>The thing that stands out to me is that a group of people thought it'd be appropriate to classify all or most men and women based on a specific college and Amazon's Mechanical Turk.
Do you think this is the be all and end all of the topic? Some researchers had an idea to test and did a study and tried to get as random a sample of people as they could that might be representative of the population. Or even a sub group of people. They got some results and published. If the results are interesting enough, they will try and do a better study.
The fact that your experience is different doesn't refute the study either. It's statistical. They aren't saying the 100% of all friendships are like X. I dare say it's also obvious that a sociological study has limitations and pointing them out isn't interesting.
From an academic perspective, it's no problem as the primary audience are other researchers where the limitations are a given.
However, I posted the details from the study's design for the readers of Hacker News, who might assume that the study is generalizable to the broader U.S. population, especially since only the abstract is available for most users who see the link.
Do you think this is the be all and end all of the topic? Some researchers had an idea to test and did a study and tried to get as random a sample of people as they could that might be representative of the population. Or even a sub group of people. They got some results and published. If the results are interesting enough, they will try and do a better study.
The fact that your experience is different doesn't refute the study either. It's statistical. They aren't saying the 100% of all friendships are like X. I dare say it's also obvious that a sociological study has limitations and pointing them out isn't interesting.