Results from automated polling reflect GOP-leaning effect

Follow the latest tweets from Georgia’s campaign trail @ajcgaelections

Rasmussen Reports, SurveyUSA and InsiderAdvantage all use automated polling equipment for their surveys – i.e., robo-calls. So over the next four days, you might keep in mind this observation from Nate Silver at fivethirtyeight.com:

Most of the automated polling firms have a Republican-leaning house effect. For instance, it’s about 2 points for Rasmussen Reports (our estimate for Rasmussen includes polls conducted by its subsidiary, Pulse Opinion Research) and 4 points for SurveyUSA.

Another automated polling firm, Public Policy Polling, has almost zero house effect. But some of the smaller robopoll firms, like Magellan and Merriman River Group, also have a Republican-leaning effect.

On average, the robopoll firms have a 2-point Republican-leaning house effect, whereas the live interviewer polls have a 0.7-point Democratic-leaning house effect. The difference between the two, then, is 2.7 points.

Get an insider’s take on Georgia’s political scene from Jim Galloway’s Political Insider blog

4 comments Add your comment

south georgia possum

October 29th, 2010
2:20 pm

If these polls give republicans any edge at all…you can believe Jim G and the AJC won’t print very much of their content.

Ralph

October 29th, 2010
4:10 pm

More AJC wishful thinking to use a blogger who says robo calls are Republican leaning. Almost ALL pollsters apply statistical corrections to their results that end up with likely voters. Rasmussen and SurveyUSA have good track records, where InsiderAdvantage does not (until close to the election when they take out leftist bias so they don’t look quite as silly).

Next week we will see how accurate the polling is – and the AJC will be very quiet about that.

There are many criticisms about Nate Silver’s poll ratings, but this one is at the heart of it:

[Silver needs to] broaden the scoring of poll accuracy beyond the final poll conducted by each organization before an election. He includes all polls with a “median date” (at least halfway completed) within 21 days of the election. As he writes, we have seen some notable examples in recent years of pollsters whose numbers “bounce around a lot before ‘magically’ falling in line with the broad consensus of other pollsters.” If we just score “the last poll,” we create incentives for ethically challenged pollsters to try to game the scorecards.

This is why the InsiderAdvantage poll is worthless until the last week, when ALL polls usually align to reality. Rasmussen has much higher ratings on other poll ratings than Silver’s and has been consistent while others are now just removing their bias.

Those who want to delude themselves that these polls this week are wrong will be deeply disappointed next Wednesday.

pete

October 29th, 2010
11:22 pm

lets not forget women voters Deal voted against equal pay for women four times.Be sure you do your home work dont set us back

Ralph

October 30th, 2010
9:47 am

Pollsters probably have statistical corrections for conservative people like me who don’t take part in their polls. I vote, but don’t share my intentions with pollsters.

I use caller ID and don’t answer toll free numbers, when it comes up “Private”, strange business names, or out of state area codes unless I recognize the number. If it is somebody I might want to talk to, they can leave a voice mail. This works well avoiding the robocalls from pollsters, political parties, and solicitors. I even have call block where I block a list of numbers from repeat offenders. Isn’t technology great to keep these pests away?

If I accidentally answer and it is a spiel I’m not interested in, I merely interrupt and say “No thank you”, and hang up. It’s not rude – their companies are the ones being rude with their unsolicited intrusions.