Blunt's people need to learn about polling. Being a "likely" voter is subject to social desirability bias more than simply being registered. Can we really trust someone who says they are "likely" to show up on Election Day to necessarily do so? This is probably why Gallup almost always goes with a registered voter sample. Second, there is nothing purported about undecided voters being non-existent. The results are only given for those who have a position on a candidate. It doesn't say there are no undecided voters in the least. Third, there is nothing "predicted" about 37 percent of the voters being men (learn to read!). The only thing it says is that the sample generated garnered 37% of men, which was then weighted to reflect the actual balance of males and females in Missouri. It would be nice if people who actually understood how polling works would comment as opposed to political hacks working as campaign shills. After all, you Blunt people put all kinds of faith in Rasmussen polls--a firm that releases nothing on its weighting procedures--or other polls that don't even report the demographic breakdown of its respondents as a regular practice. I guess, though, that you like all of these other results--which may not even be weighted to be representative of Missourians--because your candidate always comes out a few points ahead.
It's also noteworthy that the problems the Blunt campaign sees in the poll didn't affect the outcome of the Long, Skelton, and Emerson races, which had the outcome that one would expect given the characteristics of their districts.
Since the poll was conducted with Random Digit Dialing. Since most younger people use only cell phones (cell numbers are not included in this type of poll) there are more older respondents. Again, though, the data is weighted to compensate. The poll results gained from this sample can be generalized to the larger state population with a 95% confidence level.
![[Avatar]](/images/avatar26.png){kind=avatar}
Contraption