As Amichai-Hamburger expected, the Wikipedians were more comfortable online. "They feel the internet is a more meaningful place to them," he says. But to his surprise, although Wikipedia is founded on the notion of openly sharing and collecting knowledge as a community, they scored low on agreeableness and openness.

http://www.newscientist.com/article/dn1634...ml?DCMP=OTC-rss

http://www.liebertonline.com/doi/abs/10.1089/cpb.2007.0225

I'd heard of this study earlier, and it seems to be one of those studies that's interesting in theory, but probably useless in practice. Seeing that people outside the Wikipedia community are interested in it, I decided to take a look.

The first thing that jumps out at me is that the report does not describe how the participants were found—the closest thing to a hint is that "[t]he data were collected via online questionnaires". This does not inspire confidence. From the news I've heard, it's my understanding that it collected (mostly?) Israelis; this calls into question whether Israelis, Israeli Wikipedians, or Israeli non-Wikipedians are representative of the wider community (and non-community). I suspect that it is not representative, not to mention the small sample size.

The second thing that jumps out at me is that if you look at the data table provided in the report, Wikipedians scored *higher* on the openness trait. The table gives a mean score of 3.75 for male Wikipedians (SD 0.63) compared with a mean score of 3.55 for male non-Wikipedians (SD 0.51), and a mean score of 3.92 for female Wikipedians (SD 0.38) compared with 3.64 (SD 0.59) for female non-Wikipedians. There's clearly a major error here, given that the paper is claiming, in its conclusion, that Wikipedians have less openness.

This may have been caused by what appears to be a copy-and-paste error evident in the text. Below the text reading "the average of the agreeableness trait among the Wikipedia members is significantly lower as compared with that of participants who are not Wikipedia members" the text mentions "In addition, a significant difference was found in the openness trait […]" followed by a verbatim copy of the previous text regarding agreeableness—an incongruous phrasing, to say the least!

This paper looks to be a trainwreck: when the conclusion is not supported by the data published in the paper, with such an elementary mistake in the text, the paper fails to be convincing. The fact that the sample is probably not representative of the population to be studied is also troubling.

I'd find it interesting if Wikipedians did overall score much differently—and I'd find it further interesting if Wikipedians were markedly different from the denizens of other Internet fora—but this study simply doesn't inspire confidence at a basic level, let alone one worthy of rigorous peer review.

(Disclosure: I am a volunteer administrator on Wikipedia.)

P.S. As far as I can tell, no one on the Wikipedia Review has noticed this basic error, despite a number of snarky anti-Wikipedia comments. Surely that says something about the quality of the discussion there, thekohser, that I need not put into words.

I'd heard of this study earlier, and it seems to be one of those studies that's interesting in theory, but probably useless in practice. Seeing that people outside the Wikipedia community are interested in it, I decided to take a look.

The first thing that jumps out at me is that the report does not describe how the participants were found—the closest thing to a hint is that "[t]he data were collected via online questionnaires". This does not inspire confidence. From the news I've heard, it's my understanding that it collected (mostly?) Israelis; this calls into question whether Israelis, Israeli Wikipedians, or Israeli non-Wikipedians are representative of the wider community (and non-community). I suspect that it is not representative, not to mention the small sample size.

The second thing that jumps out at me is that if you look at the data table provided in the report, Wikipedians scored *higher* on the openness trait. The table gives a mean score of 3.75 for male Wikipedians (SD 0.63) compared with a mean score of 3.55 for male non-Wikipedians (SD 0.51), and a mean score of 3.92 for female Wikipedians (SD 0.38) compared with 3.64 (SD 0.59) for female non-Wikipedians. There's clearly a major error here, given that the paper is claiming, in its conclusion, that Wikipedians have less openness.

This may have been caused by what appears to be a copy-and-paste error evident in the text. Below the text reading "the average of the agreeableness trait among the Wikipedia members is significantly lower as compared with that of participants who are not Wikipedia members" the text mentions "In addition, a significant difference was found in the openness trait […]" followed by a verbatim copy of the previous text regarding agreeableness—an incongruous phrasing, to say the least!

This paper looks to be a trainwreck: when the conclusion is not supported by the data published in the paper, with such an elementary mistake in the text, the paper fails to be convincing. The fact that the sample is probably not representative of the population to be studied is also troubling.

I'd find it interesting if Wikipedians did overall score much differently—and I'd find it further interesting if Wikipedians were markedly different from the denizens of other Internet fora—but this study simply doesn't inspire confidence at a basic level, let alone one worthy of rigorous peer review.

(Disclosure: I am a volunteer administrator on Wikipedia.)

well, what do our anecdotal experiences tell us about the reality and truth of the situation?

Still, it's nice that someone is trying, I suppose... I guess primarily-anecdotal studies aren't the sort of thing that get you an NSF grant, unfortunately.

It's always nice when scientific studies verify and validate common-sense conclusions that are obvious to anyone with direct experience of the situation.