Monday, February 19, 2007

carnal sociology coda

"I did some googlestalking of Erich Goode after you wrote that post about him."
"Do tell."
"His father was a sociologist. Si Goode."
"Former American Sociological Association President. There's a rumor that if you are a legacy in sociology you get a special EZPass through human subjects committees."
"His stepmother was also a sociologist. Lenore Weitzman."
"Really? She was the source of arguably the most spectacular error in the history of quantitative sociology." [full ASR debate here]
"About the consequences of divorce for the standard of living of men and women?"
"Yes. If her study had also looked at the financial consequences of out-of-wedlock births, Erich and 'Peggy' probably could have been a great case study."
"He also wrote this critical review of Carolyn Ellis's book on autoethnography."
"Autoethnography is where people do participant observation research that primarily focuses on themselves."
"You don't need to tell me that. You may need to clarify it for people who read your blog, which is why you've inserted this despite it not really being part of our conversation."
"In autoethnography, what is important is not literal truth but narrative truth."
"Anyway, Goode doesn't like how Ellis calls the people in her studies 'participants.' He prefers 'subjects.'"
"I can see where that would help for maintaining objectivity and distance, as well as keeping roles straight so that one doesn't cross any lines."
"He also writes: 'Like Leo Buscaglia, the guy from the seventies who always wanted to hug everybody to make them feel better, Carolyn Ellis hugs her students a lot.'"
"I read part of a Leo Buscaglia book when I was in high school. It made me want to die."
"Goode ends his essay by saying: 'Me, I'm not a hugger.'"
"It's true that in that in his essay, he doesn't anywhere talk about hugging any of the women from NAAFA. He wouldn't even offer to refill that one subject's glass of wine when he refilled his. I guarantee you Leo Buscaglia would've asked if she wanted more wine."


Gabriel said...

JFW has got to be one of the only blogs to regularly use Socratic dialogue.

Corey said...

I don't have a comment on Goode right now (though the piece you link seems to be a reasonable criticism of AutoEthnography).

I found the Peterson - Weitzman exchange in ASR more interesting than the Goode stuff.

Weitzman's ASR rejoinder to Peterson should be considered exhibit #1 for the importance of sound archival practice. She writes: "The raw data file that is stored at the Murray Center is the original 'dirty data' file and does not include these cleaning changes." (pg 537).

and then...

"There were so many discrepancies between the questionnaires and the 'dirty data' raw data file, and between the questionnaires and the mismatched SPSS system file, that we finally abandoned the effort and left a warning to all future researchers that both files at the Murray Center were so seriously flawed that they could not be used (pg 538, emphasis in the original)."

Translation... it's not my fault that my sensational statistical claims are rubbish. By submitting a cruddy data file to Murray, she contemptuously disregarded the archival mandate issued by her funders.

All I can say is, wow.... Weitzman did this study at a time when it was necessary to have programing and computer consultants. It's more than a little convenient for her that these consultants were responsible for the errors.

But the sad fact is that it seems to me that many contemporary researchers exercise a comparable degree of carelessness in the documentation of their data. When I was at ICPSR, I worked with a high profile data file where more than 3/4 of the variables in the dataset did not map to the hard copy documentation. Despite our best efforts to communicate these inconsistencies, the PI could not be bothered to investigate the source of these problems, meaning that the public did not get access to a resource constructed with their resources. Equally serious in my view is that nobody is able to replicate the PI's findings.

Anonymous said...

This suggests that good data are what researchers seek. In fact there is a substantial distinction between what researchers consider "good" data and what an archivists or the public considers "good" data. Replication is not really the aim ---hence, any old sloppy practice will suffice. I will wager (with many days spent at the table in Vegas) that 30-40% of sensational findings in social science are data, coding, or programming errors. And we shall never discover them because our disciplines are organized around narrative not data (despite any claims to the contrary).

Anonymous said...

Would I be wrong to assume from all this that there is a freeseian takedown in the works? Will we hear about it here first? Will it be carnal?

jeremy said...

The Peterson takedown of Weitzman was quite enough, even if it was long in coming given various obstructive practices concerning access to the data.

Anonymous said...

Bummer. Just an idle daydream that this thread + earlier cryptic reference to angry rejoinder sent to journal = gen-u-wine hawkeye barn-burner.

Ookami Snow said...

googlestalking - I don't know why I haven't heard of that word before but it is great.

Tom Volscho said...

From Wikipedialrbbg

"As a sociologist, Goode relies heavily on informants for his research. For example, Goode consulted with and interviewed actual drug users for his books on drugs. In 1999, Goode admitted through the sociology journal circuit that he had engaged in sexual intercourse with many of his deviant informants, and discussed how this influenced his perspective on the subject he was studying. This caused a firestorm of articles defending or denouncing his work [2][3]"