study participants that aren't heavily skewed is very difficult, and would probably be prohibitively expensive to do for every single study.


In this case it's better to survey a sample that is skewed but in a known way (you can then address these biases in the paper) as opposed to skewed in an unknown way via mechanical turk (how many are pretending to be part of the demographic you targeted just because western academic surveys pay more?).


Oh no. I would not use mturk for psych studies. I know its been on the rise and its not a good way to sample a population at all.




Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: