Thursday, Oct. 18, 2007 | 7:15 a.m.
Sen. Harry Reid figured he was in pretty good political health after his internal polls showed him in early August with strong numbers: Well over half of Nevada voters viewed him favorably.
Then he woke up to the Review-Journal's story on Monday reporting that more than half the people the newspaper polled don't like him.
If both polls are to be believed, Reid's popularity plummeted in just two months.
Reid protested to the R-J and he went on KNPR's "State of Nevada" public affairs show to protest to whoever would listen . He might be more popular than the R-J would have its readers believe.
"I don't care about polls. I especially don't care about Review-Journal polls," he told KNPR show host Dave Berns.
But a 24-point slide over two months?
"Voters are not that fickle," said Paul Freedman, an associate professor of politics at the University of Virginia. "The challenge of making sense of polling numbers is making sure you're comparing apples to apples."
On the surface, the two polls seem the same. Reid's poll was 600 likely voters. R-J's was 625 likely voters. Both asked whether voters viewed politicians favorably.
But even subtle differences in polls can change numbers.
Reid's poll had voters choose among highly favorable, favorable, unfavorable and highly unfavorable.
The R-J's offered up only favorable or unfavorable options.
"That's a huge difference," Freedman said.
Voters with mixed feelings were forced to choose one way or the other, or go with neutral, which picked up 15 percent of the vote.
Poll results can also shift based on the order of the questions and what else the questioners ask.
Some numbers, such as Barack Obama's high unfavorable-to-favorable ratings and the big number of people with a neutral opinion of Gov. Jim Gibbons, gave Reid's camp reason to question the poll.
"There are clear indications that this is an inaccurate poll," said Jon Summers, Reid's spokesman.
Review-Journal Editor Thomas Mitchell said he stood by the poll's accuracy and fairness but otherwise declined to comment. "I don't think I'm going to chat on the topic," he said. "I don't see the point."
Brad Coker, managing partner of Mason-Dixon Polling & Research Inc., which conducted the poll, did not return a call for comment Wednesday.
Others in politics questioned some results of the Review-Journal poll - including that just 2 percent of Republicans viewed Reid favorably.
"That 2 percent, I just don't believe," said Jennifer Duffy, senior editor of the Washington, D.C.-based Cook Political Report.
But, in many ways, Reid's polling numbers reflect the national mood toward the House and Senate.
"Those numbers look a little like Congress' numbers," Duffy said. "There's no reason for him to be exempt."
In fact, she said, Reid's position as Senate majority leader makes him more vulnerable to being tied to overall disapproval with Congress, because he puts a face on the Democratic Party.
Not only is Reid facing a highly politicized atmosphere , but the liberal wing of the Democratic Party has been unhappy with congressional performance.
"The Democratic base really believed with the 2006 election, Congress would end the war in Iraq and impeach Bush," Duffy said. The reality of a one-seat majority, with a Republican in the White House, has set in. "They're angry."
But lest Reid decide to pull his candidacy - he says he's running for reelection in 2010 - one poll does not a trend make.
Experts like to look at trends in the same polls, and average different polls.
"I look at dozens of polls a week," said Karlyn Bowman, a senior fellow at the American Enterprise Institute. "It's only one useful tool to understanding a complex public."
archive
