It’s Mitt…I mean Rick…I mean Newt

If you are anything like us, you are wondering what in the world is going on with the Republican Presidential Nomination process.  Just a few days ago, we learned that Rick Santorum won the Iowa Caucus.   As far as we know, Mitt Romney still won New Hampshire and now, we know that Newt Gingrich won the South Carolina primary.  It seems like every time conventional wisdom concludes that Mitt Romney will be the nominee something happens to change that.

Really though, if you think about it, we should have expected these results.  It may simply be that Republican voters are selecting the most viable regional candidate.  In Iowa, the split may have been because of Rick Santorum’s Pittsburgh roots versus Mitt Romney’s Michigan roots.  In New Hampshire, Romney’s Massachusetts ties made him the favorite son.  In South Carolina, Georgia’s own Newt Gingrich was the obvious choice.

In spite of what the pundits have said, this may prove to be an exciting race to the wire.  It is much noted and frequently lamented that the Republican Party has become a Southern party and that may place Newt Gingrich in an interesting place as the nomination race winds it way through the South.  We would be interested in hearing what you think.

 

New Hampshire Primary Polling Report Card

The 2012 New Hampshire primary has come and gone, and after months of being bombarded with poll after poll predicting how the voters would vote, we finally have our answer: Mitt Romney took first with 39% of the vote, followed by Ron Paul (23%), John Huntsman (17%), Newt Gingrich (10%), Rick Santorum (10%), and Rick Perry (1%). To close our analysis of this event, we thought it would be interesting to look at how close the pre-election polls were to the final vote results.

There were five polls conducted between January 8, 2012 and January 9, 2012, including ARG, Suffolk, Rasmussen, PPP (D) , and UNH/WMUR.  Below we show their predictions along with the actual vote percentage for each candidate.  The results are coded bold when the poll prediction was within 1% of the actual vote, and italic when it was not.

 

 

Romney

Paul

Huntsman

Gingrich

Santorum

Poll Date Poll Actual Poll Actual Poll Actual Poll Actual Poll Actual
ARG Jan 9 37% 39% 17% 23% 18% 17% 10% 10% 11% 10%
Suffolk Jan 9 37% 39% 18% 23% 16% 17% 9% 10% 11% 10%
Rasmussen Jan 8 37% 39% 17% 23% 15% 17% 12% 10% 13% 10%
PPP (D) Jan 8 35% 39% 18% 23% 16% 17% 12% 10% 11% 10%
UNH/WMUR Jan 8 41% 39% 17% 23% 11% 17% 8% 10% 11% 10%

 

Overall, the best predictions came out of the later polls (ARG  and Suffolk), which predicted the correct vote share of 4 of the 6 candidates while narrowly missing on Romney (underestimating his support by 2% in both polls) and completely missing on Paul (underestimating his support by at least 5%). On the whole, the five polls were the most accurate predicting the final vote for Huntsman and Perry, while being the least accurate on the final vote for Ron Paul (every poll missed by more than their stated margin of error).

Truly, if we weren’t such strict graders and held all of the polls only to their margins of error, then they would have predicted all of the vote percentages correctly with the glaring exception being Ron Paul. The question then remains of why predictions for Ron Paul were so incorrect across the board: Is it because he is most popular with youth (as we noted in our earlier post) who are notoriously hard to reach, or is there some other reason?  We would be interested in hearing your thoughts in the comments below.

 

2012 New Hampshire Primary: Trends in Last Night’s Voting

As we did last week following the Iowa caucuses, today we look at the major voting trends from last night’s voting in the New Hampshire primary. In some aspects, we see similar trends as last week: for example, Ron Paul once again dominated the youth (18-29) vote, while Romney dominated among voters with the highest family incomes. In other aspects, however, voting patterns differed from what many of us expected.

The following are some of the most interesting voting trends (using MSNBC polling data) from last night’s New Hampshire primary:

  • Gender: Mitt Romney received nearly 38% of the male vote and 40% of female, distancing himself from Paul who received 26% and 20%, respectively.
  • Age: As at the Iowa caucuses, Ron Paul once again dominated the youth vote by receiving a near majority (47%) of votes casts in the 18-29 age range. This support waned, however, in older age groups: among 30-44 year olds, Paul received 32% (compared to Romney’s 36%); among 45-64 he received 19% (compared to Romney’s 42%); and among 65+ he received only 12% (compared to Romney’s 42%). The other candidates saw less variation across the ages, though both Gingrich and Huntsman received more support the older the age group.
  • Marital status: Perhaps a testament to his massive sex appeal, Ron Paul received much more support from the unmarried—35% supporting him, more than any other candidate—than the married (19%). This result is unsurprising, however, given that Paul garners most of his support from younger (and thus unmarried) voters.
  • Total family income: Romney dominated the upper income bracket, receiving 51% of the vote from voters with family income of $200,000 or more. His support weakened as family income lessened, with a low of 31% from voters with family incomes under $30,000 (compared to Paul’s 36%).
  • Political ideology: Somewhat surprisingly, Romney received a plurality of the votes  in all of the categories of political ideology, including “very conservative”. Among “very conservative” voters he received 33% (runner-up with Santorum with 26%), and among “somewhat conservative” and “moderate or liberal” voters he received 48% and 37%, respectively.
  • Support for tea party: Also surprising (and in contrast to the polling data from the Iowa caucuses) is that Romney gained the most support from voters who strongly support the Tea Party movement, with 36% compared to Paul’s 21% and Santorum’s 20%.

The New Hampshire Primary: A Primer

After months of speeches, debates, hand shaking, baby kissing , and all the other political gestures that we’ve all come to love(?), we’ve finally reached the New Hampshire primary—the first primary election held in the United States. Together with the Iowa caucuses (which we profiled last week), the New Hampshire primary is one of the most important events in a presidential election year, creating momentum for some campaigns, while killing it for others. (As we mentioned previously, every candidate that has won both Iowa and New Hampshire has gone on to win the nomination; only George McGovern (D) in 1972 and Bill Clinton (D) in 1992 won the nomination after losing both).

This article provides some background on this event for our readers who (perhaps like ourselves only hours ago) know very little about this primary, and who wonder why such a small state has such a large role in the nomination process.

Background

While primaries are normally either “open” (e.g., allowing all voters, regardless of party identification, to vote in either party’s primary) or “closed” (e.g., allowing only registered party members to vote in their party’s primary), New Hampshire is often called “semi-closed” in that people registered in one party can’t vote for the other, but “undeclared” voters who are not officially registered in either party can vote in either. Like general elections, voting is done entirely by secret ballot, with the state’s 12 delegates being divided among the candidates with the most votes through proportional representation.

Voter Turnout

Compared to the Iowa caucuses, turnout for the New Hampshire primary is very high. In 2008, 527,350 people (51.3% of the total voting population in the state) voted, with 287,557 voters in the Democratic primary and 239,793 in the Republican primary. These numbers are comparable with national voter turnout in a typical general election, which normally hovers around the 50% mark.

For this year’s primary, the turnout will undoubtedly be much lower as a result of having only one contested primary. (Democrats will also have a primary, but the 13 challengers to President Obama are weak, to say the least.) To get some sense of the expected level of turnout in this year’s primary, we have to look back to 2004, when Bush was running for reelection as an incumbent. In that election, 29.2% of the voting population turned out: 219,787 on the Democratic side, and 67,624 of the Republican side.

The following table shows in graphical form the percentage of the total voting population that has voted in the past 3 presidential primaries:

Voter Turnout

Sources: Presidential Primary Election Results, Federal Election Commission, 2000, 2004, and 2008; “Annual Estimates of the Resident Population by Sex and Age for States: April 1, 2000, to July 1, 2009,” US Census Bureau, June 2010.

 

Finally, it must also be noted that the voters who participate in New Hampshire differ from those in Iowa: whereas the Iowa caucuses tend to attract party activists and donors, the “semi-closed” primary rules in New Hampshire has the effect of attracting more independents, even taking into account the fact that the majority of these “undeclared” voters actually self-identify as either Republicans or Democrats (as polling data from the University of New Hampshire has shown).

 

How representative is the population to the US as a whole?

Finally, as we did with Iowa, we wanted to look at the social/economic/demographic characteristics of the population in New Hampshire to get some sense of how representative it is to the US as a whole. (We accept that almost no state is perfectly representative, but it is useful to get a sense of how far a given state is from the mean).

Starting with economics, New Hampshire is clearly better off than most other states in the U.S. According to survey data from the US Census Bureau’s American Community Survey, the median household income of $63,033 (in 2009 inflation-adjusted dollars) far exceeds the US average of $51,425. And like Iowa, the unemployment rate is low relative the rest of the country: at 5.2%, New Hampshire has the 4th lowest unemployment rate in the country at 5.2%, well below the US average of 8.6%.

Turning to demographics, as the next chart makes clear, the population in New Hampshire (like Iowa) is remarkably homogeneous with 94.8% white, followed by Asians (1.9%) and Blacks (1.1%).

Racial Characteristics of New Hampshire

Source: U.S. Census Bureau, 2005-2009 American Community Survey

 

In terms of education, New Hampshire is much more highly educated than the US average, with 90.5% graduating high school (compared to US average of 84.6%) and 32.4% receiving a bachelor’s degree or higher (compared to US average of 27.5%).

In short, in many ways New Hampshire is not representative of the rest of the country. The question is: does it matter? Should candidacies be made or undone based on the preferences of one small New England state? We’ll leave that for you to discuss.