Monday, January 26, 2004

Weighting the Exit Poll Numbers

My guess is that the pre-weights applied to the exit poll will be in the 60-40 range (60% registered Democrats and 40% Undeclareds). That won't have much bearing on the Kerry-Dean contest (unless, of course, it's close), but it could impact the "space" between John Edwards and Wesley Clark. The more "Democratic" the actual make-up of the primary electorate is, the better it is for Senator Edwards. Registered Democrats in New Hampshire have rejected General Clark.

The weights change as exit data, sample precincts and actual vote interact within the NEP's "modeling." So what might begin as a 60-40 designation can end up, at the end of the evening, as a 52-48 split. But 60-40 seems about right to me as a starting point. It will be interesting to see how many Republican-leaning Independents vote in the primary and whether they vote for Dean in large numbers (relative to their small percentage of the total electorate).

The end or the rebirth of the Dean campaign will be found in what as known as the Seacoast region (southeastern New Hampshire) and in the towns along the Vermont border around Dartmouth College. If Kerry beats him in both places, then it truly is over for Deano.