[Wayne07]

"So, these people believe that Texas is an R+2 state?"

No, that's just the breakdown of who answered the phone. They then weight the poll to make up for it.

This survey uses statistical weighting procedures to account for deviations in the survey sample from known population characteristics, which helps correct for differential survey participation and random variation in samples. The overall adult sample is weighted to recent Census data using a sample balancing procedure to match the demographic makeup of the population by county, gender, age, education and race. When including the design effect, the margin of sampling error for this study of registered voters is +/- 3.4 percentage points.

[Political Junkie Too]

Don't they usually over-sample and then create a sample from contacted people that reflect the demographics, instead of just taking what they get and then apply weighting sauce to it?

In other words, shouldn't they call 3,000 people and then take the 1,000 that meet the demographic measures, instead of calling 1,000 people and then weight the responses?

Otherwise, their weighting factors only assume how a respondent will reply, as if they are automatons displaying group herd mentalities.

-PJ