Reply #43: Hi TIA. I'm no fan of Mitofsky, but I'll try.

[jkd]

The raw data must be weighted because this data is not based on a truly random sample. The precincts are chosen randomly, and then those polled at these precincts are selected in the same manner. To be random all the voters would be in the mix, not just those in those predetermined precincts.

Howard B. Christensen, Chairman of the Department of Statistics at Brigham Young University, relates the following: “A simple random sample is one selected such that every possible sample of the same size has the same probability. To implement it requires a "list" of the elements in the population; in the case of an exit poll, it would require a list of all of the voters who turnout. Since this list doesn't exist, a simple random sample cannot be conducted in an exit poll setting.”

The weighting of the exit polls simply attempts to adjust the totals to a true state or national sample based on their demographics. The raw data in Utah gave Kerry 29% of the vote, but when the raw data was weighted Kerry got 26%. Some of the voters were weighted eight times greater than other voters based on their profile.

I have asked NEP about their formula to weight the
exit polls with no response from them. The 2004 results are off more than previous exit polls. Mitofsky insists that their methodology is not at fault, but how is one to know.