Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: bentway who wrote (734203)	8/22/2013 5:47:19 PM
From: puborectalis	Respond to of 1579770

Prominent pollster Scott Rasmussen has left the company that bears his last name, Rasmussen Reports LLC, in what the company said was over business strategy "disagreements." But as it stands, Rasmussen as a pollster seems to be continuing down the same shaky path that it has been following for the last two election cycles. In fact, the company even assured readers in its statement that Scott Rasmussen's "legacy remains intact." It even touted his polling methodologies as "widely acknowledged as among the most accurate and reliable in the industry," saying it would continue to employ them in the field. This is a problem. In 2010, Nate Silver provided a lengthy takedown of how Rasmussen's misjudgment of the electorate caused its results to be firmly skewed (skewed!) toward Republican candidates. In 105 polls of Senate and gubernatorial races released over the last three weeks of the 2010 election, Rasmussen's results came out to overestimate the Republican candidate's standing by 4 points. In 2012, the results were just as bad — Rasmussen's polls overestimated Republican presidential candidate Mitt Romney's standing by about 4 points, again. The poor results went back to Rasmussen's methodology. It used an online panel and automated robo-calling, and it weighted its results according to a badly misjudged electorate. They don't call people who use cell phones. Scott Rasmussen told Business Insider back in November that he projected the electorate would be similar to 2008, when white voters cast 74% of the ballots. Latinos, African-Americans and other minorities would make up the other 26%. The real split turned out to be 72-28. It ranked 24th out of 28 pollsters in terms of accuracy. "The Obama campaign made clear all along that they believed the electorate was going to be 28% minority," Rasmussen said. "They were right on the money. They nailed it." But Rasmussen hasn't undergone the same review of its polling practices as Gallup, which undertook a thorough review and announced changes to its methods in early June. As The New Republic's Nate Cohn points out, Rasmussen's answer to its problems was to simply "reweigh their tracking poll to the 2012 exit polls, even though you really, really can't do that." In short, until Rasmussen is willing to undertake a more comprehensive reform of its methodologies in the wake of Rasmussen's exit, it's still the same Rasmussen. And it's not likely, right now, that the firm will be any more reliable come 2014 or 2016. Read more: businessinsider.com