Skip to main content

Hyper-Partisan Sorting in the American Electorate, 1976-2008

An Article by Thomas L. Day



The American body politic is increasingly defined by its hyper-partisanship. Where a generation ago the nominees from both the Republican and Democratic parties could compete for moderate voters of the other party, presidential elections are be- coming an exercise of partisan unity, where voters across the ideological spectrum almost uniformly vote for the nominee of their own party. I attribute this trend to three historical developments: the flight of southern social conservatives from the Democratic Party, the rise of micro-targeting technologies and the decline of objective broadcast news.


About the Author

Thomas L. Day is a master’s candidate at the Irving B. Harris School of Public Policy at the University of Chicago. His writings have appeared in the Washington Post, The Guardian, ESPN The Magazine, Deadspin, and Yahoo! Sports. Day was a bureau reporter for McClatchy in Kabul, Afghanistan, in 2009 and 2010. He holds his undergraduate degree from Penn State University and his master’s degree from Northwestern University’s Medill School of Journalism.


Introduction to the Dissonance of Ecuador’s Yasuní-ITT Initiative

Television coverage of the 1976 election results began at 6:30 p.m. East- ern Standard Time. Thirty percent of all televisions tuned into CBS and its longtime venerable news anchor, Walter Cronkite. Another 281 percent and 18 percent tuned into NBC and ABC, respectively. Those who tuned in witnessed the most anxious presidential election night since 1960. Former Georgia Governor and Democratic Party nominee Jimmy Carter scored victories in the states of South Carolina, Tennessee and Texas. Republican Presi- dent Gerald Ford carried Illinois, New Jersey and California. Later that evening, from a hotel room in downtown Atlanta, Carter phoned New York City Mayor Abraham Beame, inquiring about voter turnout in key precincts.2 Carter could not win without New York in his column, and the networks had not called the state as the evening labored on. At 1:00 a.m. on the east coast, the states of Alaska and Hawaii had cast their final votes, and neither Carter nor Ford had surmounted the 270 Electoral College vote mark needed to win the presidency.

It was not until 3:30 a.m. that NBC and ABC were ready to announce a win- ner. At the White House, Chief of Staff Richard Cheney pulled Ford into a private room to brief the President on the latest returns.3 Moments earlier, a wire report from United Press International read simply: “FLASH: WASHINGTON—CARTER WINS PRESIDENCY.” CBS, having corrected its premature report that Carter had won the state of Oregon a half-hour earlier, waited. The state in doubt was Missis- sippi. When NBC and ABC made their announcement, CBS could no longer wait. Cronkite, the most trusted man in America, told the country just before 4 a.m. that the election was decided. President Ford went to bed a lame duck and Jimmy Carter emerged the next morning as the 39th President of the United States.

After the 1976 election, the political map would have to be redrawn. Jimmy Carter was the last Democrat presidential nominee to hold what had been known as the “solid South” for Democrats. The century-long Democratic dominance of the south—dating back to Reconstruction—was finished. For the Republicans, the rise of social conservatives would gradually strengthen the party’s hold on rural voters, conceding states with large urban populations to the Democrats.

A generation hence, the Democrats do not even bother competing to win South Carolina, Tennessee or Texas in presidential politics. The state of Missis- sippi is invariably called by the news networks the moment the polls close. Simi- larly, the Republicans have largely conceded Illinois, New Jersey, and Califor- nia—all states that went for Ford in 1976—to the Democrats in the most recent presidential elections. The “solid South” for the Democrats is now the solid for the GOP. States north of the Mason-Dixon Line invariably fall to the Democrats. If the Democratic nominee for president saw it necessary to call supporters in New York City on election night, pleading for good news, the campaign’s fate would already be sealed. When it comes to electoral strategy, we hardly live in the same country we did in 1976.

A survey that year by the American National Election Studies group (ANES) asked voters if they saw “important differences in what the Democratic and Re- publican parties stand for.” 46 percent of respondents said yes, they did perceive significant ideological differences between the two parties, and 44 percent re- sponded that they did not. ANES posed the same question to voters again in 2008. 78 percent of respondents said that they saw differences; only 21 percent said they could not.4 This political environment did not come to be without fundamen- tal changes in the American political institution. The dominant ideologies of the aforementioned states have not changed.5 Voters have simply found an ideologi- cal home in one party or the other, and have increasingly voted with that party. Somewhere on the ideological continuum, a distinct and growing divide has sepa- rated the Democrats from the Republicans. Once this division was established, a dramatic increase in partisan voting for both Republican and Democratic voters has redefined the American body politic so much so that it scarcely resembles the one that guided the Carter and Ford campaigns 35 years ago.

While the 1976 election stands as a turning point in the continuum of his- tory—at which the political map of one era was discarded in favor of a new map to define American presidential politics—it is important to note that this process of change was incremental. In this paper, I will show how the American political landscape has evolved, plotting three moments in the timeline of history that fun- damentally changed how voters identify themselves and vote. The first moment comes with the signing of the 1964 Civil Rights Act, continuing through the urban riots of the 1960s. The second moment comes with the rise of “micro-targeting” technologies, which parties use to efficiently send clear messages directly to sub- groups of voters. Finally, the emergence of partisan programming on television news networks and the explosion of blogs, correlates with a decline of objective broadcast television news. These three turning points in history define the cur- rent American body politic. While other trends and movements, such as the rise of the “Religious Right” for example, are important to understanding the current American body politic, these three moments in history have fundamentally al- tered and defined American voting patterns, where Republican voters vote for Re- publicans and Democratic voters vote for Democrats.I This rise in partisan voting will be quantified using the same metrics employed in Princeton University pro- fessor Larry Bartels’ influential 2000 paper, “Partisanship and Voting Behavior, 1952-1996,” published in the American Journal of Political Science. In that paper, Bartels demonstrates that partisanship increasingly guided voting behavior in the latter half of the 20th Century.6 I conclude that partisanship has continued to drive voters’ choices since Bartels published “Partisanship and Voting Behavior.” In fact, the trend toward increasingly greater partisan voting accelerated in the 2000 and 2004 Presidential elections.

I measure partisanship with the American National Election Studies seven- point party identification scale.II Respondents who strongly identified with the Republicans, Republicans who identified with the Republican Party and indepen- dents who lean toward the Republican Party are all coded +1; strong Democrats, weak Democrats and independents who lean toward the Democratic Party are coded -1. All other respondents are coded 0. Voting behavior is also coded using Bartels’ original indicators in “Partisanship and Voting Behavior;” Republican voters are coded +1 and Democratic voters are coded 0. Voters who voted for a third-party candidate and nonvoters are omitted from the data set. Table 1 was in- cluded in the original Bartels paper. The coefficients of his probit model measure the how strong identifiers, weak identifiers, and independent leaners depart from the vote choices of pure independents.

When Conservatives Started to Leave the Democratic Party

It is difficult to track trends in ideological self-identification among voters from one time period to another, as the terms “conservative” and “liberal” seem to take on entirely fluid meanings relative to temporal political context.III Self-identified liberals largely supported military operations in Vietnam at the beginning stages of the conflict; however, by 1968, they were so outraged by the war that they pushed Lyndon Johnson off the Democratic presidential ticket.7 Self-identified conservatives have championed the cause of limited government, yet flocked to the defense of George W. Bush in 2005 when a series of articles in the New York Times revealed the Bush Administration’s approval of extensive domestic surveil- lance and warrantless wiretapping.

The American National Election Survey has attempted to measure changing ideologies by asking voters to place themselves on a seven-point issue scale, by which the most liberal response is a 1 and the most conservative response is a 7. For example, ANES has continued to survey voters on “women’s rights,” asking respondents if they believe women deserve an “equal role…in business, industry, and government” at one extreme, or if they believe a “woman’s place is in the home” at the other extreme.8 For a generation of Americans that has long accepted women in the workplace, the question seems antiquated. By 2008, the median respondent was at the most liberal extreme of the scale.9

We can, however, observe ideological self-identification as it relates to liberal and conservative political symbols, at least in recent history. In the current discourse, the images of the financial crisis of 2007-2009 have provided powerful political sym- bols. We clearly see how the image, or the political symbol, of a Wall Street banker has galvanized liberals under the “Occupy Wall Street” movement. For a previous generation, the images of police beating civil rights demonstrators as they marched through Selma, Alabama, on March 7, 1965 (“Bloody Sunday”) fundamentally al- tered national views of the symbols associated with the civil rights movement.

On July 2, 1964, Johnson signed the Civil Rights Act into law, outlawing segregation in schools and public facilities. The act passed after national (if not southern) public opinion swung in favor of its passage as response to the violence in Selma. A bloc of southern Democrats, led by Georgia Senator Richard Russell, had filibustered the bill for more than two months. Moments after signing the bill, Johnson told an aide, “We have lost the South for a generation.”10,11 Whatever conservatives’ feelings are about the Civil Rights Act of 1964 now, at that moment in history, conservatives viewed the Civil Rights Act as an infringement on states’ rights, and they saw a Democrat signing it into law.

It was during the Johnson presidency that ideology relative to political sym- bols began to take a form resembling the current political landscape. Before the Johnson presidency and his “Great Society” program, the term “liberal” had a far different connotation than it carries today. “Liberalism was conjoined with pic- tures of workers, often unionized and almost always white, hard-working people, playing by the rules, and trying to get ahead,”12 noted Bucknell University profes- sor Christopher Ellis and University of North Carolina professor James A. Stimson in their 2009 paper, “Symbolic Ideology in the American Electorate.” After the Great Society, with its dramatic increase in welfare programs, liberalism took on an association with an “unsympathetic portrait of a largely non-white under- class,”13 Ellis and Stimson noted.

Never were middle- and upper-class whites more unsympathetic to under- class, non-white beneficiaries of the Great Society than during the wave of urban riots from 1965 to 1968. “The riots were a body shock to American politics, events which were not unprecedented in American history but certainly unprecedented in the television age,”14 Ellis and Stimson argued. “Quite probably they are a big part of the story of declining support for the idea of liberalism as well.”15 After Johnson assumed the presidency and signed the Civil Rights Act of 1964, con- servatives, particularly Southern conservatives, began gravitating toward their natural ideological allies in the Republican Party. The 1976 election would be the last time that the Democrats would even compete in, much less win, much of the Deep South.

After the Civil Rights Act, conservatives in the Republican Party demanded their national ticket reflect a conservative platform and, to a lesser extent, liberals in the Democratic Party expected their ticket to push the left-of-center agenda. In 1964, Arizona Senator Barry Goldwater captured conservative hearts that were longing for a true conservative after the middle-of-the-road Richard Nixon ticket failed them in 1960, and ultimately the GOP nomination with his galvanizing campaign slogan: “A choice, not an echo.” Eight years later, the Democrats nomi- nated South Dakota Senator George McGovern, an extreme liberal who pledged to redeploy all U.S. troops from Vietnam within 90 days of taking the oath of office. Both Goldwater and McGovern were soundly defeated, but succeeded in pulling their parties toward their respective ideological bases.


The partisan voting trend accelerated in 1980 with the emergence of Republican political strategist Lee Atwater and the ever-refining art of voter targeting. Richard Nixon had instantly grasped the opportunity to exploit southern resentment of the Civil Rights Act of 1964 to turn the conservative southern Democrats into Repub- licans; Atwater, with his thick South Carolina draw, was the one who solidified the deep red Deep South. Before Atwater, there were conservatives in the South who remained in the Democratic Party. After Atwater, there were next to none. He ended the Democrats’ dominance of the South by doing exactly what Barack Obama bemoaned twenty years later in Boston during his keynote address to the 2004 Democratic National Convention: by slicing and dicing the United States into a liberal America and a conservative America, a white America and a black America, gay and straight, Christian and Jewish, and other potential divides.17, IV

A self-described “political nymphomaniac,”18 Atwater skyrocketed up the GOP ranks from the college Republicans into the White House before turning 30 years of age. He would make a name for himself in the 1980 GOP South Carolina primary, where Ronald Reagan, having defeated George H.W. Bush in New Hamp- shire, was looking to shore up the nomination. Atwater, Reagan’s South Carolina campaign manager, had assembled a legion of volunteers to identify and mobilize Republican voters before Reagan stepped foot on Palmetto State soil. With less than a week before the primary, more than a quarter of a million GOP voters had been contacted from Atwater’s web of phone banks.19

The Reagan campaign’s coup de grace of Bush came with a television ad- vertisement attacking Reagan’s future Vice President as a liberal, narrated by Reid Buckley, brother of conservative hero William F. Buckley (the two men had virtually indistinguishable voices20). It was the sort of dog whistle that the Re- publicans would soon make into an art form. Only the rock-solid conservative voters of South Carolina would recognize and respond to the voice of William F. Buckley (or an imitation of it). Reagan soundly defeated Bush in South Carolina and coasted to the Republican Convention from there.

Messages no longer were aimed at a mass audience, but instead they tar- geted smaller, more defined audiences and voting blocks. One unmistakable “dog whistle” was particularly effective for the Reagan campaign. In August of 1980, Reagan opened his general election campaign with a speech about “states’ rights,” a coded phrase for resistance to desegregation during the civil rights movement, to an audience in Philadelphia, Mississippi, where members of the Ku Klux Klan murdered three civil rights activists 16 years earlier. For many observers, Rea- gan’s speech was a direct appeal to southern whites who had shunned Richard Nixon in 1972, voting instead for Alabama Gov. George Wallace, an arch segrega- tionist.21 It was certainly effective. Reagan easily won the state of Mississippi and swept every state in the south except Carter’s home state of Georgia.

After he defeated Carter to win the presidency, Reagan hired Atwater to join the White House Office of Political Affairs. Consigned to the bowels of the Old Executive Office Building, Atwater penned several memos that would guide Rea- gan’s reelection strategy and his Vice President’s campaign for the White House in 1988. One Atwater memo, a 72-page manifesto penned in March 1983, divided the Southern voting electorate into three blocs: African-Americans, country-club elites, and populists. African-American voters were to be conceded to the Demo- crats, and country club elites were assumed to go for Reagan.22 Atwater called for the Reagan campaign to aggressively target the populists, defined as middle-class, church-going, blue-collar voters who usually vote with the Democrats, but who could be persuaded. Increasingly voluble Christian evangelicals had begun to coalesce around a strongly conservative social agenda. Reagan’s 1984 campaign, steeped in patriotic imagery and social conservativism, would be largely defined by this target voting bloc.

Four years later, Atwater was the campaign manager for George H.W. Bush. He not only saw the need to remake Bush’s image—that of a blue-blooded, for- merly pro-choice, Yalie son of a U.S. Senator—into that of a cornpone Texan, but he actually succeeded in doing it. The strategy was simple and aggressive: approach conservative Democrats, speak to their concerns, and rope them into the GOP fold. The appeals sometimes exploited race, as with the infamous Willie Horton advertisement. Horton, a black man convicted in a 1974 murder case of a 17-year-old gas station attendant, was permitted 10 weekend furloughs from his life sentence at a Massachusetts state prison. On his tenth furlough, Horton fled to Maryland, broke into a home, stabbed and pistol whipped a man and raped his fiancé. The man who signed off on his furlough, Massachusetts Governor Michael Dukakis, was Bush’s general election challenger. A brutal campaign ad, produced by a GOP-allied political action committee, recalled the Horton furlough with the sardonic bookend, “Weekend prison passes: Dukakis on crime.” Horton’s mug shot appeared in the ad, darkened.

The Bush campaign aggressively painted Dukakis as unpatriotic, giving prominence to “wedge issues,” or issues that pit voters of an opposing party against one another. Bush campaigned on a proposed Constitutional amendment to ban the burning of the American flag (which Dukakis opposed), and charmed the emerging “religious right” by choosing archly conservative Indiana Senator Dan Quayle as his running mate. For the populist voters in the Democratic fold that Atwater had targeted, the effect was magnetic.

The product of his campaign was evident in post-election surveys archived by the American National Election Studies. The Republican Party brand had be- come clearer. Blue-collar voters—frequently from rural areas, socially conserva- tive, and white—began migrating to the GOP. In the 1976 election, an election that sent Jimmy Carter to the White House by 2.1 percent of the popular vote, only 34 percent of voters identified as “blue-collar” workers voted for President Gerald Ford. On Election Day 1988, the Bush-Quayle ticket would get 50 percent of the self-identified blue-collar vote.23

Bush rewarded Atwater by naming him the chairman of the Republican Na- tional Committee. A year into his term, Atwater collapsed at a fundraiser for Texas Senator Phil Graham. He was diagnosed with an inoperable brain tumor shortly after. Atwater wouldn’t make it to see Bush’s 1992 reelection campaign. Another strategist would emerge from the 1970’s generation of college Republicans to carry on Atwater’s legacy: Karl Rove.

Rove’s first notable job in politics was to raise money for Bill Clement, a Re- publican candidate in 1978 for governor of Texas. Clement started out a long shot. All Republicans in Texas were long shots then. In 1978, of the 18 members of the Texas delegation in the U.S. House of Representatives, two were Republicans. In the Texas State Senate, three of 31 senators were Republicans. Not one of the 22 statewide elected officials was a Republican.24 Yet, the Texas Democrats shared little in common with the liberals that dominated the party in New England. By 1978, the Christian right was emerging in full splendor in Texas. Social conserva- tives only needed a nudge to leave the Democratic Party. They got it when Texas Attorney General John Hill, the Democratic nominee for governor, sued a network of Christian boarding schools over alleged abuse of the students. Texas conserva- tives sided with the schools, and then turned to Clement and Rove, who already knew their names and their phone numbers.25 In one year of work for Clement, Rove took a list of about 5,000 donors and expanded it to nearly 50,000, collecting $1 million in donations, which was a tremendous amount of money to fund a run for governor.26 Clement would become the first Republican governor of the state since Reconstruction. Once in Austin, Clement would hire Rove to be his chief of staff, and Rove’s path would soon converge with a young George W. Bush.

The 1990s and 2000s saw the appeals to subgroups become more and more refined. Where Reagan targeted southern conservative Democrats, the campaigns of the next two decades targeted much more narrowly defined groups of vot- ers. Technological advances allowed campaigns to not only assemble names and phone numbers, but data on personal beliefs, habits, and lifestyles. Campaigns in turn figured out how to push the right buttons to get liberals and conservatives to come to their ideological homes.

Perhaps the goals of micro-targeting were most eloquently explained in Applebee’s America, a 2006 book coauthored by Matthew Dowd, chief political strategist for George W. Bush and Douglas Sosnik, Bill Clinton’s political director. (Former Associated Press reporter Ron Fournier also coauthored the book.) Simi- lar to how Applebee’s tailors its restaurants to fit the character of the local area they serve—like posting an image of a local high school football hero at the front door—the authors explain that political strategists must also tailor their mes- sages to subgroups of voters based on their individual preferences: “Who are their friends? Where do they get their information? Who do they turn to for advice? What are their hobbies? What magazines do they read? Where do they live? What car do they drive? Where and how do they shop? What do they do for vacation? What angers them? What makes them happy? What do they do for a living? These and thousands of other lifestyle questions form a vast constellation of data points that Presidents Bush and Clinton (employed).”27

As the Clinton reelection campaign headed into the home stretch of 1996, his team had amassed a vast database of phone numbers of voters all across the country, subdivided into 55 groups based on identifying characteristics.28 With the data on hand, the Clinton team went about pushing the right buttons. Micro- targeting so strongly guided the Clinton election strategy that the President would pick vacation spots — hiking through Yellowstone National Park, for example — on their appeal to subgroups of voters.

If the Clinton team introduced micro-targeting data mining, George W. Bush’s reelection campaign eight years later would perfect the technique. Initially reluc- tant to spend large sums of money on micro-targeting, Rove immediately directed $3 million in Bush campaign funds toward micro-targeting when told that the he could reach nearly 12 million “unreliable Republicans,” ten million more than his staff estimated the Bush campaign could reach without micro-targeting.29

Voters may not vote according to what candidate is closest to all of their beliefs on the issues, but on the issues that matter, they often do. That’s where micro-targeting enters the equation. By identifying the issues that matter most to a certain group of voters, campaigns are able to take their messages right to the people they want to reach and none that they do not. Candidates will frequently take soft positions on issues when they know they are speaking to a broad audi- ence, then take much harder positions when they know they are speaking to a group of voters who are strongly and primarily concerned with one particular issue. For example, as the 2004 campaign began, President Bush endorsed amend- ing the constitution to ban legally sanctioned gay marriage. Then he hardly ever said another word about the subject. Not one of his television advertisements mentioned his support for the constitutional amendment.30 The Republican Party and groups aligned with the GOP flooded Christian conservatives with direct mail holding Bush as the only man standing between liberals and legally sanctioned gay marriage. Focus on the Family alone sent direct mail to 2.5 million households.31

“Persuadable voters in the electorate—those most likely to be receptive to a candidate’s campaign appeals—are often cross-pressured between their partisan loyalties and policy preferences,” wrote Harvard University’s D. Sunshine Hillygus and the University of Arkansas’ Todd Shields in their 2008 book, The Persuadable Voter: Wedge Issues in Presidential Campaigns. “Information and communication technologies have made it more efficient for candidates to narrowly target these per- suadable voters on the specific issues they care about,” Hillygus and Shields add.32

The Emerging Dominance of Ideologically-Biased News

In the last ten years, one element of the American political institution has rein- forced the partisan battle lines: the media, particularly the blogosphere and tele- vision news during prime-time hours.

We can now safely write the obituary of broadcast news, at least the broad- cast news enjoyed by previous generations. If there were still only three main news programs—the three network evening news programs—to inform one’s world, it’s possible that evening news programs would continue to thrive. How- ever, since the explosion of cable news and the emergence of online news, news media aimed toward a broad audience have seen their viewership and readership sharply decline. This is the natural consequence of the multitude of news outlets that exist today, and requires news networks to target specific audiences more. In 1970, as televisions reach near-universal integration into the American landscape, 80 percent of all viewers tuned into the three network news broadcasts. That fig- ure is now less than 40 percent.33 (More than half of all television screens tuned into ABC, NBC and CBS to watch the 1976 election night returns; on election night 2008, 30 percent of all television screens tuned into the three networks.)34

In their stead has come a wave of 24-hour news networks with increasingly ideologically driven programming. It should now be clear that conservative news watchers turn to Fox News, where they have their views reinforced, and liberals turn to MSNBC, where they too avoid the inconvenience of having their beliefs challenged. As Eric Lawrence, John Sides and Henry Farrell note in their 2010 paper, “Self-Segregation or Deliberation? Blog Readership, Participation, and Po- larization in American Politics,” this is a reflection of the public’s social need to interact with those who agree with us, while protecting ourselves from those who do not. “Individuals prefer social contexts populated by others who share their core political values and avoid social discourse with people who disagree with them profoundly over politics,” Lawrence, Sides and Farrell note. “Ultimately, homophily within networks likely coincides with polarization — that is, the diver- gence of competing partisans or ideologues, such that individuals who initially leaned to the left find themselves moving farther left over time, and individuals who initially leaned to the right move farther right.”35 Online news sites have also proven skilled at filtering out the disagreeable. Unlike television news, it is difficult to fault the online news industry for presenting a diverse platform of opinions; if someone has an opinion, there is someone else blogging that opinion.

In 2005, just as the blogosphere was beginning to emerge as a platform for political discourse and the Fox News Channel strengthened its dominance of the television news ratings, Princeton University’s Markus Prior surveyed 2,358 U.S. residents on their media preferences and consumption. His findings, published in “News vs. Entertainment: How Increasing Media Choice Widens Gaps in Politi- cal Knowledge,” found that media consumers responded to the increase in media choices not by finding the best, most informative reportage, but by simply find- ing the sources that best reflected their own views, or by declining to consume news altogether. A logical consequence of the explosion of media sources might have been a more informed electorate; alas, no increase in voter information was found. “In the low-choice broadcast environment, access to the medium and ex- posure to the news were practically one in the same, as less politically interested viewers had no choice but to watch the news from time to time,” Prior noted. No longer. Television and online news sources do not simply compete amongst them- selves for consumers, but with unrelated sources of entertainment. This reality has mandated that news sources tailor their programming toward consumer pref- erences. “Accidental exposure should become less likely in a high-choice media environment because greater horizontal diversity (the number of genres available at any particular point in time) increases the chance that viewers will find the content that matches their preferences,” Prior argued.36

I conclude that the cumulating effect of this media environment has been the quarantining of voters into their own ideological fortress. Where Walter Cronkite told his viewers “that’s the way it is,” his famed tagline now seems more apt for news consumers, not the news reporters. Consumers are informing the news networks and news websites of “the way it is,” and finding news that reflects their views. As voters vote on how they view the previous four years—a view reinforced by their chosen news sources—they have been increasingly likely to vote according to their party. In other words, voters are increasingly viewing the previous four years through the lenses of their party—a trend that is no doubt reinforced by the news they receive.

Updating Bartels’ “Partisanship and Voting Behavior”

It is conventional wisdom that the GOP “base” carried George W. Bush into the White House in 2000, then kept him there four years later. Conversely, by nearly any measure, no president in modern times has been as reviled by voters in the opposing party as the 43rd Chief Executive. This logic is reflected in the Table 2 and Figure 1, which extend Larry Bartels’ data through the 2008 presidential election.V In the two election cycles after Larry Bartels’ study, the trend toward greater and greater partisan voting increased to unseen levels in the modern era.37

To see how strongly party attachment has increasingly influenced voting be- havior, perhaps it is best to match the 1976 election against the 2004 election. Both Jimmy Carter and George W. Bush won the 1976 and 2004 elections, respectively, with fewer than 300 Electoral College votes and a popular vote victory of less than three percentage points. Yet, the movement of strong and weak identifiers from 1976 to 2004 toward their own parties is evident. It is important to note that a slight downtick in partisan voting is evident in the 2008 election data. This is certainly a reflection of then-Senator Barack Obama’s success in luring moderate independent and Republican voters into the Democratic fold. Is it also a reflection that the trend toward partisan voting has ebbed? I would argue not. There did not seem to be a fundamental change in the partisan dynamic of American politics during the 2008 election. A true decline in partisan voting would have brought an increase in Democrats voting for McCain as well as Republicans supporting Obama. Of course, no such increase happened. While the 2008 election marked a decline in partisan voting relative to the 2004 election, the data demonstrates that it was still a more partisan election than 1996, and a significantly more partisan election than 1976. Gerald Ford in 1976, even in losing, won a far greater share of Democratic voters than did McCain in 2008. Where Gerald Ford earned the vote of 22 percent of Democrats, John Mc- Cain only received the votes of ten percent of Democrats 32 years later.40 This is a reflection of an electorate more likely to vote based on party affiliation.

President Obama’s appeal did not, to a great extent, transcend party bound- aries. Even during his “honeymoon,” Obama’s approval rating among Republi- cans, according to Gallup, never reached above 40 percent. After August of 2009, Obama’s approval rating among Republicans sank below 20 percent and has yet to reach above that marker. There does not seem to be any evidence that the Demo- crats have gained significant support among social conservatives, rural voters, and the voters in the “solid South.” It is inconceivable that Obama will compete in this coming election with the Republican nominee in states like Alabama, Missis- sippi, or South Carolina.

In other words, more so than a generation ago, Republican voters remain in the Republican camp, Democrats in the Democratic fold. Perhaps we will know more in a year’s time. However, the 2012 campaign begins with President Obama’s approval rating among Democrats near 80 percent, and his approval rating among Republicans in the single digits, according to Gallup.42 This is roughly the inverse of President George W. Bush’s approval rating among Republicans and Democrats on Election Day 2004. Given the likelihood of a closely contested 2012 election, it is logical to assume that the trend toward greater partisan voting will resume.


In this paper, I have concluded that partisan voting in presidential elections has continued to trend upward during the past three election cycles. Drawing on pre- vious research, I have attributed this trend to three developments in the timeline of history: (1) The movement of conservative voters—conservative relative to po- litical symbols—away from the Democratic Party in response to the Civil Rights Act of 1964 and President Lyndon Johnson’s Great Society, (2) the development of micro-targeting technologies, which have enabled political parties to speak to finite subgroups of voters, and (3) the emerging dominance of ideologically-biased new programming, particularly during the primetime television news hours and in the blogosphere. This trend has been confirmed and quantified by Princeton University’s Larry M. Bartels; I have demonstrated that partisan voting has con- tinued to increase well after Bartels completed his research.

I would strongly encourage future research into the interaction between voting behavior and media consumption as news networks continue to abandon objective news coverage in favor editorialized news programming, and as more vot- ers consume news from blogs and other ideologically-biased online news sources. It would be logical that party identification would continue to gain strength in influencing voting behavior as voters continue to turn to more partisan sources for news consumption. ◗


I. The rise of the Religious Right, as we will see, would not have buoyed the Republicans without the rise of micro-targeting strategies and technologies.

II. The survey asks, “Generally speaking, do you usually think of yourself as a Republican, a Democrat, an Independent, or what?” (If Republican or Democrat) “Would you call yourself a strong (Republican/Democrat) or a not very strong (Republican/Democrat)?” (If Independent) “Do you think of yourself as closer to the Republican or Democratic party?”

III. Three regular surveys can quantify trends in self-identification, no one more accurate than the other. Public opinion research firms have frequently asked respondents if they “identify” or “regard” themselves as a conservative or a liberal. The Gallup Organization and the Opinion Research Corporation surveyed voters from 1936 to 1964, asking voters after the reelections of Franklin Roosevelt and Dwight Eisenhower, and the election of Lyndon Johnson, “Should (the President’s) second Administration be more liberal, more conservative, or about the same as the first?” From 1945 to 1979, Gallup asked respondents: “Which of the three policies would you like to have (the President) follow: 1. Go more to the left, by following more of the views of labor and other liberal groups? 2. Go more to the right, by following the views of business and conservative groups? 3. Follow a policy half- way between the two?” This final question demonstrates how ideology relates to the political symbols of labor unions against business interests.

IV. While Obama’s soaring rhetoric of cross-cultural unity in his 2004 Democratic National Convention keynote address no doubt propelled his national figure into the history books, his own campaigns have also sent direct messages to closely identified communities of voters.

V. The measure of partisan voting in Figure 2 is calculated as Bartels did in Figure 1: by multiplying the coefficients by percentage of voters who identify themselves as a Republican, Independent, or Democrat. For example, the 2008 measure of partisan voting is calculated as follows: (1.731 * .32) + (1.169 * .28) + (1.054 * .29) = 1.1869.


1. Jules Witcover, Marathon: The Persuit of the Presidency, 1972-1976 (New York: Viking Press, 1976), Pg. 5.

2. Ibid, Pg. 9.

3. Ibid, Pg. 13.

4. The American National Election Studies, “The ANES Guide to Public Opinion and Electoral Behavior,” (Ann Arbor, MI: University of Michigan, Center for Political Studies, 1952-2008. <>

5. Paul R. Abramson, John H. Aldrich and David W. Rohde, Change and Continuity in the 2008 Elections (Washington: CQ Press, 2010), Pg. 73-75.

6. Bartels, Larry, “Partisanship and Voting Behavior,” American Journal of Political Science (Jan. 2000): 35-50.

7. Allen J. Matusow, The Vietnam war, the liberals, and the overthrow of LBJ in Exploring America’s Past: A Reader in Social, Political, and Cultural History, 1865-Present (Lanham, Maryland: University Press of America, 1996), Pg. 235.

8. The American National Election Studies, “The ANES Guide,” < http://www.> ,

9. Ibid, <>

10. Noah, Timothy, “What We Didn’t Overcome: Obama won a majority of votes. He didn’t win a majority of white votes,” Slate Magazine, November 10, 2008, < http:// overcome.html>.

11. Grubin, David and David McCullough, American Experience: LBJ, Corporation for Public Broadcasting, 1991, lbj/.

12. Ellis, Christopher, and James A. Stimson, “Symbolic ideology in the American Electorate,” Electoral Studies (2009): Pg. 398.

13. Ibid, Pg. 398.

14. Ibid, Pg. 398.

15. Ibid, Pg. 398.

16. Levendusky, The Partisan Sort, Pg. 24.

17. Obama, Barack H, “2004 Democratic National Convention Keynote Address” (Delivered at the Fleet Center, Boston Massachusetts on July 27, 2004), http://www.

18. Brady, John, Bad Boy: The Life and Politics of Lee Atwater (New York: Addison Wesley Publishing Company, 1997), Pg. 26.

19. Ibid, Pg. 77.

20. Ibid, Pg. 78,

21. David Greenberg, “Dog-Whistling Dixie: When Reagan said “states’ rights,” he was talking about race,” Slate Magazine, November 20, 2007,< articles/news_and_politics/history_lesson/2007/11/dogwhistling_dixie.html>.

22. Brady, John, Bad Boy: The Life and Politics of Lee Atwater (New York: Addison Wesley Publishing Company, 1997), Pg. 117.

23. The American National Election Studies, “The ANES Guide,” (This question—http://—is regressed against the vote to come up with this data)

24. Cannon, Carl, Lou Dubose and Jan Reed, Boy Genius: Karl Rove, the Brains Behind the Remarkable Political Triumph of George W. Bush (New York: Public Affairs, 2005), Pg. 19.

25. Ibid, Pg. 20.

26. Ibid, Pg. 21.

27. Sosnik, Douglas B. Matthew J. Dowd and Ron Fournier, Applebee’s America: How Successful Political, Business, and Religious Leaders Connect with the New American Community (New York: Simon and Schuster, 2006), Pg. 3.

28. Ibid, Pg. 24.

29. Ibid, Pg. 41.

30. Cooperman, Alan, and Thomas B. Edsall. “Evangelicals Say They Led the Charge for GOP.” The Washington Post, November 8, 2004, Pg. A01.

31. Ibid, Pg. A01.

32. D. Sunshine Hillygus and Todd Shields, The Persuadable Voter: Wedge Issues in Presidential Campaigns (Princeton, New Jersey: Princeton University Press, 2008), Pg. 3.

33. Markus Prior, “News vs. Entertainment: How Increasing Media Choice Widens Gaps in Political Knowledge and Turnout,” American Journal of Political Science (2005): 577-592.

34. “TV Ratings: ABC Polls Highest on Election Night – Zap2it.” TV Listings Guide, TV Ratings, TV News, TV Shows – Zap2it. November 5, 2008. (Accessed June 3, 2011),,0,615045.story.

35. Lawrence, Eric, John Sides, and Henry Farrell, “Self-Segregation or Deliberation? Blog Readership, Participation, and Polarization in American Politics,” Perspectives on American Politics (2010): 141-157.

36. Prior, Markus, Post-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections (Cambridge: Cambridge University Press, 2007), 822-825.

37. Data generated from ANES survey, 1952-2008. (Graph was generated by regressing final vote against identification – this data toptable/tab9a_1.htm—against this data toptable/tab2a_1.htm

38. Larry M. Bartels, “Partisanship and Voting Behavior, 1952-1996,” American Journal of Political Science (2000): 35-50.

39. CNN Election Night Exit Poll, November 4, 2008. Poll questions retrieved March 20, 2012,

40. Data generated from ANES survey, 1952-2008. (Same as with Citation 38)

41. Gallup Presidential Job Approval Rating Center. Poll questions retrieved March 20, 2012, aspx?ref=interactive.


Leave a Reply

About The Review:

The Cornell Policy Review is the official public policy journal of the Cornell Institute for Public Affairs, a graduate program offering a two-year Master's degree in Public Administration (MPA).