Pollster.com

Articles and Analysis

 

Moving Day "Outliers"


Janet Elder has a must read on the history of the pre-caucus surveys done by the Des Moines Register's Iowa Poll.

Kathy Frankovic reminds us of the critical difference for presidential candidates between having "the most" experience and having "the right" experience.

Gary Langer examines the differences in the trial heat results in three recent national surveys (including his own ABC/Washington Post survey) and concludes we need to "cut back on fixation with the horse race."

Marc Ambinder looks at how the Iowa Democratic caucus math and process can turn a lead (on a poll) into a tie (on caucus night).

Michael Witney has the details and the text of an email questionnaire sent by the Clinton campaign to its donors (both via Ben Smith).

Mark Penn puts the Clinton campaign's spin on recent statewide polls. Jay Cost annotates.

AP says the Clinton campaign is testing negative ads in focus groups (via The Page).

Albert Hunt's widely linked column on "Tension in Hillaryland" also includes a summary of focus groups recently conducted in Philadelphia by non-aligned Democratic pollster Peter Hart.

Dante Scalia has another batch of media buy reports from New Hampshire's WMUR.

Carl Bialik has the story of an Oregon political consultant who has used a survey of sorts to predict the winner of the Heisman Trophy for the last six years.

And a bit of housekeeping information: We will be moving Pollster.com's world headquarters on Thursday afternoon, so our updates will likely be delayed during the afternoon.

 

Comments
Mark Lindeman:

Amidst the polls, many of us overlooked the "Blumenthal primary," in which cities around the world vied for the honor of hosting pollster.com. Some observers have speculated that the lucky winner could become the epicenter of a regional Knowledge Industry boom -- the next Silicon Valley, but on a Web 2.0 scale. With due respect, that always struck me as a bit far-fetched. Nevertheless, enjoy your new digs!

____________________

Jeff Winchell:

I know long division (though 2 digits hardly seems long) was a PITA for most school children, but when giving a report on the internet, where the reader's can access Windows Calc with a few mouse clicks, why do reporters insist on incorrectly using subtraction when they should be doing division? They often base their entire story on this rudimentary bad mathematical analysis.

Gary Langer's story included this:
"John Edwards was at 14, 11 and 10 percent respectively � similar. Obama was at 30, 27 and 23 percent � a significant difference between the high and low estimates. And in the biggest difference, Clinton was at 40, 44 and 53 percent respectively."

Even before spending a few seconds on Windows calc, I knew his analysis was completely false. But let's look at the numbers:

10/14 = 71%
23/30 = 76%
40/53 = 75%

No difference. No need for that paragraph in his story. Perhaps no need for huge sections of that story, or the story at all.

But hey, it's not about thinking, it's about persuading. And Americans love to do addition and subtraction and hate "advanced" math.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR