[lbo-talk] AAPOR on New Hampshire Polls

Wojtek Sokolowski swsokolowski at yahoo.com
Thu Jan 10 13:09:21 PST 2008


--- Marta Russell <ap888 at lafn.org> wrote:


> Marcy wrote:
> UPDATE 11:06pm PT: As we know, the presumption is
> always that the
> polls were wrong. Never the results. Despite how
> much less
> transparent the system used to count votes is than
> the system used to
> collect polling data. With that in mind, Matthew

[WS:] I would not put much faith in polls. They are spurios, becaosue they are based on a faulty behavioral model - that ideas and opinions expressed to pollsters predict behavior. However, that is often not the case - especially if social desirability or undesirability is involved.

For the sake of argument, let's assume that racism is widely spread among likely Dem voters. Racist views, however, are not socially desirable, and people who hold such views may want to hide them. One way of doing so is to tell the pollster that they are going to vote for a black candidate, even though they may not actually do so. That may explain the discrepancy in question.

Wojtek

____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs



More information about the lbo-talk mailing list