On 'Modeling' and Surveys
Mark Blumenthal | June 7, 2010
Topics: Internet Polls , Likely Voters , Michael Traugott , Mick Couper , Probability samples , Robert Groves
My National Journal column for this week looks at the "modeling" that underlies both online panel surveys and all pre-election polling. Please click through and read the whole thing.
By Mark Blumenthal | June 7, 2010 9:22 AM | Permalink | Comments (2) | TrackBacks (0)
Comments
I think our number one issue Obama needs to take care of now is immigration reform. He needs to make a trip to AZ and Texas and reach out to the HIspanic community and than take suggestions he gets to the senate floor. Immigration reform could happen this year, and I think it could save the Dems in November and also show the president cares as much about border security as Gov. Brewer, yet he will work on legislating it through more democratic means.
If he doesn't at least get the process started before the fall, Obama's presidency could be in serious jeapardy.
Posted on June 7, 2010 1:38 PM
I hope you have the chance to further discuss the substantive assumptions behind the modeling for internet, automated, and RDD surveys, as well as behind likely voters. If Mick Couper is right (and I think he is), what can we watch for in terms of sampling assumptions that are specious and those that are probably good.
Posted on June 7, 2010 5:14 PM
Post a comment