With recent successful exit polls in Montana and South Dakota, Edison Media Research concluded the busiest primary season in the history of exit poll research. In the five months between January 3rd and June 3rd, Edison staffers covered primaries and caucuses in 39 U.S. states and Puerto Rico, providing election analysis and data to ABC, CBS, CNN, FOX, NBC and the Associated Press. Edison data was used to track the demographics, attitudes and opinions of voters in both Democratic and Republican contests, interviewing more than 80,000 respondents over 18 different election days. Throughout this uniquely complex effort, exit polling data was provided in real time without delays or disruptions for the members and subscribers to the National Election Pool. Despite the complexity and logistical challenges of conducting this number of interviews during the primary season, the Edison team projected winners with 100% accuracy.

"This has truly been an historic campaign season," noted Joe Lenski, Edison's Executive Vice President and director of exit polling efforts for the National Election Pool. "I am incredibly proud of the effort everyone at Edison put into these exit polls. Thanks to them, journalists and historians will have a lasting record of this unique race, and we truly look forward to seeing this data analyzed for many years to come."

Edison now gears up to replicate this effort, as the company prepares to conduct exit polling nationwide for the General Election on November 4th. The exit polling effort for the 2008 Presidential Election will be one of the largest single-day survey research projects ever conducted.

Edison Media Research is pleased to announce that Edison Executive Vice President Joe Lenski has been elected to be the Vice President and President-elect of the New York chapter of the American Association for Public Opinion Research (AAPOR). Lenski will begin serving as Vice President this summer and succeed president Anthony Salvanto (CBS News) in 2009 as the head of the New York chapter of the nation's leading organization for public opinion and survey research professionals. Lenski has been an active contributor to AAPOR both nationally and with the local New York organization, the largest of AAPOR's seven national chapters. Lenski leads Edison's exit polling efforts, including the company's work with the National Election Pool as the sole provider of exit polling data for the six major news networks and other subscribers. He is a globally recognized thought leader in the fields of exit polling and survey research, and one of the founders of Edison Media Research in 1994.

The best predictions of the Democratic Presidential Primaries may have been made way back on February 6^th, 2008 --the day after the Super Tuesday primaries.  On that date, a set of predictions was leaked to Bloomberg News showing the popular vote and delegate breakdown for all of the remaining contests from February 9^th through June 3^rd.  It proved prescient again last week when the unofficial results from Pennsylvania showed Clinton winning 84 delegates and Obama winning 74 delegates - the prediction from February 6^th had Clinton winning 83 to 75.

So who is responsible for this nearly dead-on prediction?  The Obama campaign itself, as you can see at this site or download the spreadsheet yourself.

And Pennsylvania has not been the only spot-on prediction that was made in this spreadsheet.  It has picked the winner correctly in every contest since February 5^th except for one - Obama claimed the Maine Caucuses even though this Magic 8-ball spreadsheet had Clinton squeaking out a win.

Almost as importantly, it has been right on the button in the delegate distribution in many states - the spreadsheet predicted the exact delegate breakdowns in Rhode Island, Vermont, Wyoming and Mississippi; was only off by one delegate in Ohio and Nebraska; by two in Wisconsin; and by three in Louisiana, Washington, Hawaii and the District of Columbia.

Now some might say that this is a self-fulfilling prophecy, with the Obama campaign investing money in places where it expects to do better to begin with.  However, the Obama campaign is not limited in its fundraising and seems to be spending as much as possible in every state. What this says to me is that the campaign had realistic expectations in every state after February 5^th and it created a plan and a goal that the campaign subsequently achieved in each of these states.

So what does this spreadsheet prognosticate for the next round of primaries on May 6th? Seems like the Obama campaign is predicting a good day for the Obama campaign:

Indiana - Obama 39, Clinton 33;

North Carolina - Obama 61, Clinton 54. 

With its track record so far I wouldn't bet against the spreadsheet.

I am always fascinated by the results at Google Trends, Google's own tracker of relative search activity. Entering in our top three candidates at the moment, we see that Barack Obama would win the primary in Googlestan hands down:

The letters on this graph correspond to news stories that came out the same day of the spike, much in the same way that Wall St. analysts attempt to ascribe troughs and valleys in stock prices to external news events. I'm not sure these spikes correspond to anything more than increased search activity caused by increased news coverage in general, and not to any specific news article, but the fact that the spikes occur on some very prominent days (Super Tuesday, for one) provides empirical evidence that the graphs are not a "random walk" but do in fact provide a mirror of the interest of some subset of the population.

What subset is, of course, an open question. I'd love to dig deeper, but Google frustratingly/wisely does not give us datapoints, or even the y-axis, so we can only look at these relative indicators of search activity and wonder. Recent data released by Hitwise suggests that Google users are more well-heeled than users of other search engines, an income group that has been a strength for Obama. Yet the same data also reveals that the groups that index highly for Google tend to be 55+, and definitely older than groups that index highly for Yahoo!, which suggests a pro-Clinton lean. Certainly, whatever the biases of the population of Google search users, Obama outperforms the real vote here in Googlestan (at least, by my guesstimation of the relative lift of these lines, anyway) and, in fact, did extraordinarily well in states that he actually won. Here are the top ten states for Obama search activity in February:

Does the fact that Obama "outsearches" Clinton in Texas and Ohio portend victory? Clearly we can't tell from this--consider the margin of search activity in Missouri and the actual margin of the Missouri Democratic Primary, which was one percentage point. Consider also that even when we rank the states in order of search activity for Clinton, the past 30 days would indicate that the Senator from New York didn't even come close in her home state, which she actually won quite handily:

So what conclusions can we draw from this? Is it a comment on the online population? Doubtful--since online access is nearly ubiquitous. Perhaps, given the inverse relationship between the "Search Performance" of Obama, Clinton and McCain it is merely a function of how much available information voters have about each candidate--in other words, an inverse function of how long each candidate has spent in the public eye. In any case, Google--in case you, well, Google this and read my question--how about giving us a y-axis for this chart!

Senator Clinton has enjoyed widespread support from voters living in union households since the beginning of the primary season in Iowa. In fact, even in states that were extremely close, such as Missouri, Clinton's support amongst White voters living in union households far exceeded even her advantage with White voters in general. In the exceptionally close race in Missouri, 57% of White voters selected Clinton, while 69% of Whites living in union households did the same. In New Mexico, another close Super Tuesday primary state, the gap was even more notable, since Clinton actually trailed Obama with White voters overall (43% to 54%) but beat him with White union voters 52% to 44%.

We took a look at the White union vote from Iowa through to Wisconsin (including MO and NM as representative of Super Tuesday, as they were extremely close in the total vote). Clinton's dominance with this voter group was initially split with Edwards, but Clinton took a clear lead in New Hampshire and did not relinquish it--until Wisconsin. Obama's slight margin of victory with this subset of voters--his first-- was perhaps the clearest signal that the Clinton campaign had suffered a serious setback in Wisconsin. After the primaries of March 4th, we'll revisit this question to see if Wisconsin was an outlier, or the canary in the coal mine for Clinton's core union support.

Since the National Election Pool, on whose behalf we conduct the Exit Polls, put the question of listening to 'conservative talk radio' onto their surveys of Republican voters last week in Virginia and Maryland, much discussion has ensued. Indeed, McCain got his worst percentages in both states among those who listen frequently to conservative talk radio.  Also, many talk radio listeners don't find McCain 'conservative enough' on the issues.

But another look inside the numbers shows that while McCain might not be the perfect candidate for the most frequent conservative-talk listeners, he might not be the disaster some people make him out to be. We also asked this question: "No matter how you voted today, how would you feel if John McCain wins the nomination: Very Satisfied, Somewhat Satisfied, Somewhat Dissatisfied, or Very Dissatisfied?" And while conservative talk listeners may have preferred other candidates, they are pretty strong in their satisfaction with McCain.
 
As the nearby graph shows, fully 67% of frequent talk listeners in Maryland said they would be satisfied to some degree with McCain winning the nomination.  While this is a tad lower than the 77% overall who said they would be satisfied, it is still an overwhelming majority.

In Virginia, an even higher 74% of frequent talk listeners will be satisfied by McCain, only a spot less than the 77% overall there expressing satisfaction.
 
So when you listen to talk radio, and hear callers who say they would 'never vote for McCain', understand that they seem to represent only a small minority of even the frequent talk listeners.

The Superdelegate Transparency Project Portal is a wiki that purports to have all the answers. Feel free to examine, contribute and/or correct where you can.

Much has been made over the past few weeks of the possible influence of conservative talk radio upon the race to decide the 2008 Republican Presidential Nominee. While John McCain now appears to be the candidate, there has been a considerable backlash against McCain from several notable conservative radio talk show hosts. Have Rush Limbaugh, Sean Hannity and others had any impact upon the hearts and minds of Republican voters? With the help of the National Election Pool, Edison decided to find out--and the results may surprise you.

Read "Does Talk Radio Make A Difference" here.

On Super Tuesday, Mitt Romney won the Utah Republican Primary with an overwhelming 90% of the vote. Based upon the high Mormon composition of the state, it is a given that Mormons overwhelmingly voted for Romney, and in fact 94% of Mormons did so (John McCain and Ron Paul split the remaining 6%). It is tempting, then, to view Utah's Mormon voters as a monolithic bloc, and indeed on first blush this appears to hold true. Certainly on "values" issues we see incredible homogeneity--91% of Utah's Mormon voters believe Abortion should be illegal; 90% attend church at least weekly; only 5% described themselves as "liberal."

Digging deeper, however, we find some clear divisions amongst Mormon voters on several key issues. On the immigration issue, 33% of Utah's Mormon voters believe there should be a chance to apply for citizenship, while 34% believe illegal immigrants should be deported. The divisions on this issue cut across age, gender and income--in fact, the best predictor of this was political philosophy: 50% of the "deporters" described themselves as "very conservative," compared to only 25% of the folks who chose a path to citizenship.

Also, while nearly two-thirds of Utah's Mormon voters were positive about the Bush administration, 35% were either "dissatisfied" or "angry." These voters overwhelmingly chose the economy as the top issue facing voters this year (67%) though economically they profile very similarly to those voters who were satisfied with the Bush administration. Lost in that concern for the economy, however (which many voters shared) was a sharp disagreement over the Iraq war: 90% of Utah's Mormon voters who approve of the Bush administration also approve of the Iraq war. This contrasts very sharply with the nearly 50% of "dissatisfied" Mormon voters who do not approve of the war. These anti-war Mormon voters were three times as likely to identify themselves as "Independents" than as "Republicans."

We'll never know, of course, how well Rudy Giuliani might have performed on Super Tuesday had he campaigned aggressively from Iowa onwards, but we do have some sense of the residual support Giuliani had on Super Tuesday, and even more interestingly, where some of those supporters may have gone. In a cross-survey analysis of the 15 Republican races we covered on Super Tuesday, Giuliani received 2% of the vote. We asked two questions across all the primaries we covered on Super Tuesday that specifically named the former New York City Mayor: which candidate is best qualified to be Commander-in-Chief, and which candidate is best qualified to manage the economy. Though 2% cross-survey voted for Giuliani, 7% named him for the former qualification, and 6% for the latter. Amongst the voters who named Giuliani for these two questions, the actual vote distribution is quite interesting:

Actual Vote of Respondents Selecting Giuliani as "Best Choice to Manage Economy (6% Of Sample)"

Actual Vote of Respondents Selecting Giuliani as "Best Choice for Commander-in-Chief (7% Of Sample)"

Overall, Romney was judged to be the best choice to manage the economy (39% cross-sample, compared to 32% for McCain), but those who thought Giuliani would best manage the economy broke much more for McCain, perhaps as a result of his endorsement of McCain's candidacy. Yet, on the "Commander-in-Chief" question, of those who thought Giuliani would make the best CinC, slightly more actually tended to vote for Romney than McCain. What makes this more notable is the fact that the 39% of Giuliani voters who disapprove of the U.S. war in Iraq much more closely resembles the profile of McCain voters (37% disapproval of the Iraq war) than the profile of Romney voters (only 22% disapproval).

Incidentally (and not necessarily related), while McCain had the highest percentage of Veterans amongst his voters (23% of those who voted for McCain cross-survey have served in the military). Giuliani had the lowest Veteran composition--only 14% of his voters indicated they had served.