When Teamwork Doesn't Work for Women | Hacker News Reader

When Teamwork Doesn't Work for Women

nytimes.com

67 pointsdsr1221 comments

From last year's epic "emotional labour" thread on metafilter [1], `barchan` made this comment that seemed to resonate with a few people:

  > Another terrible thing about emotional labor at work is that
  > women "stepping up" to get things done is not just expected and
  > unappreciated, it's called "teamwork"; but when guys do it, it's
  > called "leadership".

[1] -- http://www.metafilter.com/151267/Wheres-My-Cut-On-Unpaid-Emo...

This article is excellent, both in the way it's written and in the underlying genius of using tenure as an indicator for credit for each paper.

One interesting point is that a women-only or woman-solo paper gets more credit (9%) than a mixed-gender paper (8%).

I'm wondering why they rounded single-digit number, it's 15% imprecision (8.4% and 8.5% aren't significantly different, whereas 7.5% and 9.4% would be a great gap).

The paper is excellent and genius, the article is pure NYT political fluff.

It's too bad indeed that the labels are rounded to one digit, but at least the plot bars are from unrounded data (as the second figure with 6%, 7% and 8% bars makes it clear), so I think readers still likely get the right impression of the effect size.

"The bias that Ms. Sarsons documents is so large that it may account on its own for another statistic: Female economists are twice as likely to be denied tenure as their male colleagues."

Aren't they using earned tenure as the metric to determine whether someone did or did not get credit? Seems like that's really just one point then rather than two. Unless I'm missing something.

I'm curious about the data for women who published papers with other women. That will tell us whether the issue is that men tend to get most of the credit for collaborative work done with women, or something else.

The article seems to address that in the second chart. When a woman publishes with only women they get the 9% bump they're supposed to. When they publish with a mixed-gender group they get only 4%, and when they publish with only men they get almost no benefit. Is this what you were asking?

I wonder if they controlled for age in this study. Women in the sciences are younger than men (on average), so a dual-female paper is more likely to be from two contemporaries, while a dual-male paper could more easily be from an old prof and a young one. From what I've seen, the more established author tends to get more of the credit/prestige, even though the less-established one tends to do more of the work.

Age doesn't appear to be in the pdf. A surprising omission.

http://scholar.harvard.edu/files/sarsons/files/gender_groupw...

That does seem very flawed. If there's a strong correlation between gender and age in the field then why would that be not accounted for? I can definitely see people assuming the older person was the brains/leader of a project.

> while a dual-male paper could more easily be from an old prof and a young one.

So the best case to check if this is true is the reverse, an older female prof with a younger male.

The story seems to be that when Janet writes with George, her colleagues infer that George deserves the credit. That might be a reasonable inference if women were more likely to join research collaborations as the junior partner, but in fact Ms. Sarsons finds that they are less likely to do this.

> This leaves a sample of 552 economists ... between 1975 and 2014

Soon there will be studies about a few dozens people and latter about individuals. I just wish other issues would receive similar attention and dedication.

Lies, damn lies and statistics ... I have yet to find a person that does not do its best to find a data presentation that suits the agenda.

It's horse shit on inspection. Of course, so are virtually all economics papers, so let them eat each other alive for all I care.

Please document your judgement.

I'm usually one of the first ones to detect horse shit in women-victimizing papers, usually by matching statistics with reality. For example, given that I've seen many (if not almost all) women in my programming environment being promoted to "managers" earlier than men, statistics about women being paid less in IT "for the same experience" get discredited by the fact that a "manager with 1 year of experience" is not the same when he's had 7 years as a developer under his belt or when she's 24 years old. Studies rarely specify "with the same experience in the current and previous job". Besides the fact that studies are inherently false if they don't explain why women get promoted a lot a lot a lot in the reality I'm witnessing.

So you see, I'm open to be on your side. Just in this case, I didn't detect how this study could be biased. Please document, thanks.

The premise that coauthors on papers make equal contributions is in itself grossly biased. They don't. I'm coauthor on papers I've never even read. If you were looking at the class of Scott Locklin on papers I'm named on, assuming everyone is supposed to get equal credit, using the methods they did, I'd be considered to be unfairly not getting enough credit, because there are less of me than everyone else and most of the papers I'm in, I do not deserve equal credit. The proper comparison class is, do principal authors get equal credit for their contributions, normalized to their proportion in the population.

What about when someone does statistics to make a decision for themselves which they do not reveal?

Then pretty much by definition we wouldn't hear about it, would we?

People are also good as self deluding themselves with statistics.

We detached this subthread from https://news.ycombinator.com/item?id=10874697 and marked it off-topic.

Reply on news.ycombinator.com