I am writing today to announce the closure of the New Mexico Independent. After three and a half years of operation in New Mexico, the board of the American Independent News Network, has decided to shift publication of its news…
While environmental activists played their part yesterday during demonstrations at the capitol building, going so far as to dress up as solar panels and to sing the tune of “You Are My Sunshine,” their counterparts, the anti-cap-and-trade contingency who has…
Roughly one quarter of University of New Mexico students are unimpressed with the state’s flagship public school, according to a survey that questioned college students about their higher education experiences.
Markos Moulitsas, the founder of the liberal political blog Daily Kos, announced today that he’s filing suit against his pollster, Research 2000, alleging “that the weekly Research 2000 State of the Nation poll we ran the past year and a half was likely bunk.”
Moulitsas cited the work of three statisticians who questioned the polls’ accuracy.
The weekly polls asked about the favorable ratings of a number of political figures, including President Barack Obama. The poll also looked at whether respondents thought the country was on the right track or wrong track and whether they’d vote for a Democrat or a Republican.
Statistician and polling expert Nate Silver, who runs the blog FiveThirtyEight, said of the report by the statisticians is “highly compelling and it confirms other oddities that I had detected in Research 2000′s polling.”
Research 2000 President Del Ali told TPM Muckraker in an e-mail that he would refer all questions to his lawyer but added, “I will tell you unequivocally that we conducted EVERY poll properly for the Daily Kos.”
Among other things not easily comprehended by non-geeks, the statisticians noted that if 55 percent of men responded one way to a question, the percentage of women in that same column would almost ALWAYS also be an odd number. That’s nearly impossible and it’s one of the reasons they question the accuracy of the poll:
…the even-odd property should match about half the time, just like the odds of getting both heads or both tails if you tossed a penny and nickel. If you were to toss the penny and the nickel 18 times (like the 18 entries in the first two columns of the table) you would expect them to show about the same number of heads, but would rightly be shocked if they each showed exactly the same random-looking pattern of heads and tails.
Were the results in our little table a fluke? The R2K weekly polls report 778 M-F pairs. For their favorable ratings (Fav), the even-odd property matched 776 times. For unfavorable (Unf) there were 777 matches.
Common sense says that that result is highly unlikely, but it helps to do a more precise calculation. Since the odds of getting a match each time are essentially 50%, the odds of getting 776/778 matches are just like those of getting 776 heads on 778 tosses of a fair coin. Results that extreme happen less than one time in 10228. That’s one followed by 228 zeros.
The Independent has reported on Research 2000 polls in the past, though not on the weekly tracking polls that were the subject of the statistician’s report.
“Based on the report of the statisticians, it’s clear that we did not get what we paid for,” Moulitsas wrote at Daily Kos today.
