[nck4222]

I believe this is it. From the source paper:

"We excluded respondents who had a non-US IP address, reported having a same-sex marriage, reported an age at marriage of less than 13 years old, or were above age 60."

The paper used mTurk to get ~3000 responses to their survey. So it's basically saying "the men who answered these questions ended up having this score, the women who answered these questions ended up having a slightly different score"

EDIT: Link to paper, http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2501480

[andyjohnson0]

Apart from the small sample size, if they collected data by mechnical turk then I wonder what sort of biases that introduces? In the paper you linked, Frances and Mialon state:

"Samples of mTurk workers have been found to be more representative of the US population than in-person convenience samples, standard internet samples, and typical college samples"

I am somewhat sceptical of this, and there seems to be some evidence to back-up my scepticism [1][2][3]. In particular, [2] states:

"In sum, the MTurk sample is younger, more male, poorer, and more highly educated than Americans generally. This matches the image of who you might think would be online doing computer tasks for a small amount of money."

Which are some of the factors that the marriage study itself seeks to examine. This looks like lazy data collection methodology to me.

[1] http://www.culturalcognition.net/blog/2013/7/10/fooled-twice...

[2] http://themonkeycage.org/2012/12/19/how-representative-are-a...

[3] http://www.behind-the-enemy-lines.com/2010/03/new-demographi...

[Tyrannosaurs]

It merely says that it's more representative than the other groups, not that it's particularly representative. For instance college samples are massively unrepresentative (younger, smarter, probably wealthier than average).

Essentially it's just a more representative sample than these other horribly unrepresentative samples he's listing.