At 11:19 AM 3/12/2013, Mary Barnes wrote: >On Tue, Mar 12, 2013 at 10:00 AM, Michael StJohns <mstjohns@xxxxxxxxxxx> wrote: >> At 07:56 AM 3/12/2013, Dan Harkins wrote: >>>While these studies are interesting and thought provoking, I think it is >>>wrong, and very dangerous, to use these studies to justify blanket >>>statements about intelligence, group or otherwise. >> >> I'm laughing a bit about this thread. For example, there's also "substantial evidence" that young women and young men do better in gender segregated schools because the women's IQs plunge due to primping and displaying and men's IQs plunge due to testosterone if they're kept together. Unfortunately, there's also "substantial evidence" that doing things this way can lead to some socialization issues (where both groups tend to have warped views of the members of the other groups). (I myself doubt both versions of the "substantial evidence") >[MB] I too find your response and some others laughable. I would >agree that some of your statements with regards to all boys and girls >schools are wrong, however, there is indeed research highlighting that >girls do better in all girls schools due the fact that they are given >the attention that often goes to the boys in math and science classes. I somewhat agree with this, but other studies have indicated that this has quote a lot to do with the specific teacher and general disciplinary environment of the school than being an absolute characteristic of gender divided studies. >The rest I is untrue based on my experiences, but I would certainly >welcome someone pointing out research supporting your statements. >Both my sons have gone to or go to all boys schools (since first >grade). Their girlfriends go to all girls schools. There is no >socialization issues in general. That all said, my sons' school has >some of the geekiest kids in the DFW area, Sorry - geekiness *is* a socialization issue. I say that as one of the more geeky people at my school. With respect to supporting research - didn't you note the quotation marks? And the statement "I myself doubt both versions of substantial evidence" I've read a few main stream press articles on the gender divided schooling issue and "socialization" was mentioned pretty much in the same breath (paragraph?) as the "girls do better in all girl schools". I considered both of the statements somewhat suspect for the same reasons. But I mentioned them - with the quotation marks - to indicate that cherry picking data to support a conclusion is generally problematic. Here's where I'm at: The school studies were done with random (e.g. public school) and non-self-selected (e.g. parent selected) groups as the subjects. AFAIK the IETF is pretty much a completely self-selected group of people and most especially the women are self-selected - and I wouldn't consider that the school study applies much given those difference. I mentioned it because its conclusion - that women and men are "smarter" when separate (yup - paraphrasing) seems to be at odds with the other mentioned conclusion that "groups are smarter the more women in them". For the other study mentioned by Margaret (Wolley et al) it actually said this: >Finally,c was positively and significantly correlated with the proportion of females in the group (r=0.23,P=0.007).However, this result >appears to be largely mediated by social sensitivity (Sobel z=1.93,P =0.03), Which actually says that "the more sensitive people in the group, the better the result, and by the way women tend to be more sensitive". But the problem set for the study bears not a lot of resemblance to the problem set for the IETF. So again, I'd claim it's mostly inapplicable, hence cherry picking. > so it is likely that there >may be *slightly* more issues with socialization than the average >public school. [/MB] >> >> Seriously - diversity is generally good. I think we all get that. Going off and trying to support that general statement with (Dan's words, but I think I agree) "cherry picked" data isn't going to advance that cause much. >[MB] The data isn't cherry picked - there has been *lots* of research >on this topic over the past decade (and even those previous). Such >studies are doubted because I am sure they are not of any interest to >the folks that suggest they don't exist. So, these wouldn't have been >on your radar. Not surprising, those that are doubting that IETF has >any issue with diversity are folks that aren't in the minority- it's >really hard to understand an issue if you haven't dealt with it >yourself. There's lots of research showing lots of bias in our >society - the fact that many have never chosen to read any of it does >not mean it doesn't exist. [/MB] I've read it (them) or things like it. What I've gleaned from each and every study is that their conclusions are suspect when you try to generalize them to groups not constituted like the study group or to problem sets not tested by the study. If you want to convince me via objective standards that an IETF WG with more women in it does a better job than a WG with few women (or even the reverse of that where an IETF WG with more women does a worse job than a WG with few women), you'll need to do the study on the IETF and publish the criteria for "better" before you start. Or find a group that has been studied that has a population and work mode much more similar to ours than the studies that have been pointed to. I don't disagree with desire for more diversity, I just don't think the case needs to be made by citing studies that don't necessarily have anything useful to say about the IETF. Mike >> >> Mike >> >>