Did Under Voting Cost Mount Vernon Schools the November Levy Election? (Part Two)

In my first post on under voting in Knox County, Ohio, I introduced the concept of under voting and discussed patterns of under voting in races in Knox County involving candidates. I found that the Gambier precincts exhibited levels of under voting that were below the Knox County norm in the presidential race, but that under voting rates in Gambier were much higher than the Knox County norm in other races down ballot.

This piece examines the effect of under voting on an issue race, focusing on the Knox County School Levy election that took place on November 6, 2012.

The Mount Vernon School Levy failed narrowly on November 6^th, losing by a margin of 6813 votes in favor (49.3%) to 7014 votes against (50.7%). Had the levy gotten 202 more votes (a tie results in a loss), it would have passed. In the Gambier precincts, 241 votes or ~18.1% of votes cast were under votes. In the non-Gambier precincts, 390 votes or ~3.2% of all votes cast were under votes.

So, getting back to the central question, did the high rate of under voting in the Gambier precincts cost Mount Vernon Schools the November Levy Election? The answer to that question, of course, is complicated. Below, I will examine four alternative scenarios, each of which results in a slightly different answer.

Scenario One- Everyone votes, under voters all vote for the levy: This scenario, while perhaps unrealistic, is the most optimistic for the levy. Had the under voters in Gambier all voted for the levy, the levy would have passed by a margin of 7054 votes to 7014 votes (pending automatic recount). This scenario, however, is probably overly optimistic; unless the school levy could have generated the sort of enthusiasm as Barack Obama, it is at least somewhat unreasonable to expect that there would be no under votes at all in this race. It is also somewhat optimistic for the levy to assume that all under voters would vote for the levy if they had cast ballots.

Scenario Two- Everyone votes, under voters support levy at rate of voters: What if one assumes that everyone votes, but that the under voters support the levy at the same rate as those who already voted? This may be a more reasonable assumption than assuming that every under voter would naturally support the levy. In the Gambier precincts, 91.2% of voters supported the school levy. Had 91.2% of the under voters supported the school levy, the levy would have gotten approximately 220 more yes votes for a total of 7033 yes votes. However, under this assumption, approximately 21 of the under voters (~8.8%) would have voted no, giving the no side a total of 7035 no votes. Under this scenario, the levy would have failed by three (!) votes (a tie results in a loss). Obviously, the levy would have gone to recount under this scenario; the only thing that would be sure under this scenario is a lengthy legal battle.

Scenario Three- Under voting falls to norm outside Gambier, under voters support levy at rate of voters: The assumption that everyone votes is also somewhat optimistic; after all outside of the Gambier (and College Township) precincts there was some under voting in this race. If we reduce under voting in this race to the non-Gambier average of 3.2%, this means that ~43 under votes would still have been cast in this race, thus meaning that 198 fewer under votes would have been cast. By allocating these under votes in the same way as the formula in Scenario Two, 6994 total votes (increase of 181) would have been cast for the levy and 7031 votes would have been cast against the levy. As a result, the levy would have needed 38 more yes votes to pass under this scenario; however, as with the previous scenario, this result falls within the 0.5% margin to trigger an automatic recount in a local, county, or municipal election.

Scenario Four- Relaxing the Assumptions of Scenarios Two and Three: While the assumptions in Scenario One were likely too loose, the assumptions in Scenarios Two and Three may be too rigid. (Goldilocks had a similar problem with temperature and pudding!) In Scenario Two, I used the 91.2% support rate among all voters. However, it is likely that most of the under voters were Kenyon students as opposed to year-round Gambier townspeople (who make up a small portion of the Gambier vote). I also suspect that Kenyon-affiliated people may have supported the levy at a slightly higher rate than the year-round Gambier townspeople (although support must have been widespread in the village among all residents for the levy to get 91.2% of the vote). Therefore, I average Scenario 1 and Scenario 2 and say that 95.6% of under voters would support the levy.

Let me also relax the assumption of under voting- what if under voting in Gambier took place at a rate of 1.6% in the school levy election, half the 3.2% average for non-Gambier precincts? After all, the Gambier precincts showed in the presidential race that their voters are quite adept at filling out ballots when they want to make their voices heard. Is this assumption reasonable? Perhaps.

Under the relaxed assumption about under voting, ~220 under voters would be converted into voters. Using the assumption of 95.6% support for the levy, I find that supporters would gain ~210 votes and opponents would gain ~10 votes. As a result, the levy would have received 7023 votes in favor and 7024 against, failing by only two (!) votes (again, tie=loss). Once again, the election would have been decided by a recount.

So did under voting cost Mount Vernon Schools the November 2012 election? The answer to that question is a definitive “maybe.” That all depends on a.) which of the above scenarios one finds most convincing and b.) what one assumes would have happened in a recount.

The only other conclusion that can draw is that, had a lower rate of under voting taken place, the election administrator’s prayer most certainly would not have been answered. Most likely a lengthy recount process would have taken place that may have dragged on for weeks if not months.

Sanford v Colbert Busch: what are the odds for South Carolina’s first debate? | Harry J Enten

Debating a cardboard cutout has not helped the former governor against his opponent, and the polls show it

Republican Mark Sanford needs a breakthrough moment in tonight’s debate for South Carolina’s congressional special election, or else his comeback effort to win on 7 May is likely to fall short against Democrat Elizabeth Colbert Busch.

Sanford hopes to return to public office as a congressman, after resigning his in the wake of an extramarital affair for which he resigned from the governor’s mansion and left his wife. After winning the Republican nomination for the special election in South Carolina’s first district, Sanford has faced a barrage of bad press suggesting that he can do no good, even including the news that he violated the terms of his divorce with his former wife.

His Democratic opponent, businesswoman Elizabeth Colbert Busch, has her own claim to fame. Her brother, Stephen Colbert, has hosted numerous fundraisers on behalf of his sibling, helping her stay on the air in a district that went for Mitt Romney by over 15pt. She has also received support from a number of outside organizations, like the Democratic Congressional Campaign Committee.

The National Republican Congressional Committee, meanwhile, has abandoned Sanford. They refuse to air advertisements on his behalf, leaving the airways to his opponents. Sanford himself has little money left, and has been forced to rely on tactics such as a “debate” last week – with a cardboard cutout of Nancy Pelosi. He couldn’t afford podiums. Sanford used music stands instead. As a smart analyst told me, “It just feels like he’s going down.”

The one poll we have from the race shows Sanford trailing by 9pt, 50% to 41%. The poll, conducted last week by Public Policy Polling (PPP), shows an electorate that has Democrats as a greater proportion of the vote compared to PPP’s prior survey, which Colbert Busch led by 2pt. The new poll’s respondents voted for Romney over Obama by only a 5pt margin, compared to the 18pt margin Romney by which won the district last fall. PPP attributes this difference to low Republican enthusiasm, itself stemming from the most recent charges against Sanford.

Not everyone is buying PPP’s survey, though. Cook Political Report’s Dave Wasserman, for instance, thinks that Colbert Busch leads, but by a margin less than 9pt. My own math indicates that when you adjust PPP’s projected electorate to match its prior survey, Colbert Busch is ahead by a little over 3pt.

So just what are the chances that PPP’s poll is off by 9pt? Not very likely.

Over the past ten years, there have been 30 non-candidate/party sponsored polls conducted in the final two weeks before 18 off-cycle (i.e. not on election day during a midterm or presidential year) special congressional elections. Overall, these polls have been mostly accurate; 27 out of 30 times, the candidate leading in the poll won most of the votes in the special election. (Note: I only look at those elections where at least one Democrat and Republican are running. You can see the raw data here.)

Two of the three polls after in which the leading candidate didn’t win the election were for the 2009 New York 23rd special election, a unique race, in that there were three competitive candidates: Democrat Bill Owens, Republican Dede Scozzafava and Conservative Party candidate Doug Hoffman. In the week prior to the election, the Republican candidate dropped out, remained on the ballot, and endorsed the Democrat.

Not surprisingly, the 23rd special had PPP’s largest absolute error; they predicted Conservative Hoffman to win by 17pt. He lost by 2pt to the Democrat.

In 26 of the 30 special election surveys, the polling error was less than 9pt – Colbert Busch’s lead. The polls’ average error was only 4pt, and the median even lower, at 3pt.

These margins leave little room for Sanford. He would need an error of at least 9pt, which has happened only 13% of the time. He would also need that error to be in his favor, which means he’s looking at a 7% likelihood. That’s not outside the realm of possibility, but it’s a long shot. My guess is internal polling shows the same results, which is why the Cook Political Report and Rothenberg Political Report have put their finger on the scale for Colbert Busch.

The best comparison to this scenario would be Louisiana’s sixth special election in 2008. Louisiana’s sixth was like South Carolina’s first in that a Republican represented the district for over 30 years, and both districts were over 15pt more Republican on the presidential level than the nation as a whole in the prior presidential election. Like Sanford, Republican Woody Jenkins had run for statewide multiple times. Jenkins also had his own ethics problems: a connection with former KKK wizard and Louisiana politician David Duke.

The final poll before the Louisiana sixth special had Democrat Don Cazayoux ahead by 9pt, just like the recent poll that shows Democrat Colbert Busch leading. Jenkins did better than his numbers suggested. He didn’t lose by 9pt, but he did lose – by 3pt. Even though the poll overestimated Jenkins’ deficit, Cazayoux’s lead was too great to overcome.

Like in Louisiana’s sixth, there is a decent chance that the upstart Democrat’s lead is less than 9pt. It’s quite possible Republican voters will come around to Sanford on election day.

The chance, however, that the PPP poll is so off that Mark Sanford defeats Elizabeth Colbert Busch? Less than 15%, unless something happens very quickly.

Harry J Enten

Patterns of Under Voting in Gambier and the rest of Knox County, Ohio (Part One)

Among residents of Knox County, Ohio, the political differences between Gambier (home of Kenyon College) and the rest of the county are well-known. Gambier is populated by generally liberal students and faculty who (mostly) vote Democratic; Michelle Obama even visited the Kenyon campus in 2012. In contrast, the rest of the county is largely filled with generally conservative voters who tend to vote Republican. Indeed, 2012 Republican candidate Mitt Romney held a campaign event at the Ariel Corporation in Mount Vernon. Overall, Knox County voted for Governor Romney over President Obama by a 61 to 37 percent margin. Outside of Gambier and surrounding College Township, President Obama won the most votes in only one precinct (there was a tie in another precinct).

Using precinct-level data from the Knox County Board of Elections, this post focuses on another noticeable difference in voting patterns that exists between Gambier and the rest of Knox County: the extent to which “under voting” takes place in various contests. According to Wikipedia, an “under vote” occurs when, “the number of choices selected by a voter in a contest is less than the maximum number allowed for that contest or when no selection is made for a single choice contest.”

A close look at the Knox County Board of Elections website reveals an interesting pattern when one examines under voting by precinct. In the 2012 presidential race, not a single “presidential under vote” was cast in either Gambier precinct (the surrounding College Township precinct also saw no under votes). What makes this so interesting? In the rest of the county every other precinct had at least one under vote in the race for president. Indeed, 213 votes (~0.8% of all votes cast) in the rest of the county were under votes.

What makes this pattern even more remarkable is that it begins to reverse itself in other races down ballot. Outside of the race for president, the under vote rate in Gambier exceeded the norm for the rest of the county.

For example:

In the Senate Race between Senator Sherrod Brown (D) and State Treasurer Josh Mandel (R), there were 87 under votes in Gambier or ~6.5% of all votes cast. Outside of the Gambier precincts, there were 619 under votes or ~2.3% of all votes cast.
In the House Race between Representative Bob Gibbs (R) and Challenger Joyce Healy-Abrams, there were 140 under votes in Gambier or ~10.5% of all votes cast. Outside of the Gambier precincts, there were 1360 under votes or ~5% of all votes cast. This despite the fact that the only debate between Gibbs and Healy-Abrams was actually held at Kenyon College in Gambier!
In the “Nonpartisan” State Supreme Court Race between Incumbent Robert Cupp (“R”) and Challenger Bill O’Neill (“D”), there were 730 under votes or ~54.8% (!) of all votes cast. Outside of the Gambier precincts, there were 6453 under votes or ~23.6% of all votes cast. (Note: I called this race “nonpartisan” due to the fact that, although no partisan labels appear on ballots, candidates are nominated through partisan primaries.)
The pattern is similar in other races down ballot.

So what implications can be drawn from this?

Here are three initial takeaways:

The Power of the Obama Campaign: Young voters really connected with President Obama and his campaign did a great job of reaching out to these voters and getting them to turn out to the polls. These voters were excited to vote for President Obama and filled out their ballots in such a way as to act on this excitement. This excitement about voting for President Obama, however, did not represent increased loyalty to the Democratic Party as a whole; this was made clear in the 2010 midterms as turnout among young voters remained relatively constant with historical patterns and did not experience any noticeable surge.
Importance of Partisan Cues: The substantial drop off that took place in the Gambier precincts for the State Supreme Court race underscores the odd things that can happen in ostensibly non-partisan judicial races. While some Kenyon students were willing to vote for a candidate with a “D” next to their name, they weren’t about to go searching for the partisan affiliation of a non-partisan candidate. (Good work on non-partisan judicial elections is being done by University of Pittsburgh Professor Chris Bonneau and UNC Graduate Student John Lappie.)
Under voting isn’t a liberal thing, it’s a college student thing: While under voting rates were above average in the Gambier precincts, this was not the case in the College Township Precinct. Home to some Kenyon employees, College Township has an ever-so-slight Democratic tilt. Furthermore, under voting in College Township was in line with the rates for the rest of the county. For example, 5 voters or ~2.2% under voted in the U.S. Senate race between Senator Brown and State Treasurer Mandel in College Township.

These implications are certainly not the only ones that can (or should) be drawn from this data. Indeed, the next post in this series will examine the practical implications of under voting for low turnout races, focusing specifically on the Mount Vernon School Levy.

Immigration reform won’t deliver a Latino voter ‘bonanza’ to Democrats | Harry J Enten

To assume that 11 million undocumented immigrants potentially eligible to vote will change US political arithmetic is erroneous

I have been quite skeptical about the ability of the Republican party to win over Latino voters. As others and I have noted, Latinos don’t vote Democratic just because of immigration policy. They vote Democratic because they are more ideologically “in sync” with the Democratic party. That’s why I’ve felt that going along with immigration reform was unlikely to net Republicans many Latino voters.

That said, I can’t agree either with the math in a Politico article titled “Immigration reform could be a bonanza for Democrats”. The article starts off promisingly enough with the premise that if immigration reform passed and undocumented immigrants became citizens, Latinos would start voting Democratic in even larger numbers. I can go along with this because the main reason anyone votes for or against a political party is for its economic platform, and 81% of first generation Latino immigrants say they want a “bigger government with more services”, compared to only 48% of Americans overall.

Politico then uses the commonly quoted figure of 11 million undocumented immigrants and claims that there would, therefore, be “up to” 11 million undocumented immigrants up for grabs if they all became citizens. If these 11 million then voted along the lines of the Latinos who cast a ballot in 2012, Obama would have won the national vote by 7pt instead of 4pt. He could have carried Arizona and even Texas, which were each won comfortably by Republican Mitt Romney.

The problem I have, though, is why would anyone use the 11 million figure for reference.

First off, 1 million undocumented immigrants are under the age of 18. I don’t care what your immigration status is, you can’t vote in United States presidential elections if you are under the age of 18.

Second, of the 10 million adults, 19% aren’t actually Latino; 11% are Asian. Asian voting patterns tend to be less stable than Latinos. In the past election, Asians went for President Obama by about 45pt. Twenty years ago, they went for Republican George HW Bush.

Without more research (and there is surprisingly little of it), it’s unclear to know how undocumented Asian voting patterns would change given immigration reform. Also, keep in mind that about 60% of these Asian immigrants are in California and Washington State alone – so they’re not exactly going to be a game-changer in the electoral college.

Third, I tend to doubt that all 8 million adult Latino undocumented immigrants would go for citizenship. A Latino Decisions poll says 87% of them would, but I’m skeptical. Only 60% of all legal immigrants actually apply for citizenship. Since the Immigration Reform and Control Act of 1986, only about 40% of the newly eligible immigrants by so-said act became citizens. It would be reasonable to expect a similar percentage this time.

Only 36% of Mexican immigrants who are eligible to become citizens have gone through the process. Mexicans make up 72% of all undocumented Latino immigrants. Six in ten of the other 28% of eligible Latino immigrants have become citizens.

All together, if undocumented Latino immigrants become citizens at the rate that eligible Latino immigrants do, then we’re looking at 3.5 million new voting-age citizens. That 3.5 million is a far cry from the 11 million we first talking about.

Finally, just how many of these 3.5 million undocumented Latino immigrants can be expected to vote in the presidential election? Per the standard census Current Population Survey (CPS), only 49.9% of all voting age Latino American citizens cast a ballot in 2008. (Note, there is no report available for 2012 as of this point.) Based on pre-election surveys and work by Michael McDonald, there is reason to believe that percentage may have dropped further in 2012.

All told, it would seem that only about 1.7 million new Latino voters would be added if undocumented immigrants were granted citizenship. Nationally, this would be a net of about 775,000 votes. This would increase Obama’s vote margin, but not to 7pt; it would only go up to about 4.4pt – in other words, half a point from where it actually was in November 2012. Even adding in new Asian voters, who vote at a lower rate than even Latinos, and other undocumented immigrants (and controlling for the percentage who apply for citizenship, percentage of citizens who vote, and the percentage who voted for Obama), the margin probably only goes up to, at most, 4.6pt.

The amount this would shift individual states in elections is debatable. Take Nevada, where, at last count, there were 190,000 undocumented immigrants – the highest percentage of any state population. Most of them are Latino. Apply the same math we did above, Obama would have gained about 17,000 votes. It would have increased his state margin of victory by 1.4pt. That’s not nothing, but we’re talking about the state with the largest percentage undocumented immigrants.

Most states aren’t close to Nevada’s undocumented immigrant population, while the ones that are simply aren’t competitive at the president level: Arizona, California, and New Jersey. No state in 2012 would have had a different outcome if undocumented immigrants were given the right to vote.

The truth of the matter is that passing immigration reform won’t be a votes “bonanza” for the Democratic party because of potentially or newly enfranchised undocumented immigrants. That doesn’t mean passing immigration reform will help the Republican party among Latinos; the GOP should probably still be worrying about its Latino voter appeal. But it’s not facing a landslide from a new citizen electorate.

Harry J Enten

LA mayoral race: Garcetti leads Greuel as runoff election enters final month | Harry J Enten

Two Democrats are vying to be mayor of the second most populous city in the US – and the pollsters like Eric Garcetti

You wouldn’t know it on the US east coast, where the focus is on who will win the contest to succeed Michael Bloomberg as mayor of New York, but the election to be mayor of the nation’s second most populous city is also occurring this year. The fight to lead the City of Los Angeles is well underway. A runoff between Democrat Eric Garcetti and Democrat Wendy Greuel is slated for 21 May.

Los Angeles, as with many cities, uses a non-partisan runoff system. That is, all the candidates, regardless of party, face off against each other in the first round. If no candidate receives 50% of the vote plus one vote, a second round is held between the two top finishers. In this case, the two top finishers were both Democrats.

Garcetti is currently a city councilman and the former president of the same body. Greuel worked as a member of the city council, with Garcetti, and now as the city comptroller. Both are white, liberal Democrats. Indeed, there isn’t much in the way of policy differences between the two.

The main separation between Garcetti and Greuel is a difference of position on public unions. Greuel is supported by them; Garcetti is not. Garcetti’s main line of attack on Greuel is that she is bought and paid for by the Department of Water and Power union, which is costing the city millions in pensions. Whether or not it’s a hypocritical line of attack given Garcetti’s voting record on DWP, it’s working as a campaign strategy.

Garcetti is leading Greuel 50% to 40% in the latest USC/Los Angeles Times poll. He does better among those who give unfavorable ratings for Los Angeles’ unions. Thanks to an endorsement by third-place finisher, Republican Kevin James, Garcetti takes the Republican vote by 18pt and independent vote by 13pt. A key constituency for Greuel is black voters, with whom she’s up by 20pt.

Greuel, however, wasn’t buying the poll’s top line. In fact, according to the New York Times’ Los Angeles correspondent, Adam Nagourney, she “trashed” it. Polling for low turnout elections is difficult, especially in a city as diverse as Los Angeles. Beyond this generality, though, is there any reason to think the poll is inaccurate?

The USC/Los Angeles Times poll seems to textbook example of solid methodology. The poll was conducted by a Democratic pollster, Benenson Strategies Group, and by Republican pollster M4. Benenson was the lead pollster behind President Obama’s highly successful polling team, while M4 was the only pollster to nail President Obama’s 2012 23pt victory in California.

Cellphones were called; multiple attempts were made to contact each respondent if they didn’t respond initially; and the poll was weighted using voter registration files to ensure that a realistic portrait of the likely electorate was painted. This last point is important because the election is probably going to have low turnout, and you don’t want to cast too wide a net.

Some might complain that the voter file might miss transient residents likely to live in a growing city like Los Angeles. The issue for Greuel is that another public pollster SurveyUSA uses the random digit dial technique. That is, they contact all types of Los Angeles adults, regardless of voting history. Based on a number of questions, they whittled their results down to likely voters. Their last poll conducted two weeks ago had Garcetti ahead by a very similar 49% to 40% margin.

Moreover, the voters who pollsters are most likely to miss are actually Garcetti voters. Garcetti led among younger voters by at least 20pt in both the SurveyUSA and USC/Los Angeles Times polls. He was up by 20pt or more among Latinos – the ethnic/racial group most likely to be undercounted by traditional polls. Garcetti was ahead with cellphone respondents by 23pt in the SurveyUSA poll.

Past polling accuracy lends credence to the belief that these surveys are correct. In 2001, an average of the two final polls for the first round had Antonio Villaraigosa ahead of Jim Hahn by 2.5pt. Villaraigosa won the first round by 5pt. An average of the final two polls in the runoff had Hahn over by Villaraigosa by 8.5pt. Hahn emerged victorious by 7pt. The average error for the two rounds was only 2pt.

In 2005, an average of the two final polls for the first round had Villaraigosa leading Hahn by 14pt; he won by 10. An average of the two final polls in the runoff had Villaraigosa winning by 17.5pt; he became mayor by a 17pt margin. The average error of the two rounds was only 2.25pt.

I could find no public surveys from 2009, though we do have one round from this year. An average of the final two polls had Garcetti and Greuel tied. Garcetti emerged victorious by 4pt.

All in all, the polls showing Eric Garcetti leading race for Los Angeles Mayor are almost certainly correct. Wendy Greuel would be better-off concentrating on trying to close the gap with Garcetti, rather than complaining about public surveys. Recent election polling for Los Angeles mayoral elections have been marked by accuracy. None have featured significant errors. The possible fault lines in the surveys in 2013 would, if anything, make Garcetti appear weaker than he actually is.

Harry J Enten

LA mayoral race: Garcetti leads Greuel as runoff election enters final month | Harry J Enten

Two Democrats are vying to be mayor of the second most populous city in the US – and the pollsters like Eric Garcetti

I could find no public surveys from 2009, though we do have one round from this year. An average of the final two polls had Garcetti and Greuel tied. Garcetti emerged victorious by 4pt.

Harry J Enten

For It While They Were Against It

Sometimes, people respond in strange ways to survey questions.

For a recent project with Jim Stimson and Elizabeth Coggins, I spent a fair amount of time analyzing data from the Cooperative Congressional Election Study (CCES). Here’s a fun nugget from my exploration: a sizable proportion (21 percent) of respondents both support and oppose Obamacare. Simultaneously.

We can speculate wildly about why a fifth of respondents — in a sample that is disproportionately educated and interested in politics! — would give such a puzzling answer.

But in a bigger sense, surveys — as useful as they are — offer highly artificial settings where respondents will give answers. Not attitudes, nor opinions, nor preferences per se — just answers. We should keep that in mind before reading too much into public opinion reports.

The Conflict

Part of the CCES comprises a set of “roll call” votes. These present respondents with a policy position and require a simple yea/nay answer. Two of these questions ask about the Affordable Care Act: one asks the respondent to vote for or against Obamacare, the second asks respondents to vote for or against repealing Obamacare.

There is a logical connection between these two questions. In general, someone who wants to repeal the law would probably not vote for it; and those who want to keep the law around should vote for it to begin with.

Generally that works… but as the ‘jitter’ plot shows above, it doesn’t work that way for everyone. Each dot on the figure represents a single respondent. (I like imagining that I’m assigning people to stand in a corner of the room depending on their answers to questions. Maybe I have a power complex…) There are clearly a good number of respondents in two quadrants: those who either support Obamacare and want to keep it, and those who oppose Obamacare and want to repeal it. Makes sense.

But who are those respondents in the other two quadrants? Slightly more than 12 percent of them want to repeal Obamacare, despite saying that they would vote for the bill; and 9 percent would vote against the bill, but wouldn’t repeal it.

The latter group — the Vote Against / Don’t Repeal group — may be reasoning through the path dependency of Obamacare. Something like, “Well, I don’t like it, but it would endanger the health care system to repeal it now.” Or maybe they’re just ardent believers in the Democratic process: elected officials passed the bill, any who would I be to usurp them? I doubt either of these stories, but it’s not impossible.

The other group — the Vote For / Repeal It! group — is weirder, though. There’s really no logical connection between the two answers.

Surveys are weird…

Well, they are! Despite having used public opinion data in research for several years now, I took my first “real” political survey over the winter holidays. Gallup called and wanted to talk to me about global warming, and that sounded like fun.

It wasn’t. First, you get pretty tired of answering questions after the first twenty. Second, even as a well-educated, highly-informed and engaged observer of the political world, the survey made me feel dumb. There’s this unusual pressure in a survey to answer questions promptly, which is fine but sometimes you don’t have an easy answer right at the top of your mind. Besides, these issues are complicated! Global warming? Economics? Coal, nuclear, wind, oil? Health care mandates?

Stressed yet? Even informed and engaged respondents get a bit overwhelmed by the survey items, and by the need to provide clean answers to complicated questions. And sometimes the questions aren’t entirely clear. Are we asking if you would have voted for Obamacare back in 2010? Or would you vote for it today? Do some respondents miss the “repeal” part of the question? These are all possible points of confusion, introduced in a highly artificial environment, but for which it’s impossible to test without a specific instrument.

Here’s the uncomfortable truth about polls: we use them because they’re what we have. On many questions, they’re good for giving the general feeling in the public. “Will you vote for Mitt Romney, the Republican, or Barack Obama, the Democrat?” isn’t terribly difficult, and most respondents can give a decent answer.

But as the questions become more complicated, responses become less reliable. Accessing “true” attitudes on policy questions with a survey can sometimes be like removing a splinter from your finger with an axe: In a sense it works, but it’s awfully messy.

And it gets messier when we try drawing relationships between multiple items, all of which have some weird characteristics, like non attitudes, weak attitudes, and non response. Aggregating to reduce the high dimensionality of multiple responses can help filter out some of the noise, but that’s a topic for another post.

Pundits and commentators roll out polls daily to elicit support for some position or another. Being an informed consumer of surveys means going beyond “What’s the Margin of Error?” (We are the Margin of Error, duh!)

It means realizing that a fair number of responses might carry little objective meaning. When pressed I’ll answer, but I honestly don’t know, don’t care or haven’t quite figured out my views yet. Treating these responses as some true-to-life measure of how the American people feel, or how they’ll act, can go pretty far afield.

–

Note: The CCES sample above is limited to the UNC module of 1,000 respondents. Expanding this to the full CCES sample of 55k+ doesn’t change anything, though, but does make the figure a bit messier.

The Manchin-Toomey gun law defeat heralds trouble for Senate Democrats | Harry J Enten

Demographic shifts in the US, with Democratic support concentrated in big cities, are driving ever more partisan politics

On Wednesday, the United States Senate defeated the Manchin-Toomey proposal on background checks. The defeat was a setback for gun control advocates, though it should not have come as a surprise. The defeat with a manifestation of growing problem for Democrats: their coalition is bad for winning many seats in the Senate, so it’s bad for passing legislation, too.

Each state, regardless of size, gets two senators. The least populous state, Wyoming, at 576,000 residents, has two senators, as does the most populous state, California, at 38 million. These less populated states tend be mostly rural. The more populated states like California gain their population because of large cities.

The issue is that Democratic voters tend to congregate in larger cities. Per the 2012 exit polls, President Obama won cities with populations over 500,000 (or cities with populations about the size of all of Wyoming) by 40pt. They won cities with a population of between 50,000-500,000 by a still impressive 18pt. They lost the suburbs by 2pt, cities with populations between 10,000 and 50,000 by 14pt, and rural areas by an astounding 24pt.

Indeed, look at a key part of the Democratic coalition: Latinos. Forty-five percent of Latinos live in the ten largest metropolitan areas. Obama won 12 out of the 16 top states with the largest Latino population. Latinos have helped to solidify the Democratic hold on what was once the swing state of California with 38% of the state’s population – far greater than the national average of 16.3%. Forty-one states, however, have a Latino population at a lower proportion than the nation’s 16.3%; 33 have a population that is less than 10% Latino.

So, how did this manifest itself on the state level?

President Obama won a 7.3pt victory in 2008, yet only carried 28 states. Four years prior, President Bush won only a 2.5pt victory and carried 31 states. Eight years prior to that, in 1996, President Clinton won by 8.5pt and took 31 states. George HW Bush emerged victorious by 7.7pt and won 40 states. That’s right: GHW Bush’s national margin was about the same as Obama’s was 20 years later, but he took 12 more states.

Democrats have tended to win fewer states than Republicans given the same national vote victory, but it’s become worse. To win 30 states in the 2012 election, Obama would have needed to carry the national vote by about 9pt more than he did. He lost Arizona by 9pt, even as he won nationally by 3.9pt. It would have taken a true landslide for Obama to have won by 13pt nationally.

Twenty years ago, this concentration of Democratic strength might not have been too big of a deal, in terms of Senate representation. As I noted back in December, 49% of the Democrat’s 1993 senate caucus came from states that were more Republican the nation as a whole in the prior presidential election. Today, that percentage has been cut in half to only 25%.

When you combine the fact that Democratic presidential nominees are winning fewer states with the fact that there is more straight ticket-voting, the Democrats have a major problem on their hands. It’s simply going to get harder here on in to win a Senate majority, let alone a super-majority of 60 seats, which a party really needs to overcome the growing use of the filibuster.

The Democrats who are in “red” states recognize this fact and you saw it this week in the gun ownership background checks amendment vote. The four Democrats who bucked their party line were from states more Republican than the nation as a whole in the last presidential election. Three of the four are up for re-election in 2014. Of the red state Democrats running for re-election in 2014, three of five voted against the Manchin-Toomey compromise.

Just as bad for Democratic legislation is that there are no Republicans out there willing to compromise. Only 16% of the Republican Senate caucus comes from states where Obama won by a greater percentage than he won nationally. The percentage of blue state Republicans is also down by about half from the 28% it stood at after the 1992 elections. The Republicans who are elected now come from red states and just have no electoral need to compromise. If anything, they’re more fearful of a primary challenge from a stricter conservative.

Only one Republican from a red state, John McCain, voted for Manchin-Toomey. The other three were all from states that were more Democratic than the nation as a whole. The bill’s co-sponsor Pat Toomey, who is no liberal, likely benefitted from the politics of being on board with a bill that is popular in his home state of Pennsylvania.

None of this means that Democrats won’t get back to 60 seats, or can’t bring Republicans on board, in the future. It just means that it’s just much harder than it used to be and will get harder still. The current Democratic coalition is concentrated in a few states, which has resulted in fewer states being Democratic on the national level. Combine this with less split-ticket voting between presidential and congressional races, and it’s bad news for Democrats in getting proposals through the Senate.

The immediate result was the failure of Manchin-Toomey – an event that I expect to be repeated with much future Democratic legislation.

Harry J Enten

Can Weiner win the Democratic primary for New York mayor? | Harry J Enten

As the former congressman polls second behind Christine Quinn, Anthony Weiner looks set to come back from his sexting scandal

Anthony Weiner looks primed to shake up a dormant New York City mayoral race with his entry – should he get in the race officially. After appearing in print and on television, as well as putting out a policy notebook in the last week, the former New York congressman can now add “second-place poll finisher” to his resume.

Marist has published a survey of the Democratic primary race showing frontrunner Christine Quinn at 26%, Weiner at 15%, John Liu at 12%, Bill de Blasio at 11%, and Bill Thompson at 11%.

I take the poll as mostly good news for Weiner in his quest to take Gracie Mansion. Why?

1. Weiner has gone straight in at second spot

The fact that Weiner can shoot up to second place without even officially entering the race is a strong sign. What makes it even stronger is the fact that second place in a New York City mayoral primary is a ticket to the next round. When no candidate reaches 40%, the top two finishers go forward to a runoff. Quinn’s not anywhere close to 40%, while Weiner’s sitting pretty in second.

New York City runoffs can be odd ducks. You can end with a situation like the 1993 Comptroller primary where the incumbent Liz Holtzman finished second in the first round, then lost the second round with a lower percentage of the vote than she’d won in the first. You could have a 2001 mayoral primary scenario where Mark Green lost the first round by 5pt and won the second round by 2pt.

The point is that anything can happen in a runoff. The key is to live in the first round to fight another day.

2. Weiner has the potential to build a stronger position

The one thing I was struck by upon reading the Marist poll is that Weiner’s in a far stronger position than he was just two months ago. In Marist’s February poll, Weiner’s favorable rating among Democrats stood at 34%. That’s now up to 45%. His unfavorable rating is down 2pt, from 43% to 41%. Overall, that’s a 13pt gain in net favorability.

The percentage of Democrats who want him to run for mayor is also way up. The percentage of Democrats who wanted Weiner to run just after the sex scandal was only 27%. That’s up to 40% now. The percentage who don’t want him to run is down from 65% to 46%. Together, it’s a net 32pt turnaround for Weiner.

The point is that it would be a mistake to treat Weiner’s favorables now as stable. The more time that has passed since his sex scandal, the more people seem to be warming to him again. His recent press tour mostly garnered positive press; it certainly did not backfire – people were not turned off. With his wife Huma Abedin by his side, it’ll be awfully difficult for opponents to try and hang the Twitter-sexting scandal around his neck.

3. Quinn is an unusually weak frontrunner

Under attack for a record that many believe is not all that liberal, Quinn’s campaign has faltered. Quinn was at 37% two months ago; she’s now at 26% with Weiner in the field. It doesn’t take a genius to recognize that her numbers are going in the wrong direction. That’s good news for candidates not named Quinn.

The 26% is the lowest polling result for a Democratic frontrunner since 1977. Keep in mind that that field was evenly divided not because the candidates were weak, but because they were all so strong. It included women’s right leader Bella Abzug, future Governor Mario Cuomo, and Ed Koch, among others. This year, Quinn is at 26% even as the third-placed John Liu’s former campaign treasurer is on trial for campaign funding fraud.

Even the “meh” candidacy of Ruth Messinger in 1997 was in the upper 30s at this point. And Fernando Ferrer, who would go on to lose the general election by 20pt, was also polling in the upper 30s at in 2005.

For Quinn to have been running for many months, to have been on the cover of New York Magazine, and still to only be at 26% will be a dismayingly feeble result for her campaign managers. Her base, too, seems one that may be unstable.

Quinn is polling at 28% among black voters. The problem is that New York City politics does tend to be racially determined. Even in her subpar run of 2005, African-American candidate C Virginia Fields managed to win a plurality of the black vote. Right now, Bill Thompson, who was the Democratic candidate for mayor in 2009, is only at 14% among African Americans. I’d expect him to eat into Quinn’s percentage of black voters, or at least freeze her where she is.

4. Weiner v Quinn level-pegging in name recognition

The one piece of what might be seen as bad news for Weiner in the Marist poll is that with 85% name recognition, he’s still only polling at 15%. But I don’t see that as much of a problem: he wasn’t polling much higher before the sex scandal hit, but his numbers are in flux and have trended positive. Meanwhile, his frontrunner rival, Christine Quinn, has a similar level of name recognition, yet she’s slipped back to 26%.

Conclusion

The facts that he’s already in position to make the runoff as it stands, that his favorables are going in the right direction, and that the frontrunner is stalled, are all good signs. That doesn’t mean he will win. I am still partial to the possibility of Bill Thompson overhauling the others, given the math above.

At this point, all the candidates’ hopes are very much alive. But do I think Anthony Weiner can win this thing? Absolutely, he could.

Harry J Enten

Terrorism in the US: what is the real threat level? | Harry J Enten

In perspective, terror attacks are at a historic low. But the picture is more complex: major cities like Boston are still likely targets

Though we know nothing yet about who committed the bombings at the Boston Marathon, or why they did so, the assumption at this point is that these were acts of terror – and, as the president affirmed Tuesday, is being investigated as such.

The shadow cast by the 9/11 attack means that every such incident now tends to be seen as a new episode in a distinct and frightening era of terrorism in the mainland United States. But does this picture actually fit the historical record?

Let’s put Boston in the context of the history of terrorist attacks in the United States over the past 40 years. One graph that tries to do this, posted by the Washington Post on Tuesday morning, is this:

It’s taken from the National Consortium for the Study of Terrorism and Responses to Terrorism at the University of Maryland (pdf), and it plots the number of terrorist attacks in the 50 states. It’s difficult not to be immediately struck by the peak on the left hand side of the graph: no other year comes close to the near 500 attacks that occurred in 1970. In fact, only two other years even surpass 100 attacks, and both of those were also in the 1970s. But this graph doesn’t tell the whole story.

First, let’s start with the fact that most terrorist attacks occur in major metropolitan areas. Manhattan and Los Angeles are two of the most at-risk targets for major terrorist attacks throughout the period: 13.1% of all terrorist attacks from 1970 to 2008 occurred in Manhattan, New York alone; 6.0% were in Los Angeles County, California. Combined, that’s nearly a fifth of all terrorist attacks.

Other areas were less consistent at registering in the terrorism chart. San Francisco was a center of leftist terrorism in the 1970s, but has since cooled off. Maricopa County, Arizona has seen more attacks in the 2000s as “single-issue terrorism” (such as animal rights or hate crime) has become more dominant.

Second, over the last decade, very few terrorist attacks have occurred across the central swath of the country. When you look at a heat map of terrorist attacks over the last 40 years, you see terrorism occurring in many different states. That, however, shows where attacks have been most likely to occur by decade.

Only five of the terrorist attacks coded by type by University of Maryland occurred in the center of the country during the 2000s. Almost all the attacks that took place in that region were during the 1980s and 1990s, when rightwing and religious groups were the most likely to carry out attacks.

Third, terrorism attacks are most likely to be in places where crime is high. There’s not a perfect correspondence, to be sure; but the correlation between high crime rates and terrorist attacks is highly significant at 0.25.

Fourth, terrorist attacks tend to occur in areas that are most ethnically diverse. Even when taking into account population density, you are more likely to see a terrorist attack in an area where many languages are spoken. You are also more likely to see a terrorist attack in a city where people live with less residential stability. Poverty and inequality, however, are not a factor: you are more likely to see a terrorist attack in cities with a lower degree of concentrated disadvantage.

A key point, however, is that the percentage of foreign-born residents is not a significant predictor of terrorist attacks. The same goes for racial identity: a city with more black or Hispanic residents is no more likely to see a terrorist attack than a majority white city. Language likely differs from these factors because you usually see different languages in cities with foreign business interests or government interests, which is exactly where terrorist attacks are more likely to occur.

Fifth, the chance of surviving a terrorist attack is the same as it has always been. Deadly terrorist attacks may be down, but the the percentage of those attacks that are deadly is the same. In 1970, only about 5% of terrorist attacks were deadly. That spiked to 41% in 1973, but it has mostly hovered between zero and 20% in the years since. The two years with the biggest spikes were 2001 and 2006, when 25% of attacks were deadly.

Sadly, most of these factors tend to make the events that occurred in Boston predictable or likely, compared to other places. Boston is among the top ten major metropolitan areas, and it’s on the coast. It’s in the top third for United States cities by violent crime rate. A number of the places within the metropolitan area have high population densities, including the city of Boston itself. The city ranks highly for non-English speaking households (pdf). Finally, the attack had a fatal outcome.

About the only characteristic that perhaps doesn’t fit the conventional pattern is that it occurred at a sports event, rather than the target being a government building or business. Indeed, the targeting of ordinary civilian members of the public is the truly scary part of the Boston Marathon bombing: with the exception of the Atlanta Olympics bombing of 1996, the idea of attacking a sports event in the United States is novel and unusual.

In other ways, Boston is typical: from a statistical point of view, it’s not all that surprising that nearly 2% of all US terror attacks occurred in the Boston metropolitan area between 1970 and 2008. This week’s tragic incident sadly matches the pattern.

Harry J Enten