Slashdot Mirror

Then again there is this one http://dailycaller.com/2012/08/23/university-of-colorado-prediction-model-points-to-big-romney-win/ that has been correct every time since 1980 that says Romney will win big.

That model is a joke. It didn't exist 6 months ago. Just look at it! PA for Romney? It's not even considered a battle-ground state anymore and the Romney campaign has pulled ALL advertising and has none scheduled.

It's a peer-reviewed scientific journal article. Don't be a Denier. The science is IN!

. . . despite being a comedy program The Colbert Report appears to exercise "disproportionate real world influence" -- likely due to the "elite demographic" of its audience.

Disproportionate to what? Do other programs with a 1.3 million-person audience not produce similar effects on fundraising?

Are there no Republican fora with Republican audiences that produce bumps for Republican guests and dips for Democratic ones? I wonder what the O'Reilly Factor data look like.

The graphs only show +/-60 days, a time period in which TFA suggests that the appearance on tCR is causing a shift in fundraising. What we need, to get a sense of the impact of tCR is to look at these same candidates in time periods when tCR is not a possible influence, e.g. +/- 120 days.

It's all well and good to look at non-tCR candidates at the same time as the appearance, but we also need to look at tCR candidates at other times.

Even less explicable in the original article is the bump Republicans exhibit in the 45 days prior to their appearance. It's unclear whether they have even agreed to do the show at that time. Certainly, no one is advertising their appearance.

Lacking any explanation of how tCR could cause that pre-bump, we must consider whether the pre-bump might be causing the appearance on tCR. That is, tCR may invite Republicans who are making headlines and getting donations.

If you look at the charts in the original article, the bump starts around two weeks before they actually appear on the show. Which makes me suspect that the Bump is more likely due to the candidate making a round of interviews (of which Colbert is one), rather than it being due specifically to the Colbert interview.

You overestimate the uniformity in thought amongst academics and underestimate their breadth of experience. I can assure you, the academics I've had the pleasure of meeting in rural Africa are well aware of the things you so dismissively think they haven't experienced merely by fact of being academics--and their views are not so divergent from those of the supposedly fat and rich academics in cushy jobs in the U.S.

I would direct you to any APSA event at which you may find illumination on this topic.

http://www.zogby.com/search/ReadNews.dbm?ID=259 Gore up 5.4%

http://www.zogby.com/search/ReadNews.dbm?ID=276 Gore now up 7%

http://www.gsm.ucdavis.edu/visitors_center/news/ha gerty_predicts_gore_win.htm UC Davis "statistical analysis" predicts Gore win

http://www.newsmax.com/archives/articles/2000/11/6 /184917.shtml Zogby predicts Gore win (Nov. 6, 2000)

http://www.bizjournals.com/milwaukee/stories/2000/ 11/06/daily19.html Online (yeah, I know) Harris poll predicts Gore win

http://www.apsanet.org/PS/march01/lewisbeck.cfm Gore post-mortem off by almost 7%

It's not that hard to dig this stuff up.

I quoted from the article you referenced, BTW - I didn't think that was too hard to follow.

Perhaps my initial post was overstated (as are yours, BTW, even moreso) - but again, more precisely, all that *I* had heard from the news was (generally speaking) larger margins for Gore, and it turned into a statistical dead heat - and similar errors in the many other issues at hand during previous elections. My links "prove" that you are a liar (or dishonest) by your logic....even with your subsequent artificial constraint about weighting later polls more heavily, and focussing exclusively on the popular vote for the president.

Are you being dishonest now, because I dug up evidence to the contrary? Hmmm....

Obviously, the polls are often off by more than their statistical margin of error. Sometimes significantly so. Perhaps I was factoring in more local, and other national issues - I did not constrain my observation to national polls of the presidential popular vote. Even still, it is easy to dig up plenty of polls that support that.

I have no intent to deceive, and I am also not a Bush supporter.