Murray’s Poll Means Different Things To Different People
Meanwhile, the APP ( All Phoney Palloney?) read Patrick Murray’s poll and wrote a headline claiming that Pallone is leading Little “by nearly 2-1.” 2-1? Murray said Pallone has a 12% lead. 2-1 would be a 33% lead. 12% is not nearly 33%.
Even my own analysis on Murray’s poll has been bothering me:
Murray classifies a likely voter as someone who has voted in 2 of the last four general elections. Of the 333,519 registered voters in CD-6, 60,053 voted in 2 of the four last general elections; 23,750 (40%) Democrats, 27,791 (46%) Independents and 8,512 (14%) Republicans, according to GOP records of voter activity.
CD-6 has 333,544 registered voters. In the last five elections, an average of 157,000 people voted. The average is 144,000 if you don’t count the 2008 presidential election (254,543 voters) and the 2007 state legislative election(96,950 voters). Yet, only 60,053 people voted in 2 of the last four elections, the criteria Murray used to select “likely voters.” Granted, Murray used a different list than I did to measure who voted in 2 of the last 4 elections. If there is a wide disparity between the two, my analysis of his poll is as flawed as the APP headline. I think I’ll ask him about his list.
Incidentally, Real Clear Politics also rates the CD-3 race between Jon Runyan and John Adler as a toss up. RCP rates Scott Sipprelle’s race against Rush Holt as “Leans Dem.” I find the CD-12 rating hard to believe. I’m not on the ground in CD-12 as much as I like to be, but I wake up every morning to “I’m Scott Sipprelle and I approved this message.”