autumnus: A purple monochrome portrait of Zoe from Dreamfall, with drawn stars in background and "the Dreamer" written on bottom. (Default)
So, apparently while I was busy with my own life last week, the fan community went into a bit of a chaos.
The story involves

two so-called-researchers who are no longer really affiliated with BU

an homophobic, biased, and ethically dubious research that involves an online survey posted in several adult fanfic hubs, about sexual preferences and fanfic.

a transformative works (or fanfiction/fanart) community that is rightfully angry at being treated like freaks.

Below articles tell the story.

Fanlore Article about SurveyFail
The curious case of the game show neuroscientists, or how NOT to research an online community
Sex, Lies and IRB Tape: Netporn to SurveyFail

If you want more details, a link list is here:

SurveyFail Masterpost


Honesty I am not a writer, I don't even consider myself a blogger most days, so I am not even going to try to summarize the events in a coherent way. I haven't been there, and the links tell the story clearly.

However this is too horrifying to not mention. Gender issues aside, how can two Ph.D. graduates can come up with a "research" that is so ethically unsound and badly constructed is beyond me.

Several articles state that these guys are computer scientists, as if it is an excuse. It is not. Basic idea of how to conduct a user study of any form is something one learns in grad school. If you are dealing with human subjects you learn it, and more importantly you always pass it through an IRB committee, for good reason. I am a computer science grad student, MS, not even doctorate. I learned this from an undergrad class during my first semester.

I really hope these guys are two unusual cases and that there will be consequences for what they did not only within our field, but also legally,  that the data they have will not be allowed to be used, not even for a pop-psychology book.  For one thing there should be implications about doing surveys with minors without parental consent, especially when asking questions about subjects like rape fantasies, or even their sex lives or drug use, with no apparent safety protocols in place.

I really hope these guys are aberrations, because if not, I really don't know what that means about the credibility of computer scientists. In the mean time how did these two even earned the title "Dr." should be asked. I almost feel bad for BU at this point, they don't really deserve this kind of a PR nightmare.

Now we have to see what Penguin does, the idealist part of me wonders if they will rise above the petty greed and do the right thing and shot down the project.

Yikes.

Edit: Thinking more calmly, we do force students to take writing classes to learn how not to plagiarize. Action such as done by these two people can potentially hurt more people that breach of ownership might. Maybe a similar approach is required for human research and ethic as writing: undergraduates should be required to take an human subject research and ethics class regardless of their major.


Date: 2009-09-07 04:56 pm (UTC)From: [personal profile] codeman38
codeman38: Osaka from Azumanga Daioh, with a speech bubble reading 'Contemplation No. 1'. (contemplation)
From another CS grad student... word.

I had to go through human subjects training for a human-computer interaction course that involved surveying other students on how user-friendly they found an interface design. Interesting thing is, though, that wasn't actually a required graduate course, so it is possible that someone could do an entire CS grad program and not encounter that, if their thesis also didn't involve human subjects.

Date: 2009-09-07 05:12 pm (UTC)From: [personal profile] jalendavi_lady
If nothing else, there needs to some sort of warning everyone gets about warning signs a human subject project is something to stay away from. Stick it in a required basic science course at some point in high school, or in a social studies course.

I knew things looked funky about the survey once I heard what was going on because I'd been used as a survey subject as a PSY 101 student and knew what a consent-with-privacy form for a proper study was supposed to look like. My college classmates who filled that particular requirement with a non-psych course were never survey subjects.

18-year-olds in high school or people who don't go to college aren't going to have the benefit of any required undergrad course, and neither of those groups will be protected by keeping surveys like this away from minors, either.

Date: 2009-09-07 05:14 pm (UTC)From: [personal profile] elf
elf: Quote: "I found a special snowflake;" 5-point snowflake (Special Snowflake)
how did these two even earned the title "Dr."

Their degrees are in visual cortex info & AI; they're mainly computer scientists who study how to make computer circuits act more like brain circuits. They don't have any education whatsoever in how people act.

It's reasonable that their understanding of psychology & culture is useless, kind of like how you wouldn't expect a PhD in history to have any understanding of suspension bridges even though many suspension bridges have had important historical impacts.

The fun part (and by "fun" I mean "reason we want to spork our eyes out when we read their posts") is them thinking that because they study "human brains" they know something about human behavior.

Um. Knee surgeons don't necessarily understand why marathon runners compete.

Date: 2009-09-08 03:06 am (UTC)From: [personal profile] starlady
starlady: "They don't play by the rules, why should we?" (dumbledore's army)
Maybe a similar approach is required for human research and ethic as writing: undergraduates should be required to take an human subject research and ethics class regardless of their major.

I think you're exactly right.

Profile

autumnus: A purple monochrome portrait of Zoe from Dreamfall, with drawn stars in background and "the Dreamer" written on bottom. (Default)
autumnus

January 2016

S M T W T F S
     12
3456789
1011 1213141516
17181920212223
24252627282930
31      

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Sep. 23rd, 2017 09:33 pm
Powered by Dreamwidth Studios