I've now read the relevant chapters of _The Impact of Television_,
the naturalistic experiment by Tannis MacBeth Williams.
On Sat, 13 Apr 2002, Stephen Black went:
> I've gone back to my dusty old files to find out why I mistrusted
> the much-cited similar Williams study of the '70s on the
> introduction of TV to an isolated Northern community. I have a
> handout from a symposium at a meeting of the Canadian
> Psychological Association in 1977 where Williams first presented
> her findings (there's a book now, but this is what's handy for me
> at the moment). The results are complicated. True, the community
> ("Notel") initially without TV showed an increase in both verbal
> and physical aggression after the introduction of TV. But
> curiously, the amount of aggression in Notel afterwards was
> _higher_ than for two other control communities, Unitel and
> Multitel, which had long-standing access to TV.
More curiously still, the "Phase 1" levels of physical aggression were
*identical* in all three towns--Notel, with no TV; Unitel, with only
the Canadian station; and Multitel, with the Canadian station plus
ABC, CBS, and NBC. I scanned in the figure, which shows observers'
ratings of children's playground behavior in "Phase 1" and "Phase 2":
http://www.charm.net/~dhe/Williams.jpg
(In "Phase 2," Notel has gained access to the Canadian station, while
Unitel has gained access to a second Canadian station, and Multitel
has retained access to all stations.)
The Williams data suggest that TV can increase children's physically
aggressive behavior two years after its introduction into a no-TV
community. But they don't suggest that any effect remains once TV has
been available for a few more years.
> So if TV was so bad, why did it produce more aggression in the
> community that had it for a short time than for those that had it
> long-term? Shouldn't there be a dose-response effect? Was the
> influence of TV a contrast effect that quickly wore off? Perhaps
> she deals with this in her book, perhaps not.
Answer: not. No mention that I can find.
It's a pity, because the first few chapters of the book are quite
reasonable and persuasive, paying plenty of attention to issues of
causal inference. Prior data are reviewed informatively and with
seeming evenhandedness.
I'll post more after I read the St. Helena study.
--David
---
You are currently subscribed to tips as: [EMAIL PROTECTED]
To unsubscribe send a blank email to [EMAIL PROTECTED]
