Syndicated Columnist Michael Barone says the Republicans may have a huge margin of victory in the take over of the House of Representative in November. However, Barone says the Gallup polling is volatile and shaky at best.
From the Washington Examiner:
Late yesterday, Gallup came out with new numbers on the generic ballot question—which party’s candidates would you vote for in the election for House of Representatives? Among registered voters Gallup shows Republicans ahead by 46%-42%, about as good a score as Republicans have ever had (and about as bad a score as Democrats have ever had) since Gallup started asking the question in 1942.
However, Gallup also shows the results for two different turnout models. Under its “high turnout model” Republicans lead 53%-40%. Under its “low turnout model” Republicans lead 56%-38%.
These two numbers, if translated into popular votes in the 435 congressional districts, suggest huge gains for Republicans and a Republican House majority the likes of which we have not seen since the election cycles of 1946 or even 1928. For months, people have been asking me if this year looks like ’94. My response is that the poll numbers suggest it looks like 1994, when Republicans gained 52 seats in a House of 435 seats. Or perhaps somewhat better for Republicans and worse for Democrats. The Gallup high turnout and low turnout numbers suggest it looks like 1894, when Republicans gained more than 100 seats in a House of approximately 350 seats.
Having said that, caution is in order. Gallup’s numbers tend to be volatile. Its procedures for projecting likely turnout are very sensitive to transitory responses. They’re useful in identifying shifts in the balance of enthusiasm. But they can overstate the swings to one party or the other. Scott Rasmussen’s latest generic ballot numbers among likely voters show Republicans with only a 45%-42% lead, much less than the 48%-38% lead he reported two days ago. That’s based on a three-day average, indicating Democrats fared relatively well on the most recent night of interviewing. Perhaps Barack Obama’s attempts to gin up enthusiasm among Democratic voters are bearing fruit. Or perhaps one night’s results were an anomaly. Polling theory tells us that at least one out of 20 polls is simply wrong, that is, the results differ from what you would get from interviewing the entire population by more than the margin of error.