Hacker News new | ask | show | jobs
by noobermin 1372 days ago
Does "gender gap" here matter more than what larger fractions show? For example, they say "women desire more Affirmation" but 49% have it as a top two while 56% have "time" in the top two. Men also have 54% for time in the top two. I feel like that's more interesting than looking at percentage differences (which tbh don't really seem to matter that much for this statistic).

EDIT: a striking example of this, they say "women give more gifts" in terms of LL they express, as a finding. But only 32% have this as a top two expressed LL, so a good majority of women don't even have gifts as a top two expressed LL. So, who cares about the percent difference 8%, does it matter? What does it even mean? What if 1% of men did X vs 9% of women? Why would the interesting takeaway be "wow, women are more likely than men to do X!!" vs. "gee, a sheer minority of people do X at all."

Saying the words "women are more likely than men to do X" leaves the reader unfortunately with the impression "many women do X," whereas that might not be true.

2 comments

>Saying the words "women are more likely than men to do X" leaves the reader unfortunately with the impression "many women do X," whereas that might not be true.

I wonder if this is related to "X is better than Y." leading to "How dare you say X isn't bad!" I didn't, X can be extremely horrible, it can be the second most horrible thing to exist. The only qualifier I gave was that Y was at least one rank more horrible than X. If instead I had used "Y is worse than X" this likely wouldn't have happened.

I'm wondering if many people, when reading a relative comparison, imagine it in some absolute setting and then assume that was the intended information to convey. If your case they imagine a case where more women than men do X, meaning they imagine a whole lot of women doing X and very few men doing X, even though it could actually have been 2 of every million women vs 1 of every million men.

>>>Saying the words "women are more likely than men to do X" leaves the reader unfortunately with the impression "many women do X," whereas that might not be true.

Is there an accepted term for the extra caveat-ing and qualifying that people do with language these days to avoid the non-logical misinterpretations of factual statements by stupid people?