Does Voting Democratic Make You More "Manly"???

Apparently the feeling of winning or losing an election actually affects one's physiology. A newly released study suggests that on election night, the testosterone levels of male voters changed when the results were broadcast. Male McCain voters saw a precipitous drop in their testosterone upon learning that their candidate had lost, whereas male Obama voters saw no change in their testosterone. In the parlance of the anthropologists and pharmacologists behind the study (which had participants submit saliva samples throughout the evening) elections are "dominance contests" which, like other competitions in society, "are a critical component of determining the leadership of social hierarchies." The feeling of victory of defeat, despite the fact that one did not win or lose personally, but vicariously, apparently triggers something quite deep in our psyches. Considering a variety of other variables, the authors find that:

the pattern of testosterone change remained significant even when variance in a multitude of factors was controlled for including voters' political values, support intensity for the candidates, timing of saliva collection, levels of conservatism, consumption of alcohol on the night of the election, and social surroundings on the night of the election.

Also of note is the fact that the testosterone levels of females was unchanged, regardless of whether one supported Obama or McCain. Apparently women are much more even keeled than men in this regard.

One can imagine all the arm chair extrapolation one could make from this study. Is Republican opposition to Obama not just based on policy differences but rooted much more deeply in a fear of "dominance"??? Can this be applied more broadly to some geographical variations in the vote--i.e. a southern inferiority complex??? One could run wild with this type of stuff, so feel free to hypothesize in the comments section. Anyhow, this is certainly a different way of looking at voting than what we normally consider here.

Why is John Cornyn Being So Nice to Sonia Sotomayor???

To continue our look into the political, geographic, and demographic ramifications of the nomination of Judge Sonia Sotomayor to the Supreme Court, I thought I'd speculate a bit on how various Senate Republicans might approach the confirmation process. Over the past week the lines seem to have been drawn around the question of whether biography should affect one's view of the law. As we approach the confirmation hearings and vote, we'll move (thankfully) from the stage that we're currently in--all sorts of "legal analysts," interest group representatives, and other flaks pontificating their positions and advancing their agendas--to one in which those actually voting on her nomination (i.e. Senators) are at the forefront. Unlike the current crop of voices, Senators are actually accountable to voters. Thus, they tend to approach these decisions with a bit more tact and caution.

To get a sense of the calculations that some Republican Senators might be going through, I thought I'd look at the size of the Hispanic populations in their relative states. Opposing the first Hispanic nominee to the Supreme Court leaves one open to retribution from Hispanic constituents. If it comprises a sizable component of the electorate, an energized Hispanic bloc bent on retribution could prove decisive in defeating the Senator who votes "No." Given the degree to which Hispanics moved to the Democratic ticket last fall, and given the rate of growth in the Hispanic population, a number of GOP Senators are going to spend the next few months walking a delicate tight rope.

I've created the following chart that looks at each of the 40 current GOP Senate seats. Note that because we're focusing on the electoral dimension of the confirmation vote I haven't included the retiring Senators but have listed their seats as open. I've coded, first, when that seat is up for re-election. We might expect those Senators with re-election races next year to perhaps be most fearful of the charge of being anti-Hispanic. I've next coded the percentage of their state that is Hispanic, based on numbers from the Pew Hispanic Center. Obviously, those states with a larger Hispanic vote present "No" voting Senators with a greater risk. Next I've coded the margin of victory in each Senator's last election. Those GOP senators who had extremely large margins of victories might have little to fear (especially if they come from states w/a small Hispanic population) in voting against Sotomayor compared to those who had close races. Next I identify the seven current Republican members of the Senate Judiciary Committee. These members will be in the spotlight as they lead the questioning of the nominee and cast the initial votes on her. Committee members also tend to "carry the water" for the party at large during the nomination process so we will see the emerging GOP strategy on Sotomayor implemented by this group. Finally, I've indicated how Senators in office at the time voted on Sotomayor's elevation to the U.S. Court of Appeals in 1998. It will be interesting to see whether those who supported her in 1998 now vote against her, or vice versa,--and how they justify such a vote.

A couple of interesting things to note. As I noted in the title of this post, Texas Senator John Cornyn has been conspicous in his even-handedness toward the Sotomayor nomination. One reason for this, perhaps, is the demographic reality that is modern Texas. Percentage wise, the Lone Star State has the largest Hispanic population in the country. While Cornyn was just re-elected last fall with a relatively healthy margin, any future ambitions that he may have in the state (or nation-wide) are going to be affected by his standing among Hispanic voters. The fact that he also sits on the Judiciary Committee would seem to be further evidence of why he is taking such a moderate tack at this point. Finally I'd note that Cornyn is currently the head of the Republican Senatorial Committee and thus in charge of the effort to recruit and fund GOP Senate candidates this cycle. Unlike other Senators he needs to be especially attuned to the national mood and electorate. His recent comments that the GOP needs to become a "national" party again show that he is at least aware of how much work needs to be done among certain groups (including Hispanics) of the electorate. Cornyn's colleague from Texas, Kay Bailey Hutchison, is also worth watching. While her most recent victory was double that of Cornyn's, she has perhaps a more difficult decision to make. First, Hutchison is widely expected to declare herself a candidate for the upcoming governor's race in which she will challenge incumbent GOP Governor Rick Perry in what will no doubt be a blockbuster primary fight. Thus, each decision she makes will be parsed for its electoral ramifications. Complicating this is the fact that she voted against Sotomayor in 1998. The result is some pretty complicated political jujitsu---how do you 1) win a Republican primary in which the incentive is to move right (anti-Sotomayor) yet leave yourself in a position to 2) win the general election where over 1/3 of the electorate is Hispanic???

A similar dynamic exists further west in Arizona. Here, John McCain is on the ballot next year. While he's had relatively little difficulty in his recent re-elections, the trendlines have not been favorable for the GOP in Arizona over the past few cycles. Democrats have gained 3 House seats since 2006 and many believe that it will go Democratic in the next presidential race. With 30% of the population being Hispanic, one can see why these trends have emerged. One can envision a scenario in which McCain sides with the Sotomayor pick in the end. It fits his image of an "independent" or "maverick" while also, I'd suggest, being an easy vote. He will have to explain why he changed his mind after his 1998 vote against Sotomayor, though. Because McCain doesn't sit on Judiciary, the fate of the Sotomayor nomination will probably be well known by the time it gets to the Senate floor. If it appears that Sotomayor is going to be easily confirmed, McCain can vote "Yes"without worrying that his decision will change the outcome. He maintains good standing back home while not incurring the wrath of his party--he wins on all counts. John Kyl, the other Arizona GOPer, is in a much more tricky spot. He faces the same treacherous demographics, yet 1) had a closer re-election last time; 2) sits on Judiciary (and will thus be forced to stake a position relatively early on); and 3) voted against Sotomayor in 1998.

Other Senators to watch??? I'd pick out Richard Burr from North Carolina. Electorally he's in a state that has been moving Democratic and has a Hispanic population that while not huge, can be decisive--see Nate Silver's discussion of the Hispanic vote in Obama's Tarheel State win. He's also been pulling some pretty underwhelming poll numbers for an incumbent. The fact that he's not on Judiciary and wasn't in the Senate in '98, and thus didn't vote on Sotomayor then, gives him a lot of room to maneuver now. He could vote "Yes" in the end too.

What Do We Know About White People???

I’ve been meaning to do some number crunching on the role of race in the presidential election, given how much discussion and coverage it received during the campaign. While I wrote about this subject a bunch prior to the election, I haven’t done much in depth analysis in the aftermath so let’s return to the topic. For this post I want to focus pretty much exclusively on the “white vote.” We know that among African American voters, not only was Barack Obama’s vote overwhelming (about 95%) but also that there wasn’t much variance across regions or states. While some of the states with high African American populations moved into the Democratic camp this year (North Carolina, Virginia) others did not (Mississippi, Alabama, Louisiana, etc.). It is this latter group of states that was the focus of the oft cited (and discussed here) “Schaller Effect”—the tendency of southern states with large black populations to have white electorates that vote overwhelmingly Republican. Essentially, Schaller argues, the politics of these states have become so defined by the issue of race over the decades that white voters, regardless of class, education, religion, etc., tilt Republican. With an African American on the ballot this year we got a chance to test this theory like never before.

To look at how white voters cast their ballots this year, and to determine whether any underlying patterns exist, I gathered the exit poll data compiled by CNN. The resulting spreadsheet is available here. Nationwide, Obama won 43% of the white vote. When we look at the vote preference based on race, we see that Obama won a majority of the white vote in 18 states, plus the District of Columbia. The map at left highlights those states and shows a pretty clear geographic clustering in the northeast, upper Midwest, and Pacific states with the exception of Colorado--a new battleground between the parties. If the electorate was 100% white he would have received 222 electoral votes, not enough to win the presidency. Thus, we look to those states where he lost the white vote (to varying degrees), but found a large enough minority electorate (black, Latino, Asian, etc.) to overcome his poorer showing among whites. This netted him an additional 10 states (FL, IN, MD, NV, NJ, NM, NC, OH, PA, VA) and provided him the solid electoral college victory that propelled him to the presidency. While there’s probably not anything overly surprising here—Democrats have for years had to rely on a multi-racial coalition to win the White House—looking at the raw numbers is instructive.

Conversely, if we look at the McCain vote among whites, we see a strong geographical dimension to his support. Nationwide, McCain received 55% of the white vote. The map left highlights those states where his white support was the highest--over 10% above his national average. Thus, in a cluster of mostly southern states the Republican nominee got 2/3 or more of the white vote.

Another way we might approach these numbers is to go to those states with the largest black populations relative to the state as a whole. Here we get to the phenomenon described by Schaller—how does Obama do among whites in the “blackest” states?? At left is a chart showing his performance among whites in the 15 states with the most significant black electorates. Overall, he won 8 of these states. However, with the exception of Washington, DC, Obama received a majority of the white vote in only 3 (Joe Biden’s Delaware, his home state of Illinois, and New York). He received more than his national average (43%) in only one additional state—Maryland. When we look at his performance in the Deep South states we seem to see evidence of the white backlash Schaller talked about with his extremely poor showing among whites in Alabama, Mississippi, Louisiana, Georgia, and South Carolina standing out. These states have large black populations but saw whites vote strongly Republican, a point noted in the map above as well.

What happens when we look at the inverse of these states—i.e. the “least black” ones?? Does McCain do as well as he did in the Deep South or did we see, conversely, Obama do really well? If McCain does equally well we might conclude that northern and southern whites aren’t really that different. If there is a geographical difference, maybe there is something to the hypothesis of racialized voting in the South. At left is a chart showing the 15 “least black” states in the country and Obama’s % of the white vote in each. Here his performance is much better. He won 8 of these states overall and in 7 of these 8 (with the exception of New Mexico) he not only got higher than his national average among whites, he won a majority of white voters. In fact, had some states in this group had a sizable enough minority population—Montana, the Dakotas—he may have added their electoral votes to his total as well. Clearly, it seems, there is a difference among white voters based on geography.

From this, can we conclude that white southerners voted against Obama because he was black?? Not necessarily. An alternative hypothesis is that these voters voted against Obama not because he’s black, but because he’s a Democrat. To get a better sense of what happened in November we need to go back further in time. Let’s look at 2004 when you had two white candidates on the ballot. If the results among white voters are quite similar to what we saw in 2008, our conclusions might need a bit more nuance. By looking at 2004 we can also get a sense of what changes might be taking place across the electorate and whether there is a geographic pattern. Thus, I also gathered the statewide CNN exit polls from the Bush/Kerry contest. What do we see?? Nationwide there wasn’t much difference between the support Obama and Kerry received among whites. In fact, Obama did better, getting 43% of the white vote to Kerry’s 41%.

Looking at the results state by state, we see that in all but 7 states, Obama did better among white voters than Kerry. In 8 states the performance of the two Democrats was identical. In the remaining 35 states, plus DC, Obama did better among whites than Kerry. If you look at the 18 states where Obama’s gains were the greatest (5% or more), he won 13. Four of these states voted Republican in 2004 (Indiana, Colorado, North Carolina, and Virginia). Thus, while the black and Hispanic vote was crucial in helping bring these states over to Obama’s side, we can’t ignore his increased performance among white voters as well.

Whereas Obama won 8 of the blackest states, Kerry only won 5. In these states, Kerry received below his national average among white voters (41%) in 8. Aided partially by increased support among African American voters (and also changes in turnout among whites and Republicans, plus greater support among Hispanics), Obama was able to outperform Kerry electorally. Narrowing further in on the white vote (and getting at the Schaller Effect) we see that in these states, Obama’s performance was worse than Kerry’s in four (Mississippi, Alabama, Arkansas, and Louisiana). Given that 2008 was a much better environment for Democrats than 2004 this is quite interesting. With a faltering economy, an unpopular president leaving office, and unprecedented national organization and fundraising, he did worse among some whites than did a candidate that, while white, was hardly a “good ‘ol boy.” Kerry, portrayed as elite, effete, “French,” and the epitome of northeastern liberalism outperformed Obama in much of the old Confederacy. Plus, given Bush’s cultural affinity with the South, his evangelical faith, and his political rearing in Texas, you’d have expected him to have high levels of support among southern whites. That McCain did better among some states’ white voters lends credence to Schaller’s argument, it seems.

So where does that leave us? With the low level of support Obama received among white voters in some states--relative to John Kerry and in an extremely favorable electoral environment--its hard to discount the assertion that race played some role in the vote. While some of these states have been heavily Republican, we saw Obama outperform Kerry among whites in equally Republican states in other parts of the country. We've also seen many of these states elect Democrats to other offices. The fact that white voters behaved differently in various parts of the country suggests that race is an issue that cuts across the electorate in a multitude of ways. Because race has been intertwined with politics in some places, and completely absent in others, we shouldn't make sweeping assertions about how important this cleavage is. Rather, we should approach the issue with nuance, an appreciation for historical context, and a willingness to explore multiple explanations for the outcomes we've observed. Given that we've had only one election to test our hypotheses, we should hold off on definitive statements until there's more data to parse.

Micropolitics in NYC: Don't Ignore the Trees for the Forest

For those of us who are fascinated by way ethnicity, religion, culture, race, income, and other variables affect political behavior, there's no place more compelling than New York City. Too often, I think, people gloss over the Big Apple's voting patterns because of its overwhelming Democratic tilt. In 2008, Barack Obama captured over 75% of the vote in 4 of the city's 5 boroughs (he got 47% in Staten Island). Underneath these numbers, though, are a myriad of stories. With so many voting blocs--some newly empowered, others in decline--New York politics is extremely fluid.

I recently came across this story from last summer in the NYT about the transformation in the city's electorate over the past several years. Whereas we normally think of "ethnic politics" being about the Irish, Italians, Jews, and other groups, the reality of today's Gotham is that Koreans, Russians, Pakistanis, Haitians, and dozens of other groups comprise the patchwork of voters that politicians must now cater to. The result, while perhaps more complicated for those running for office, is also empowering for these groups seeking to gain a foothold in America. As we celebrate St. Patrick's Day and Irish Americans' ascendance to our country's socio-economic and political heights, it is appropriate to appreciate how, historically, the ballot box has been a vehicle for groups' advancement. A similar story from October looks at how New York's Muslim community has experienced a political awakening.

Some of the most interesting scholarly work on New York's changing demography is being done by John Mollenkoph at the Center for Urban Research at CUNY's Graduate Center. Back in 2003, Dr. Mollenkoph co-authored this incredibly detailed study of the changes, and political implications, taking place in both New York and Los Angeles. It also provides some wonderful historical context for urban demographic change in other large cities. His findings challenge our conventional way of understanding urban politics. Whereas in past generations we saw a white/black divide and cleavage in big cities (and also Latino based in some places), the new urban landscape is much more balkanized and complex, requiring a more sophisticated approach to campaigning and governing.

To complicate things even further, we must remember that these ethnic or racial groups are not monolithic in their political behavior. Consider the maps at left, produced by the Center for Urban Research. They show the 2008 vote overall in the city as well as Barack Obama's performance vis a vis John Kerry's. While Obama did better than Kerry in most parts of the city, there are patches where McCain improved upon Bush's numbers. Given that Bush himself did relatively well in NYC--in the aftermath of 9-11--this further bump for the Republicans is quite interesting. The part of this map that immediately jumped out at me, and illustrates this need to dig deep into the micropolitics of the city, is southern Brooklyn. New York reports its election returns by Assembly district. Thus, Assembly Districts 45, 46, 47, 48, and 49 are of interest.

What caused these districts to be so different from the rest of the city in terms of its voting? One thing of note about this part of the city is that it has a large concentration of Orthodox Jews. Neighborhoods such as Midwood, Bensonhurst, and Borough Park have seen large numbers from this community settle in recent years. Likewise, Brighton Beach and Sheepshead Bay have seen an influx of Russians (many Jewish). The conservatism of the Orthodox Jewish electorate oftentimes gets overlooked given its relatively small size. Despite the discussion during the campaign season of Barack Obama's "Jewish problem," he ended up receiving 78% of the Jewish vote on election day--on par with the performance of John Kerry and Al Gore. Among Orthodox voters, however, there has always been a greater tendency to support Republican candidates given their positions on social and foreign policy issues. The Orthodox community is more likely to be pro-life, against gay marriage, and in favor of a hawkish defense and foreign policy than the Conservative and Reform movements. During the campaign the Jewish Press, which is a strong voice and based in this community, endorsed John McCain along these lines. Given this, then, we perhaps shouldn't be surprised with the voting results shown above.

While most of the data on demographic changes in the U.S. points toward a bright future for Democratic candidates--especially in regard to the Latino vote--its important to recognize that diversity entails tremendous complexity as well. Different voting blocs will bring with them different interests, agendas, and perspectives on policy. These agendas will compete with each other in an increasingly crowded environment, creating the potential for tension. While cities like New York and Los Angeles are on the front lines of this dynamic, it won't be long until its effects are felt much more broadly. Thus the need to dig deep and examine not just the forest, but the trees as well.

ElectionDissection.com Book Club--The Rosetta Stone for 2008

The first comprehensive examination of the '08 election has been published this week and I'm just beginning to get through it. They'll be a lot of posts generated by this analysis so I'm going to wait until I've had a chance to digest the numbers. The book is "How Barack Obama Won" by NBC's Chuck Todd and Sheldon Gawiser. The best thing about the book is that it gives a state by state and regional analysis rather than simply a national one. As readers of this site know, we much prefer these micro level examinations over more generalized descriptions of electoral outcomes. There's too much variance across the country. Also at first glance, it seems, is that their analysis is heavily driven by exit polling data. While this is certainly a useful tool--and something I've relied on in some of my analysis--it can't be the only data we draw upon. Until I'm able to get through the whole text, though, I don't want to get too critical. So stay tuned.

Obama's 365th Electoral Vote

When the votes were finally finished being counted, Obama's electoral vote count got increased by 1 due to an ununsual feature of one state's allocation of its votes. Nebraska is one of two states (Maine being the other) that awards its electoral votes by congressional district rather than giving the statewide winner the state's total haul. While Obama won each of Maine's two congressional districts, the Nebraska vote was split. Despite the fact that McCain won the state by about 15 points, Obama actually captured one of the congressional districts--the 2nd. As a result, he captured its electoral vote with the remaining 4 going to McCain.

While this outcome might be an interesting bit of trivia in its own right, its useful to compare 2008's numbers with 2004. In '08, Obama was able to claim the district with a narrow 3,370 vote margin of victory (138,809 to 135,439). In 2004, however, Bush won the district handily, capturing 61% of the vote (153,041 to 97,858). This historically strong Republican flavor of the district is seen in the fact that only once since voting for Harry Truman has it gone Democratic. It has a Cook Partisan Voting Index of R+9. Given this 11 point shift, I thought I'd explore the district a little more.

With the aid of Jack Huerter, one of my students and Omaha native, I got a sense of the district's demographics. The 2nd District is comprised of Douglas County and a small slice of neighboring Sarpy County. Douglas County is made up of the city of Omaha plus its suburbs extending westward from the city center. While being the most diverse of the state's 3 congressional districts, the 2nd is still overwhelmingly white with 10% of the population being African American and 6% being Hispanic. Overall, Douglas County provided 81% of the district's votes in 2008 while the Sarpy County portion provided the remaining 19%. Comparing the Douglas and Sarpy votes, we find that, using the 2 party vote totals, Obama beat McCain in Douglas 52% to 48% while McCain beat Obama in the Sarpy portion 57% to 43%.

To aid in the exploration of the district, I created the map shown above which focuses on Douglas County. Thanks to the Douglas County Elections Commission I was able to get both the precinct level voting data and precinct map. From it I created the color coded map that shows the distribution of Obama's and McCain's support throughout the county. I've created an excel file for the precinct level voting for Douglas County that is available here. Here is the original map without the color coding that provides a clearer picture of the boundaries and streets.

What we see is an Obama vote that is concentrated within the city of Omaha and some of its more western neighborhoods and a McCain vote that is, as we'd probably expect, heavily suburban. Within the city, the African American community is concentrated on the north side (see precincts in Ward 2 especially). For the Hispanic population of Omaha, the city's south side is where the largest concentration exists, especially in the areas around I-80. These south side neighborhoods were historically white working class and Omaha, like many cities, has seen this group move to the suburbs over the years. As we move westward out of the city, Jack suggests that the suburbs begin in the area of around 90th Street. We see from the map that this is where the McCain strength intensifies and we enter areas of solid red--the urban grid ends and is replaced by neighborhoods of winding streets and cul de sacs. A final group I was interested in was the Omaha student population. Creighton University is a mid sized Jesuit school located in the heart of downtown (precincts 2-15, 2-16). Its area went solidly for Obama. The University of Nebraska at Omaha has a sizable student body of about 15,000 but according to my Omaha expert is largely commuter based and thus not concentrated in any one voting area.

I had hoped to do a close comparison between the 2008 and 2004 numbers at the precinct level. However, in talking to someone at the Elections Commission, I was told that there were some pretty dramatic changes to the precinct boundaries, thus making a comparison difficult. Nonetheless, I did get the precinct level numbers that are available here so that you can make your own observations. On a very basic level we can get some sense of Obama's improvement over Kerry. In 2008, Obama won 151 of Douglas County's 350 precincts (43% of all precincts). In 2004, Kerry managed to win only 94 of the 341 precincts (28% of all precincts). While Obama's best precincts were no doubt some of Kerry's best, especially among African Americans, Obama would seem to have benefitted from a greater number of votes coming out of these areas.

The Black Student Vote and Barack Obama

Last week I came across this article about Obama’s win in North Carolina. Mentioned at the end is the claim that Obama’s win can be partially explained by increased turnout not just among African American voters, but especially students at Historically Black Colleges and Universities (HBCU's). The Tarheel State has 11 HBCUs. This got me thinking about the role of this student group across the country and whether or not we can point to any clear gains made by the Obama camp as a result of this energized student base. Also bolstering my interest in this question was the fact that last weekend saw the playing of the annual Bayou Classic in New Orleans, one of my favorite sporting events of the year to watch. If you don’t think the Obama win was important to the student bodies of Southern and Grambling State Universities, check out the halftime performance of the “Battle of the Bands”—in many ways a more passionate and heated contest than the football game itself (Southern here, Grambling State here).

Thus, I decided to build a database of voting across the HBCU universe and do some number crunching. The database has been posted here (Google account required) along with the explanatory key for each column.

HBCU’s were created, beginning in the aftermath of the Civil War, to allow African Americans access to higher education. The vast majority of these schools are located in states of the Deep South, which isn’t surprising given the degree to which blacks were excluded from college enrollment in these states’ flagship universities. Thus, HBCU’s have for decades acted as the route to the professions for many of the country’s black students. Currently, about 14% of all African Americans in college attend an HBCU and about 25% of the bachelor’s degrees awarded to African Americans come from one of these schools. Over time, several HBCU’s have not only welcomed non-black students but have become predominantly white or other minority. The University of Texas El Paso, for example, is now over 70% Hispanic. This school, formerly known as Texas Western, has an important place in collegiate athletics and HBCU history in that its 1966 national championship men’s basketball team was the first to start 5 black players (against the all white University of Kentucky dynasty). Thus, their role not only in educating the black community, but the nation as a whole, has been vital.

There are currently 103 HBCU’s across the country, located in 22 states and 76 counties. Only a small handful are outside of what we would consider “the South”—Lewis College of Business in Detroit; Cheyney and Lincoln Universities in Pennsylvania; Charles Drew University of Medicine and Science in Los Angeles are examples. Alabama is the state with the most HBCU’s—14. Next among the states with many such schools are North Carolina (11), Georgia and Texas (9), and Mississippi and South Carolina (8). Another thing one notices, and which is included in the database, is the diversity in the size of each of these schools. Some have only a few hundred students while others are quite large—Howard University, Florida A&M, Texas Southern, and Southern University are among the largest. Overall, however, in looking at enrollment size, it doesn’t seem as if these schools—within their respective counties—provide enough votes to truly sway an election.

When we look at how each of these schools’ counties voted this year—and compare this year’s race to 2004’s--what do we see? First, Barack Obama won 52 of the 76 counties in which HBCU’s are located. While on the surface this might have been useful for Obama in capturing several states, the reality is that many of these wins took place in states that went solidly for McCain. For example, as mentioned Alabama contains 14 HBCU’s yet was won by McCain with 60%. At the county level, Obama won 3 of the 8 counties containing these schools. Likewise, in Mississippi, Obama won all 6 of the counties containing its HBCU’s and in South Carolina he won 4 of 5 HBCU counties (McCain won each state with 56% and 54% respectively). This fact, it seems, illustrates a much larger point not only about the history of the HBCU movement but about America’s complicated political history.

In numerous posts over the past year or so I’ve dwelled upon the debate about white vs. black voting in the Deep South. What we’ve seen, through not only exit polling, but also simple geography and demographics, is that there has been a clear divide between black and white voters in the region. For example, most of these HBCU’s are physically located in counties that have very sizable black populations. 58 of the 103 HBCU’s reside in counties that are at least 30% African-American. Thus, it would seem as if the decision about where to locate these schools was very deliberate. Because these schools in many cases served a population that was underserved and lacking in resources it made sense to have them close to where the student body would hail from. Also, one wonders how successful these schools would have been (or even if they could have been created) had they been located in overwhelmingly white areas. For example, only one HBCU (Clinton Junior College in Rock Hill, SC) is located in a county that is less than 20% black in the Deep South states of South Carolina, Georgia, Alabama, Mississippi, or Louisiana. Worth remembering is the degree to which the residence of the black population in the South (as seen across counties) has been remarkably stable. Thus, returning to the vote this year, it seems as if the Obama vote we see at the county level is not so much a result of these schools bringing political change to an area, but rather amplifying the underlying preferences that have always been there. Obama’s candidacy energized this underlying Democratic allegiance and gave it reason to turn out in greater force than it had in years past.

When we compare this year’s vote to 2004, while we do see evidence of change, some of it seems due to other factors. Whereas Obama won 52 of the 76 HBCU counties, John Kerry managed to win only 39. Obama’s improvement on Kerry’s number took place across 8 states, flipping three counties in North Carolina (Cumberland, Forsyth, and Wake); 2 in Louisiana (East Baton Rouge and Caddo) and Texas (Dallas and Harris); and one each in Alabama (Jefferson), Delaware (Kent), Florida (Leon), Georgia (Peach), Pennsylvania (Chester), and Virginia (Henrico). The Texas wins in the counties containing Dallas and Houston can be attributed to no longer having native Texan George Bush on the ballot. Likewise, the addition of Delaware’s Kent County can be tied to Joe Biden’s VP nomination. Here is a breakdown of these counties, their Obama vote %, Black % vote, and HBCU enrollment:

Jefferson AL (Birmingham) 52.6% Obama, 39.4% Black, 3000 Students
Kent DE (Dover) 55.0% Obama, 20.7% Black, 3000 Students
Leon FL (Tallahassee) 62.2% Obama, 29.1% Black, 10000 Students
Peach GA (Ft. Valley) 53.4% Obama, 45.4% Black, 2500 Students
Caddo LA (Shreveport) 51.5% Obama, 44.6% Black, 1100 Students
E. Baton Rouge LA (Baton Rouge) 51.1% Obama, 40.1% Black, 11000 Students
Cumberland NC (Fayetteville) 58.9% Obama, 36.7% Black, 4000 Students
Forsyth NC (Winston-Salem) 55.3% Obama, 25.9% Black, 2800 Students
Wake NC (Raleigh) 57.3% Obama, 19.7% Black, 4300 Students
Chester PA (Cheyney) 54.6% Obama, 6.2% Black, 2800 Students
Dallas TX (Dallas) 57.7% Obama, 20.3% Black, 700 Students
Harris TX (Houston) 50.8% Obama, 18.5% Black, 10000 Students
Henrico VA (Richmond) 56.2% Obama, 24.7% Black, 1500 Students

Of those remaining counties, the only county with both a close margin of victory for Obama and a sizable HBCU population is East Baton Rouge Parish in Louisiana. Thus, whereas Southern University may have lost the Bayou Classic on Saturday, their vote may have helped tipped their county to Obama’s side, something Grambling State wasn’t able to do in Lincoln Parish which went heavily for McCain. Whereas Leon County Florida has a big HBCU population (Florida A&M) it was quite lopsided in its support for Obama. The presence of Florida State University there and its big student population might explain some of the margin and could perhaps show a greater tendency of white and black students to vote in a more colorblind way. For the North Carolina county flips, we see both relatively large margins for Obama and a relatively small HBCU population.

What are some other things we see when crunching these numbers??? Overall, Obama’s performance across these counties greatly improved on that of Kerry, something McCain was not able to replicate vis a vis George Bush. Compared to 2004, the Democrats got a higher % of the vote in 72 of the 76 counties considered here. Obama only underperformed Kerry in Jefferson County Arkansas (which he still won handily), Mercer County West Virginia, and Harrison and Wood County Texas. While McCain’s county %’s versus 2004 are dismal (only gaining in the 4 Obama declined in), an even worse scenario emerges when we look at raw vote totals. Given that the size of the electorate increases over time and that this year saw higher turnout than 2004, one would assume that both McCain and Obama would get more votes in each county than Bush and Kerry did respectively. Not so. Whereas Obama got more votes than Kerry in 70 of 76 counties, McCain only got more votes than Bush in 31 of the 76!!!! And remember, we’re mostly talking about the Deep South here. When you compare the improvements that Obama made with these declines suffered by McCain, you see how Obama was able to turn some areas from red to blue. Here, North Carolina jumps out.

Obama won North Carolina by 14,177 votes. In the HBCU database, we saw how Obama flipped Cumberland, Forsyth, and Wake County. Looking at the net vote gains (Democratic gains in 2008 over 2004 minus Republican gains in 2008 over 2004) in these three counties, Obama had a net increase of 126,563 votes. That’s your win right there. If you take other strong Obama counties in the state included here—Mecklenburg (Charlotte),
Guilford (Greensboro), and Durham (Durham)—you add an additional 162,154 votes!!! That’s nearly 290,000 votes gained in just 6 counties. Not only did Obama vastly overperform in North Carolina, McCain didn't even break even with past Republican showings. That is a sure fire way to lose.

Another state that I thought might show a similar dynamic was Virginia. Overall Obama had a 234,527 vote cushion. In the database here, there are 3 counties with an HBCU (Brunswick, Henrico, and Prince George) plus Hampton and Norfolk cities which are not part of a county jurisdiction. When you add up the net gains made by the Democrats one sees that Obama had a 70,928 vote advantage. Thus, while the Democrats increased their performance against the Republicans vis a vis 2004 in these counties, they had to rely on other counties, especially in northern Virginia to put them over the top. I should be clear that in neither North Carolina nor Virginia do I suggest that the HBCU presence and student vote were the reason these states switched from red to blue. I’d make the same caveat about the county level results as well. Rather, using this lens through which to view the election returns allows us to pick up on things we may otherwise have missed.

By focusing on HBCU’s, the data shows that we are for the most part looking at counties that are southern, black, and consistently Democratic. The Obama campaign and the 2008 election seem to have energized this electorate and given them the opportunity to elect the country’s first black president. Given the politics of the states in which most of these schools exist, we didn’t see the Obama campaign rack up massive numbers of electoral votes. States like Alabama, Mississippi, Louisiana, and South Carolina were never really in serious danger. Most of the gains made by the Democrats in these states and elsewhere seem attributable just as much to Republicans staying home as new Democrats turning out. In those states that did switch (North Carolina and Virginia) other factors surely contributed to the gains—greater white support for Obama, the economic crisis, etc. The HBCU’s, in short, didn’t make Barack Obama our next president. Nevertheless, beyond the more academic exercise of crunching election returns by using HBCU’s as a prime variable, one can’t deny the meaning that this election had to these schools’ students. If you don’t believe that, watch this.

County Flips, the Rural Vote, and Changes Out West

During the primary season, I came across the great site Daily Yonder which focuses on rural politics, economics, and society. The site is on outgrowth of the Center for Rural Strategies. In the aftermath of the election, Tim Murphy and Bill Bishop put together the above map that shows the counties that switched their vote between 2004 and 2008. The provide analysis here and here. While Obama's strength was certainly in the urban cores of the country, he did improve upon Kerry's performance in rural counties. Looking at the counties that flipped this year, the midwest was clearly Obama's strength. While his Illinois performance is unsurprising, the Wisconsin changes jumped out at me immediately after November 4 (see earlier post here). Also, as we know, Indiana (post here) and Iowa changed hands this year helping Obama solidify the entire region. I'm curious about the Democratic gains that took place in the string of counties running along the Minnesota/Dakotas borders. Finally, I'd note the Democrats' improved performance in the mountain west, a dynamic that has gotten a lot of play this year. A couple of points about the mountain west:

• Beyond the pickup of Colorado and New Mexico on the presidential level, Democrats gained Senate seats in both states.
• Democrats gained 5 seats in the House (2 in New Mexico, and 1 each in Idaho, Colorado, and Arizona). Democrats now have a majority in the House delegations of AZ, CO, and NM and split evenly with Republicans the Idaho delegation.
• Obama actually won Salt Lake County, Utah. While Utah's electoral votes won't change hands anytime soon, this flip was nonetheless pretty interesting.
• Note the rural county pickups in Montana. In the end, McCain ended up winning the state by only about 12,000 votes.

Demographic and State Shifts: Temporary Change or Realignment???

Charles Franklin of Pollster.com has created the above visual showing the performance of Obama versus Kerry across a range of demographic groups. What we see as a broad shift across virtually every group in the direction of greater Democratic support. Thus, like we saw at the state and county level, the movement toward the Democrats was quite impressive. What is no doubt the more fundamental question, however, is the duration or permanence of this shift. In the academic discussion of "realignments," what is necessary for a fundamental re-orientation of the electorate is not just movement of numerous social groups from one party to the other, but the durability of this movement. Also, with every demographic group here moving more Democratic, save three, one must wonder the degree to which the economic downturn and not other factors was ultimately responsible for the magnitude of Obama's win. Should many of these groups move back to their previous levels of partisan support over the next few years then November 4th's results won't seem so dramatic. In other words, with one election we don't have enough evidence to conclude we've had a realignment. These can only be viewed in the rear view mirror. However, much of the analysis of the exit polling that's been done over the past few weeks suggests that Democrats should be very happy about the trend lines. John Judis, who wrote "The Emerging Democratic Majority" with Ruy Teixeira discusses the possibility of this enduring majority here.

The above image of the state by state change compared to 2004 was created by Andrew Gelman at Princeton.

Population Stability and Polarized Voting in the Deep South

The above map, via Strange Maps, overlays the 2008 election returns(Obama counties blue, McCain counties red) with 19th century data on cotton production. The resulting image is striking. We know from exit polling that the white vote in the Deep South was strongly skewed to McCain (Whites in Mississippi, Alabama, Louisiana, Georgia, South Carolina, and North Carolina gave McCain 88%, 88%, 84%, 76%, 73%, and 64% respectively) and that the black vote was even more overwhelmingly pro-Obama. Beyond that, though, what we also see quite clearly is the degree to which the African American population in the south is largely concentrated in the same areas it was over a century ago. This is something that I noticed during the primary season as well (see posts here and here).

Thus, while the Great Migration (see post here) of the early and mid 20th century saw millions of southern blacks leave the Deep South for northern cities like Chicago, Detroit, New York, Philadelphia, and elsewhere, many millions also stayed. Also, many narratives of the Great Migration note that it was not unusual for individuals or families to return to south, either because the opportunities up north were not as bountiful as they believed or because the strong ties of family, community, and familiarity beckoned them home.

On a similar note, I highlighted recent Census data on mobility a while back (see post here). One thing that you note about several of the states in this region--especially Alabama and Mississippi--is that they have seen relatively little inward and outward movement of their populations.

Over at Pollster.com, Charles Franklin does an excellent analysis of the state by state change in Obama's support among whites vis a vis Kerry's 2004 performance. He notes not only the decline in white support for the Democratic nominee in the Deep South states, but also points out the improvement Obama made in Virginia and North Carolina. Ed Kilgore over at the Democratic Strategist also comments on this dynamic.

Obama's the Big Ten and McCain's the SEC But the Key to the Presidency Might Be the ACC

Rather than spend all of my time on the more traditional number crunching and map making, I thought I’d take a different approach to looking at last week’s election. One thing we know about American politics—and something I’ve written a lot about—is that it’s very regional. Different parts of the country tend to have different types of politics—driven by different demographics, economies, cultures, etc. Another part of American life that is heavily regional is sports. Where we live, where we go to school, and where we were raised tend to affect not only which sports we tend to be interested in (if any) but also the teams to which we’re loyal.

Both candidates this year, we’re told, are rabid sports fans. Much has been written about Obama’s love of basketball and his election day ritual of starting his day with a pick-up game. McCain, from what I’ve read, is a rabid boxing fan. So, can we merge politics and sport—in other words, how might we use the lens of sports to view the returns? As the college football season is approaching its most crucial weeks and college basketball is just gearing up (Go Marquette!!), I thought I’d see there were any parallels to the regionalism of college sports and the election returns.

College athletics is organized around a series of regionally based conferences, the six largest of which are part of the BCS (Bowl Championship Series). These six conferences are the Big East, Atlantic Coast Conference (ACC), Southeastern Conference (SEC), Big Twelve, Big Ten, and Pac Ten. The membership of each conference is as follows:

Big East: Cincinnati, Connecticut, DePaul, Georgetown, Louisville, Marquette, Notre Dame, Pittsburgh, Providence, Rutgers, Seton Hall, South Florida, St. John’s, Syracuse, Villanova, West Virginia

ACC: Boston College, Clemson, Duke, Florida State, Georgia Tech, Maryland, Miami, North Carolina, North Carolina St., Virginia, Virginia Tech, Wake Forest

SEC: Alabama, Arkansas, Auburn, Florida, Georgia, Kentucky, LSU, Mississippi, Mississippi St., South Carolina, Tennessee, Vanderbilt

Big 12: Baylor, Colorado, Iowa St., Kansas, Kansas St., Missouri, Nebraska, Oklahoma, Oklahoma St., Texas, Texas A&M, Texas Tech

Big Ten: Illinois, Indiana, Iowa, Michigan, Michigan St., Minnesota, Northwestern, Ohio St., Penn St., Purdue, Wisconsin

Pac 10: Arizona, Arizona St., California, Oregon, Oregon St., Stanford, UCLA, USC, Washington, Washington St.

These conferences cover the entire geographical breadth of the U.S. Thus, they represent the diversity of the American voting public. What I thought would be fun to do is look at how each conference voted last Tuesday. I took each school and looked at how its state and county (in which the school resides) voted. You see some pretty interesting things. The Big Ten, for example, is the most pro-Obama conference. All 8 states in which its schools reside voted for Obama, as did all 11 counties (there are 11 schools in the Big Ten since the addition of Penn St.). The Big Ten schools occupy the industrial Midwest from Pennsylvania to Iowa, the region of the country that Obama dominated, including his pickups of Indiana and Ohio.


The most pro-McCain conference?? Not surprisingly it’s the SEC. Encompassing the deep south, McCain’s base of support, McCain won 8 of the 9 SEC states (losing only Florida). At the county level, however, Obama actually did much better. He won 7 of the 12 counties in which these schools reside (Alachua—Florida; Athens-Clarke—Georgia; Fayette—Kentucky; Oktibbeha—Mississippi St.; Richland—South Carolina; and Davidson—Vanderbilt). Thus, we may see some evidence of the youth vote (which gave Obama 2/3 of their vote) coming through for the Democrat as well as all those liberal faculty members.

The most interesting conference—and the one that got me thinking about this question in the first place—is the ACC. Since the flip of North Carolina and Virginia from the Republican to the Democratic column, many are wondering if the “solid South” is beginning to disintegrate. Given that the ACC encompasses the border area of the east coast—part northern, part southern—its voting might reflect the bipolar nature of the region and whether it is now being pulled in one direction more than the other. Indeed, 5 of the 7 ACC states went to Obama (Clemson in South Carolina and Georgia Tech in Georgia went McCain). At the county level, Obama won 11 of the 12 ACC counties (losing only Pickens County South Carolina, home to Clemson). In most of these counties, the Obama margin was quite large. Only in Montgomery County Virginia (Virginia Tech) did he get less than 55%.

To round out the remaining conferences, the Big 12 was McCain’s next strongest at the state level. This conference encompasses the central plains, another area of Republican strength. Of the seven total states McCain won 5 (losing Colorado and Iowa). Like with the SEC, however, things at the county level were more evenly matched as each candidate won 6 counties.

The Big East is the largest conference geographically (I’m using its expanded basketball membership rather than its 8 team football membership). Stretching from Rhode Island to Wisconsin (Go Marquette!!) and also having South Florida, it’s the least geographically distinct. Nonetheless, it was strongly in the Obama camp. He won 11 of the 13 Big East states (losing Kentucky and West Virginia). At the county level, though, he made a clean sweep of the 16 counties. Monongalia County West Virginia (Morgantown) gave Obama a slim win with 51% as did Hamilton County Ohio (Cincinnati) with 52% and Hillsborough County Florida (USF) with 53%.

Finally, the PAC 10 was strongly Obama and might have given him a clean sweep had it not been for Senator McCain’s presence on the balance. With the exception of Arizona, Obama won 3 of the conference’s 4 states. At the county level, he won 9 of 10, losing only Maricopa County Arizona (Arizona St.) where McCain got 55%.

So, what can we learn from this?? All in all this may not teach us much new but rather allow us to use a different lens through which to view our politics. The dominance of Obama at the county level—he won 89% of the counties surveyed (57 of 64) illustrates, it seems, not only the role of the youth vote but also something about areas that revolve around university life. Even in the Deep South, Obama did well in those areas that have universities in their midst. Beyond students, one is likely to find a populace in these cities and counties with a higher degree of education and income than in the surrounding parts. As the Democrats’ share of the vote among the high education/high income demographic continues to grow, we might begin to see more change at the state level. The concept of the “ideopolis” is key to “The Emerging Democratic Majority” thesis. What we saw last week in North Carolina and Virginia (home to 6 schools in this survey) might be a harbinger for things to come.

Tippecanoe and Obama Too

Perhaps the biggest surprise of last Tuesday was Indiana's move to the Democratic column. Not since LBJ's landslide 1964 victory had the Hoosier State gone blue. In looking at the returns we see that Obama was able to drastically improve upon the performance of past Democratic candidates. His vote share grew in every county versus Kerry's 2004 performance, as the map at left shows (for some coverage from the Indianapolis Star, see here). He was able to win 12 more counties than Kerry (15 total), and his performance in Marion County (Indianapolis) allowed him to generate a huge number of votes to help tilt the state. In 2004, Kerry beat Bush in Marion County by 2% (6,000 votes). This year, Obama beat McCain there by 28% (107,000 votes)!

As during the Democratic primary, Lake County was also crucial to Obama's success. Essentially a part of the Chicago metropolitan area, Lake County's Democratic share increased 6% over 2004 and gave Obama a 71,000 vote margin.

You might also look at the Indiana college population for part of this year's margin. Monroe County (Bloomington) is the home of Indiana University. In 2004, Kerry won the county by 8%. This year, Obama's margin was 32% (20,000 vote margin). Tippecanoe County (West Lafayette) in north east Indiana is home to Purdue. Whereas Bush won Tippecanoe by 20% in '04, Obama switched the county, winning it by 12% (8,000 vote margin). Finally, the University of Notre Dame resides in St. Joseph's County (South Bend). Here we saw another switch--whereas Bush won the county by 2% in '04, Obama enjoyed a 17% win (20,000 vote margin) this year.

For all the details on Indiana politics, see Howeypolitics.com.

The Appalachian Problem Returns

The above map has been getting a lot of attention over the past day. It shows the counties where McCain actually overperformed Bush '04. We see a very clearly defined geographical region--Appalachia, especially the southern stretch--as the source of McCain's strength. During the Democratic primary, there was much discussion of Obama's "Appalachia Problem." For a refresher, the map below illustrates how deep Clinton's support was there (the more blue, the more pro-Clinton).

In a further examination of this, Ben Smith at Politico brings forward the final map below that charts American's self-identified ancestry. The same pattern jumps out as this part of the country is dominated by people who identify their ancestry as "American." In short, this is a region in which large numbers have family ties that stretch back multiple generations--very low levels of movement in and out. Thus, unlike parts of the neighboring states of Virginia and North Carolina, which have seen a much more dynamic population flow (especially Northern Virginia and Research Triangle of N.C.), this interior is perhaps more isolated from the trends and changes that have affected some of their neighbors. Virginia and North Carolina (and seemingly Colorado out west following a similar track) may have thus crossed the threshhold necessary to vote for a Democrat for the White House.

Big Blue Badger

As I hinted yesterday, the results from Wisconsin were staggering for Obama, compared to past Democratic performance. He flipped 32 counties from '04 and won statewide by 13 points. The map at left gives you a sense of where the greatest upticks took place. As I've been saying for a long time, the key for Wisconsin Dems is not just Milwaukee and Madison, its the Mississippi River counties. You can see the big Obama performance there.

Also of interest to me was the voting in the city of Milwaukee. Obviously Obama was going to win big there--and he did with with 78%--but I was curious about the south side. The south side of Milwaukee has always been Democratic territory but it is the home of Milwaukee's white ethnic (mostly Polish and German) blue collar community--Joe the Plumber land, in short. In 1964, George Wallace actually kicked off his campaign at Serb Hall, one of the south side's venerable gathering spots and home to a wicked Friday night fish fry. Given what we saw during the primaries in Chicago (see post here) I wondered whether McCain might actually win parts of the south side. Well, now that we've got some data, it appears that Obama actually carried it. See the map below which shows this year's citywide vote in comparison to 2004. McCain may have actually won some individual precincts here but until that data becomes available, we can't be sure. I'll also be interested in some of the inner suburbs that come off the south side--Greenfield, West Allis--to how Obama performed.

Lots more Wisconsin analysis to come.

Maps courtesy of the Milwaukee Journal Sentinel.

Some Thoughts On the South

If you’ve been reading my posts over the course of the campaign, you know that the issue of race is something I’ve written a lot about. Well, in a week we’re going to get a lot of data that will clarify many of the questions I’ve been raising and obsessing over.

One of the big questions that has been posed is how Obama will fare in the south. Will the legacy of George Wallace be purged forever? Will black turnout be so high in some states, as to make their electoral votes attainable for the Democrats? Do white southern voters think and vote differently than white non-southern voters? Does the Bradley effect exist, and if so, does it exist everywhere or is it more regionalized?

However these questions get answered, one thing that can’t be ignored is the fact that by having a reason to ask these questions—the reality of the Obama candidacy—we have demonstrated progress. We’ll no longer have to ask ourselves “If an African American got nominated for President, would he win?” Now we’ll know. While race won’t be the only reason Obama wins or loses, it will be a variable that we can finally start to put some flesh on. As a political scientist, I can’t wait to start digging into the numbers.

In thinking about the historic nature of the Obama campaign, I came across on old Time magazine from 1971 focusing on the “New South.” Featuring the just elected Governor of Georgia Jimmy Carter, the issue looked at how the region was trying to move beyond its old, race dominated politics:

Throughout the South, there are signs that the region is abandoning the fateful uniqueness that has retarded its development and estranged its people. William Faulkner’s South—heavy with ghostly Spanish moss, penumbral myths and morbid attachment to the past—is giving way to a South that has discovered it does not need fable to shore up its pride or the past to cloud its future. Moreover, a generation after the process was largely completed in the rest of the U.S., the South is caught up in an economic expansion that is reshaping the social order. The South has become at last a region of investment, both human and economic.

Making history—not living in its vainglories and myths—is the challenge and promise of the South today. The Southern frontier closed in that awful moment when the first man came to the South in bondage, locking the Southern experience into its tragic course. Three and one-half centuries later, the thrall can be broken, the frontier reopened. The South can grown rich while there is still time to safeguard the land from despoliation. It can acquire once more the political power of the sons who helped articulate the nation’s independence. Above all, it has a chance to shed its old hatreds and show the U.S. the way to a truly integrated society.

A little too optimistic or premature?? Perhaps. But should the election returns coming out of places like Virginia, North Carolina, Georgia, and other states next Tuesday show high levels of support for Obama, and produce electoral votes, we should remember that these votes were not generated in an instant or in a vacuum. Southern voters, like all voters, have grown up, lived, and been politically socialized in a historical and social context. This context evolves and develops over time. Thus, it’s appropriate to appreciate the history of this evolution.

Of course it may turn out that the southern vote will continue to be the outlier in terms of willingness to support a black candidate. Here we return to the thesis of Thomas Schaller in “Whistling Past Dixie.” In short, Schaller believes that Democrats are unable to compete in these Deep South states because of how race clouds the vote choices of white voters. An ingrained racial backlash—especially heightened with a black candidate at the top of the Democratic ticket—works to the advantage of the Republicans. While I don’t normally focus much on polls in my posts, recently released polling suggests that this explanation for southern attitudes might not be pure fiction.

Some good test cases for the role of race in these states might be found down-ballot on Tuesday. In both Georgia and Mississippi, two U.S. Senate races have become highly competitive. In Mississippi, incumbent Republican Senator Roger Wicker (appointed upon the retirement of Trent Lott) finds himself in a tough race against former governor Ronnie Musgrove. In Georgia, Saxby Chambliss is seeking a second term against Democratic challenger Jim Martin in a race that few thought would be close. What I’m interested in comparing is the difference between the votes for Musgrove/Martin and Obama. If we assume that every African American in Georgia and Mississippi who votes for Obama will also vote for Musgrove or Martin (not a perfect but, it seems, pretty reasonable assumption), then the difference between the Senate votes and the Obama vote should show us, roughly (we also need to account for the Hispanic vote) whether white voters had different evaluations of the presidential and senatorial candidates. If the Senate candidates outperform Obama, then there would seem to be reason to suspect that race was a factor.

Whatever happens on Tuesday, we’re going to be in a better position to answer some of the most vexing questions that have infused our politics going back generations. Stay tuned for the analysis.

Where They're Campaigning--A Macro Picture

In a number of posts over the past weeks, I’ve been looking at where each of the campaigns have been visiting to get a sense of their strategies and objectives. What can sometimes be missed in these daily snapshots is the larger picture or context in which the campaign takes place. Candidate visits are a part of the overall strategy of a campaign. In structuring their campaign, candidates must choose how to allocate their time, resources, and organization as well as how to frame their message and that of their opponent. With limited time, money, and organization the campaigns seek to best maximize the return on these resources within the context of a campaign’s constantly shifting position. If a candidate feels that a state is no longer in reach, they will no longer visit or air advertisements there, as witnessed by McCain’s pullout of Michigan in recent weeks. A candidate confident that a state is on their side won’t waste time with a visit—see Obama and Iowa, for example. If, however, a state seems either in danger of falling out of one’s column or potentially moving into it, a candidate will shift attention there. Here, witness the Obama campaign’s recent shift to Florida and Missouri.

Thus, by looking at where the two campaigns are spending their time, and how this has shifted over the recent weeks, we can get a sense, perhaps, of the larger direction the race is moving in. So, for example, are the campaigns spending most of their time in states won by Bush in 2004? If so, this would seem to bode well for the Obama campaign. If the Democrats are devoting resources to these states at a greater rate than states they won in 2004, they would seem confident in their ability to expand the electoral map and gain the electoral votes necessary for victory (assuming they aren’t sure they’ve already lost states they won in 2004). If McCain, likewise, is spending most of his time in the same states, this would imply a defensive stature. The Republicans can only afford minimal losses from the 2004 Bush map to win the presidency. Given the polling in some of these Bush states (Iowa, New Mexico, Virginia) McCain must find a way to win some states won by Kerry in 2004.

So, then, where are the campaigns actually spending their time? I’ve built a spreadsheet (see here) that tracks each visit by each presidential and vice presidential candidate since the end of the Republic an National Convention. These visits are tabulated by the Washington Post in their “Campaign Tracker” and updated daily. In the first column I’ve listed the date of each visit. Column 2 lists the candidate (Biden, Obama, McCain, Palin) making the visit. Column 3 is the state of the visit followed by the city in Column 4. In Column 5 I’ve noted the few visits that were fundraisers and not traditional campaign rallies or events. In the next 4 columns I’ve captured the “offensive” or “defensive” nature of each visit. This, I feel, will give a sense of the campaign’s real movement. Thus, we see Democratic visits to Bush 2004 states (offensive); Democratic visits to Kerry 2004 states (defensive); Republican visits to Kerry 2004 states (offensive); and Republican visits to Bush 2004 states (defensive). To add a visual dimension to this coding, I’ve color coded each column red (Republican) or blue (Democratic) to allow one to get a quick sense of what’s going on and to see these changes over time. I’ve also coded the visits that coincide with each debate yellow and not counted them in my tabulations as these cities were not chosen by the campaigns. I’ve also not counted a few visits that candidates made to their home state (Biden to Delaware and Palin to Alaska) because these visits do not seem part of the campaigns’ electoral strategy but were rather part of these VP nominees “roll-outs.” I’ve also not counted visits to New York as part of September 11 commemorations as official campaign stops.

In tabulating these visits, a few trends become clear. First, we see that by an overwhelming margin, the race (in terms of visits) is being conducted on Republican turf. This is very good news for the Obama campaign. 115 visits were to states won by Bush in 2004 (64 by the Obama campaign, 51 by the McCain campaign). Only 58 visits were to states won by Kerry in 2004 (26 by the Obama campaign, 32 by the McCain campaign). Put another way, the Obama campaign has spent 71% of its visits in states won by Bush while McCain has only spent 39% of its visits in states won by Kerry. Looking at this in more detail one sees, perhaps, a more dire picture for McCain. In the last ten days, the McCain campaign has only visited 3 states won by Kerry (Pennsylvania, New Hampshire, and 1 stop by Palin in Maine). Over the same time period, the Obama campaign has visited 7 states won by Bush (Florida, Missouri, Nevada, New Mexico, North Carolina, Ohio, and Virginia). Clearly, as Obama’s polling has improved over the past month, the campaign has felt emboldened to venture into a broader array of states, some of which were not thought to be winnable a few months back.


Another thing one notes is that the Obama campaign has been able to use its candidates—at least in terms of ground covered—more efficiently. One notices in the data a number of events attended by both Republican or Democratic candidates. These joint events, while probably larger in scope, are a less efficient use of the candidates’ time. Having two candidates in two places allows a campaign to accomplish more than having two candidates in one place. Here again we seem to have a Democratic advantage. Whereas the Republican ticket has had 19 joint events (thus coded as 38 visits), the Democratic ticket has had only 6.

When we look at the visits candidate by candidate, we don’t see a noticeable difference in terms of the candidates’ workloads. Obama has made 48 visits; Biden 42; McCain 42; and Palin 41. We do see some interesting trends, though, when we look at where each of these candidates has ventured. Joe Biden is probably the most interesting. He has made 43% of his visits to just two states—Ohio (13 visits) and Pennsylvania (5 visits). Thus, the “boy from Scranton” with his blue collar persona is clearly being used to try and shore up the segment of the electorate that has been most elusive to Obama throughout the campaign. Obama has made only 9 visits (19%) to those two states and concentrated his efforts on targets outside the Rust Belt like Colorado, Virginia, and Florida.

Thus, by looking at the campaign in a more “macro” sense, we can get some indication of the direction things seem to be moving. Some states that were earlier seen as competitive have seemed to solidify. Michigan, for example, hasn’t been visited by the McCain camp since September 23. It appears solidly in the Obama column. Likewise, the McCain campaign last visited Wisconsin and Minnesota on October 10. The decision by the RNC to stop airing ads in Wisconsin might signal that the Badger State is out of reach. Other states have suddenly gotten a lot more attention. Prior to October 9, Missouri had gone a month without being visited by the Obama campaign. Since the 9th, though, the Democrats have made 6 visits in a newfound sense of bullishness.

Over the remaining two weeks of the campaign, I’ll continue to monitor these visits to see if any further shifts take place. Should the polling begin to tighten in some states we’ll no doubt see the campaigns flock there. I’d also note, in closing, that the campaigns’ use of advertising will be a part of this larger strategy of resource allocation. The data for this year’s campaign advertising provided by the Wisconsin Advertising Project can be compared with these candidate visits to get a more complete picture of how the Democrats and Republicans view the current state of play. Like with visits, their recent findings suggest a battle being fought on Republican turf. The recent announcement by the Obama campaign that they had raised a staggering $150 million in September ensures that for the remainder of the contest, the Democrats will not be hindered by a lack of resources. While the laws of physics may prevent Obama and Biden from being wherever they would like to be, the opportunities available via the airwaves appear limitless.


**Update: Nate Silver takes a slightly different, but equally interesting, look at the data on candidate visits at FiveThirtyEight.com