nimh wrote:I'm wondering how different the pic might look if they'd identified the trendline for the Bush support from March rather than from May onward. Envisioning it, I'd say it would still give the advantage to Bush, but a lot less drastically so.
What site is the graph from?
Never mind, found it. It's from
here.
Somehow my inkling of a degree of partisanness involved in the analysis seems confirmed by what I read there.
The maker claims, first, that Kerry's high numbers in August were the result of "the polls [..] being deliberately gimmicked, in hopes of helping Kerry", namely through "an attempt to engineer a 'post-convention bounce'" that "failed and was abandoned after about two weeks".
He then claims that Kerry's
low numbers in September were
also the result of "a deliberate attempt" to help Kerry, by "depress[ing his] numbers, so as to set up an 'October comeback'" - "Of course, the goal was to engineer a bandwagon."
Hmmmm ... we're talking a variety of up to (in my last count) 35 different polls by different media and research institutes with different leanings - but both the boost they showed for Kerry in July/August
and the one they showed for Bush in August/September were the result of an elaborate media conspiracy to help the Democrat?
I saw that Daly expressed respect for the guy even while disagreeing with the conspiracy theory. I'm not that patient. Anyone who spins tales like that is way off into wacko zone.