Star-Ledger/Eagleton Rutgers Poll
Here’s what we don’t understand. The poll supposedly surveyed 600 registered voters and included this question:
“If the election for governor were held today, for whom would you vote Jon Corzine, the Democrat, Doug Forrester, the Republican or some other candidate for governor?”
The number of males surveyed was 282 and the number of females was 318 for a total of 600. Yet the total surveyed by party, including independent (unaffiliated) for this question, according to the poll was 540. What happened to the other 60 people?
We also don’t understand the sample weight by party affiliation used in the survey. The chart below shows the breakout of registered New Jersey voters by party declaration. How do the numbers by party surveyed by the Star-Ledger/Eagleton Rutgers Poll relate to the actual population?
Thirty-nine percent surveyed by the poll were Democrats, 28 percent were Republicans and 32 percent were independent. Is this a properly weighted poll sample? Are these poll results really meaningful? President Bush received 48% of the vote in New Jersey in the 2004 election, is this fact relevant?
Perhaps someone can enlighten us.
Registered New Jersey Voters By Party and Number Surveyed
Unaffiliated - 2,611,586 (56.48%)...Surveyed by poll - 176 (32.6%)
Democrats - 1,130,812 (24.46%).....Surveyed by poll - 211 (39.1%)
Republicans - 865,237 (18.71%)......Surveyed by poll - 153 (28.3%)
Other – 16,459 (0.35%)...................Surveyed by poll - 0 (0%)
Total Registered - 4,624,004...........Surveyed by poll - 540
1 Comments:
The sample is 0.017% of the total registered Republicans, 0.018% of the Dems. Unaffiliateds are under-represented at 0.006%. That could be significant, because by my calculation the split on unaffiliateds between GWB and Kerry was < 2%.
Post a Comment
<< Home