3 ms·

I only skimmed the article, but I didn’t see one measure that I was looking for: the performance of women who participated in the study but were unmodulated com

I only skimmed the article, but I didn’t see one measure that I was looking for: the performance of women who participated in the study but were unmodulated compared to the performance of women who were not participants of the study. My (completely instinctual, and explicitly not fact-based) theory is that perhaps women expect to do less well because they expect that women do less well. This should cause women who believed they were modulated (or that they could be) to be more likely to continue with the process because they believe that they are more likely to make it through than they otherwise would be.

antihero 4y ago
Confidence is a huge factor in so many things.
- Ancapistani 4y ago
  True. My thought behind this is that if women believe that being a woman negatively impacts their employability, and their behavior changes as a result of that belief in a way that negatively impacts their employability - it would be a self-fulfilling prophesy.
brigandish 4y ago
More data or access to the data would be nice but I believe this was addressed: > In the spirit of not giving away what we were doing and potentially compromising the experiment, we told both interviewees and interviewers that we were slowly rolling out our new voice masking feature and that they could opt in or out of helping us test it out. Most people opted in, and we informed interviewees that their voice might be masked during a given round and asked them to refrain from sharing their gender with their interviewers. Women couldn’t be sure they would sound like women and thus be judged as women.
hedora 4y ago
It is buried, but they say that women that were modulated to male did slightly worse than unmodulated women. Men modulated to female did slightly better than unmodulated men. Men did better than women on average, regardless of modulation. To me, this suggests the interviewers are slightly biased toward women, but the interview process is strongly biased toward men. In a throwaway comment, they say that ignoring people that dropped out after 1-2 failed interviews eliminated the gender gap. To me, that suggests there are confounding factors at play.

antihero 4y ago

Confidence is a huge factor in so many things.

Ancapistani 4y ago
True. My thought behind this is that if women believe that being a woman negatively impacts their employability, and their behavior changes as a result of that belief in a way that negatively impacts their employability - it would be a self-fulfilling prophesy.

brigandish 4y ago

More data or access to the data would be nice but I believe this was addressed: > In the spirit of not giving away what we were doing and potentially compromising the experiment, we told both interviewees and interviewers that we were slowly rolling out our new voice masking feature and that they could opt in or out of helping us test it out. Most people opted in, and we informed interviewees that their voice might be masked during a given round and asked them to refrain from sharing their gender with their interviewers. Women couldn’t be sure they would sound like women and thus be judged as women.

hedora 4y ago

It is buried, but they say that women that were modulated to male did slightly worse than unmodulated women. Men modulated to female did slightly better than unmodulated men. Men did better than women on average, regardless of modulation. To me, this suggests the interviewers are slightly biased toward women, but the interview process is strongly biased toward men. In a throwaway comment, they say that ignoring people that dropped out after 1-2 failed interviews eliminated the gender gap. To me, that suggests there are confounding factors at play.