How the Presidential Candidates Use the Web and Social Media

by Project for Excellence in Journalism / Originally published on Aug. 15, 2012

Obama Leads but Neither Candidate Engages in Much Dialogue with Voters.


If presidential campaigns are in part contests over which candidate masters changing communications technology, Barack Obama on the eve of the conventions holds a substantial lead over challenger Mitt Romney.

A new study of how the campaigns are using digital tools to talk directly with voters-bypassing the filter of traditional media-finds that the Obama campaign posted nearly four times as much content as the Romney campaign and was active on nearly twice as many platforms. [1] Obama’s digital content also engendered more response from the public-twice the number of shares, views and comments of his posts.

Just as John McCain’s campaign did four years ago, Romney’s campaign has taken steps over the summer to close the digital gap-and now with the announcement of the Romney-Ryan ticket made via the Romney campaign app may take more. The Obama campaign, in turn, has tried to adapt by recently redesigning its website.

These are among the findings of a detailed study of the websites of the two campaigns as well as their postings on Facebook, Twitter and YouTube-and the public reaction to that content-conducted by the Pew Research Center’s Project for Excellence in Journalism.

In theory, digital technology allows leaders to engage in a new level of “conversation” with voters, transforming campaigning into something more dynamic, more of a dialogue, than it was in the 20th century. For the most part, however, the presidential candidates are using their direct messaging mainly as a way to push their messages out.  Citizen content was only minimally present on Romney’s digital channels. The Obama campaign made more substantial use of citizen voices-but only in one area: the “news blog” on its website where that content could be completely controlled.

The study of the direct messaging of the candidates also reveals something about the arguments the two sides are using to win voters. Romney’s campaign was twice as likely to talk about Obama (about a third of his content) as the president was to talk about his challenger (14% of his content). That began to change some in late July when the Obama campaign revamped its website.

And while the troubled economy was the No. 1 issue in both candidates’ digital messaging, the two camps talk about that issue in distinctly different ways. Romney’s discussion focuses on jobs. Obama’s discussion of the economy is partly philosophical, a discourse on the importance of the middle class and competing visions for the future.

This is the fourth presidential election cycle in which the Project for Excellence in Journalism has analyzed digital campaign communications. This year, in addition to the campaign websites, PEJ broadened its analysis to include an in-depth examination of content posted on Facebook, Twitter and YouTube, areas that were either in their infancy or that candidates made no use of four years ago. The study encompassed an examination of the direct messaging from the campaigns for 14 days during the summer, from June 4 to June 17, 2012, a period in which the two campaigns together published a total of 782 posts. The study also included audits of the candidates’ websites in June and again in late July.

The changes from 2008 go beyond the candidates adding social media channels. The Obama campaign has also localized its digital messaging significantly, adding state-by-state content pages filled with local information. It has also largely eliminated a role for the mainstream press. Four years ago the Obama campaign used press clips to validate his candidacy. The website no longer features a “news” section with recent media reports. Now the only news of the day comes directly from the Obama campaign itself. (In the recent redesign, the Obama campaign also highlighted its “Truth Team” section which includes its criticism of the Romney economic plan as well as their accounting of Obama’s initiatives-also as determined by the Obama campaign.)

The Romney website, by contrast, contains a page dedicated to accounts about the candidate from the mainstream news media, albeit only those speaking positively of Romney or negatively of Obama.

 Among the study’s findings:

  • Obama’s campaign has made far more use of direct digital messaging than Romney’s. Across platforms, the Obama campaign published 614 posts during the two weeks examined compared with 168 for Romney. The gap was the greatest on Twitter, where the Romney campaign averaged just one tweet per day versus 29 for the Obama campaign (17 per day on @BarackObama, the Twitter Account associated with his presidency, and 12 on @Obama2012, the one associated with his campaign). [2] Obama also produced about twice as many blog posts on his website as did Romney and more than twice as many YouTube videos.
  • The campaign is about the economy, but what that means differs depending on to whom one is listening. Roughly a quarter, 24%, of the content from the Romney campaign was about the economy versus 19% of Obama campaign posts. But Romney devoted nearly twice the attention as Obama to jobs. Obama’s attention to the economy was almost equally divided between jobs and broader economic policy issues such as the need to invest in the middle class and how the election presents a choice between two economic visions. Another striking finding in the topics of the digital conversation is how much the agenda has changed in just four years. Gone from four years ago are web pages focused on veterans, agriculture, ethics, Iraq and technology. New are pages about tax policy-and the two campaigns overlap on fewer issues than Obama and McCain did.
  • The economy may have dominated both candidates’ digital messaging, but it was not what voters showed the most interest in. On average Obama’s messages about the economy generated 361 shares or retweets per post. His posts about immigration, by comparison, generated more than four times that reaction; and his posts about women’s and veterans’ issues generated more than three times. This was also true of attention to Romney’s messaging. His posts on health care and veterans averaged almost twice the response per post of his economic messages.
  • Neither campaign made much use of the social aspect of social media. Rarely did either candidate reply to, comment on, or “retweet” something from a citizen-or anyone else outside the campaign. On Twitter, 3% of the 404 Obama campaign tweets studied during the June period were retweets of citizen posts. Romney’s campaign produced just a single retweet during these two weeks-repeating something from his son Josh.
  • Campaign websites remain the central hub of digital political messaging. Even if someone starts on a campaign’s social network page, they often end up back on the main website-to donate money, to join a community, to volunteer or to read anything of length. A July redesign of the Obama page emphasized the centrality of the campaign website further. Rather than sending users to the campaign’s YouTube channel, the video link now embeds the campaign videos directly into the website, where the only videos are the ones Obama wants you to see.

Obama’s digital strategy targets specific voter groups-as it did four years ago-to a greater degree than Romney’s.

Visitors to Obama’s website are offered opportunities to join 18 different constituency groups, among them African-Americans, women, LGBT, Latinos, veterans/military families or young Americans. If you click to join a group, you then begin to receive content targeted to that constituency. The Romney campaign offered no such groups in June. It has since added a Communities page that by early August featured nine groups.

How important is digital campaigning: does more digital activity really translate into more votes?

In 2004, Howard Dean used the web to generate early support and fundraising, but he failed to convert that into caucus or primary turnout. Barack Obama more successfully converted his use of the web in 2008 to stage an insurgent campaign and win younger voters.

But some may question whether younger voters were attracted to Obama because of his digital activity or whether Obama used digital platforms because it was a logical way to reach a natural voter base.

While there may be no simple answer, throughout modern campaign history successful candidates have tended to outpace their competitors in understanding changing communications. From Franklin Roosevelt’s use of radio, to John F. Kennedy’s embrace of television, to Ronald Reagan’s recognition of the potential for arranging the look and feel of campaign events in the age of satellites and video tape, candidates quicker to grasp the power of new technology have used that to convey a sense that they represented a new generation of leadership more in touch with where the country was heading.

PEJ began studying the role of digital technology in presidential politics in 2000. In our first report, “ePolitics,” candidate websites were yet to emerge; news websites and “web portals” were the gatekeepers of digital campaign information. PEJ that year studied 12 of the most popular sites and portals providing campaign news, a list that included Salon, the Washington Post and Netscape. The study found an emphasis on updating tidbits of information throughout the day, so much so that sometimes the most important event of the day-or week-never became headline news. On February 28, 2000, for instance, AOL never led with John McCain’s speech in Virginia Beach attacking Pat Robertson and Jerry Falwell, even though it was not only the story of the day but a critical event of his campaign.

In 2004, PEJ re-examined the sites still in existence (and added two others); websites that year made a significant push toward offering users a chance to compare candidates on the issues-something almost entirely absent in 2000. News websites were also beginning to provide opportunities for users to manipulate and customize information; navigation, however, was often difficult. It was also the election cycle in which candidate Howard Dean transformed political campaigns by becoming the first candidate to use blogging, to use his website to organize “meetups,” and to use other internet technology as a major part of his campaign.

By 2008, candidate websites were standard, and campaigns were clearly taking steps to try to control their message in ways that bypassed the traditional media. This was the year that Hillary Clinton announced her candidacy on her web page; and Barack Obama, albeit not entirely successfully, announced Joe Biden as his running mate on his website. Obama also used his site widely to invigorate a national grass roots campaign and built substantially on Dean’s 2004 efforts to raise millions in small donations using the web. Our analysis also found that different candidates’ campaigns differed widely in how well each had mastered technology.

In 2012, in short, voters are playing an increasingly large role in helping to communicate campaign messages, while the role of the traditional news media as an authority or validator has only lessened.

Obama’s digital effort stands out first for its scale on various platforms compared with Romney’s.

At the time of our analysis (June 4-17, 2012), the Obama campaign had public accounts on nine separate platforms: Facebook, Google+, Pinterest, Tumblr, YouTube, Flickr, Instagram, Spotify and two accounts on Twitter (@BarackObama and @Obama2012).

That is twice that of the Romney campaign, which had public accounts on Facebook, Twitter, YouTube, Flickr and Google+. Romney has since expanded his presence, adding accounts on Tumblr and Spotify.

The Obama campaign is also substantially more active in these domains. Across all the platforms studied, the Obama campaign posted nearly four times as much content as the Romney campaign-614 Obama posts compared with 168 posts for Romney.

The gap in activity was greatest on Twitter. Romney averaged one tweet a day. Obama averaged 29 tweets a day, (17 per day on @BarackObama, the Twitter Account associated with his presidency, and 12 on @Obama2012). The Obama campaign also had about twice as many blog posts on its website than did Romney’s and more than twice as many YouTube videos. The two campaigns were closer in the level of activity on Facebook, where both candidates averaged about two posts a day (and Romney was slightly more active).

Overall, the Romney campaign was most active on Facebook and blogs, while Obama was most active on his two twitter accounts, posting here even more frequently than on his website blog.

For both campaigns, YouTube had the fewest number of total posts (where the production time and resource investment may be greater). After posting a video, each campaign then promoted it widely, linking to it again and again across platforms. (See below for a larger discussion of each candidate’s use of YouTube and citizen response.)

Compared with 2008, the Obama campaign has made its digital messaging more targeted and it has reduced the role of the traditional press.

One clear area of expansion is in how active the campaigns are in posting new content in the digital space, particularly on social media channels. Much of that is tied to general trends in digital communications. Obama now has more than 27.6 million Facebook friends, 207,000 YouTube subscribers and over 18 million Twitter followers, compared with 1.7 million Facebook followers and 83,000 YouTube subscribers in 2008. (Twitter was yet to emerge as a major platform.)

But there are other changes in Obama’s digital messaging that go beyond an embrace of social media-changes in the way his campaign has organized its website. One major addition is in how customized information now is. The Obama campaign now allows users to tailor content by state. Every state has a dedicated page filled with state-level news and information. Users can sign up to join any state group for localized updates, as well as the most up-to-date blog content specific to that state.  (Since the time of the detailed content examination in June, the structure of the site has changed, and the state level information is accessed differently.)

The site also offers 18 different constituency groups that users can join-groups very similar to those the campaign offered in 2008.

Another change is that the range of issues Obama is talking about has substantially narrowed. Four years ago Obama’s website featured different pages delineating the candidate’s position on 22 different issues. This election cycle, the campaign offers pages on eight-and foreign policy is not among them.

The Romney campaign this year has pages delineating his positions on 24 separate issues-eight of them tied to foreign policy and seven of them tied to economic issues. But, here, too, there is evidence of a different emphasis from 2008. McCain offered 16 different issue pages but only seven overlap with those on the Romney site.

The change in the list of issues being debated in 2012 versus 2008 offers a hint of how much the campaigns believe that the concerns of the American voters have been altered by the end of the war in Iraq and the effects of the recession. In 2008, among other issues now off the digital agenda, McCain talked about Iraq, agricultural policy and space. Romney does not address these but talks about Iran, China, Israel, trade and fiscal spending-all new. Obama four years ago had pages focused on about three times as many issues as this year. But taxes and women’s health are there in 2012. They were not subjects that had dedicated pages in 2008.

Also gone in 2012 from Obama’s website is a channel dedicated to featuring how he has been described by the news media, featuring clips and links to external news reports. Instead the information offered all comes from the campaign in the form of images or words produced internally or selected content from citizens.

Both candidates have focused mainly on promoting themselves and their campaign, but Romney was more than twice as likely to focus on Obama as the other way around.

Across all platforms studied, 55% of the posts from the Obama campaign focused on promoting his record and accomplishments. Similarly, 52% of the posts from the Romney campaign focused on its own candidate.

But Romney also devoted substantial space to discussing Obama. In June, roughly a third of posts from the Romney campaign (34%) were about Obama-largely attacking him for a policy stance or action. That was twice that of the Obama campaign (14% of which focused on Romney).  That difference held true across all platforms studied, except for Facebook where both Romney and Obama devoted the majority of posts to themselves (65% for Romney and 74% for Obama).  In late July, however, Obama began to focus more on attacking Romney, particularly in his “Get The Facts” section of the website.

An examination of the format of the content published also finds a difference in the way the candidates communicated. Romney tended to rely more on visuals-graphics, photos or videos. Obama tended to lean more heavily on text.


Still neither campaign has devoted much content to anyone other than the candidates themselves or their opponent. The Obama camp found a few occasions to post about Congress (4%), or Republicans in general (2%), and there was just one post (on Twitter) that brought Joe Biden into the spotlight. But nearly a quarter (23%) did not focus on any individual. That was the case for 11% of Romney’s posts.

The economy was the No. 1 issue for both campaigns in their digital messaging.

Another way of measuring the focus of the campaigns was to track the posts they made in social media and on their websites. For the two weeks studied in June, researchers tracked all of the posts on Twitter, Facebook, YouTube and on the candidate’s website blogs and homepages. The economy was the biggest subject. Nearly a quarter (24%) of Romney posts and 19% of Obama’s focused on the economy.

Romney’s messaging about the economy was distinctly different than Obama’s.  While Romney focused on the overall issue of unemployment and/or jobs more often than any other single aspect of the economy, he used the jobs issue primarily to criticize President Obama. Roughly three-quarters of Romney’s posts that touched on jobs were either disparaging of Obama’s comment that “the private sector is fine,” or described him as hostile to job-creators and out of touch with the middle class. Only one-quarter of Romney’s posts on the economy either focused directly on his own plans or contrasted his plans with Obama’s.

In these digital posts, Obama’s campaign also covered a broader range of other domestic issues besides the economy.  Of a total of 24 possible domestic concerns, the Obama campaign posted content on 14.  The Romney campaign posted on nine, and in those discussions often circled back to the economy. When addressing Hispanic concerns, for example, the overall focus remained on jobs and unemployment and the campaign argued that the president had not done enough to help Hispanics. On June 6, Romney’s campaign posted three separate website blog entries to this effect, one of them by Marco Rubio, who wrote “Unfortunately, President Obama’s failed policies of new regulations, higher taxes, and Obamacare and his anti-business rhetoric have hit Hispanics especially hard. Big government really hurts those who are trying to make it.  And with unemployment still abysmally high, the Obama economy is crushing Hispanics’ dreams for their children to live a better life. The Hispanic community cannot afford four more years of double-digit unemployment and higher levels of poverty.”

The Romney campaign did not touch on five of the 14 issues addressed by the Obama campaign. Those included immigration, education, public employees, gay rights and women and equal pay.  Here the Obama campaign tried to make a case for itself.

“Every GOP senator just voted against helping women get #EqualPay for equal work,” read one Obama campaign tweet.  “Celebrate Pride month by looking at the progress we’ve made over the past three years under President Obama,” read another.

While Obama’s messaging on his campaign website covered fewer issues than Romney, in other words, during the June period studied, his social media messaging was more expansive than Romney’s.


During the June period studied, both campaigns focused more on domestic issues than foreign policy, personal issues or even the political horserace, accounting for fully 50% of posts from the Obama campaign and 40% from the Romney camp.

Obama’s next biggest focus, accounting for 20% of all posts, was fundraising and volunteering. That came through strongest on the website blog, where posts tended to try to convey the spirit of supporters of Obama, such as young volunteers who became involved because of Obama’s policies on women or Hispanics, someone who started an Obama fundraising page the day the president announced his support for gay marriage, and an Obama campaign field director answering questions about what it was like to volunteer for the campaign.

The Romney campaign’s biggest focus, after domestic issues, was Romney campaign activities, which accounted for 28% of posts. Multiple blog posts from the Romney campaign focused on the candidate’s “Every Town Counts” bus tour and included embedded Storify posts featuring photos and videos from various campaign stops.

Neither candidate talked much about either candidate’s public record. Romney’s team gave it a little more emphasis-but on Obama’s record not his own.

Foreign policy issues were all but absent-a sharp contrast from four years ago. During the two weeks studied, there was one post from Obama about foreign policy and not a single one from Romney.   (As noted, Romney has several tabs on his homepage that take users to his policy stances on a variety of foreign countries and Obama offers a page on various National Security issues. But these issues were not a part of what the campaigns discussed on their digital platforms over the two weeks studied.)

The study also measured the extent to which persons other than the two candidates were prominently featured in a photo or video to see which groups Obama was attempting to appeal to. The Obama campaign was somewhat more likely to include these images of these target groups than the Romney campaign was and in doing so displayed a wider range of constituencies. About half (49%) of Obama’s posts on Facebook, YouTube and the news blog featured citizen images; 40% of Romney’s did. (Twitter posts were not included in this count.)

The group most often featured by Obama was young people, followed by celebrities, women and the first lady-all speaking positively of Obama.

For Romney, the group shown more than any other was Hispanics, mainly voicing criticism of Obama’s handling of the economy.

The digital communications of the campaigns often did not invite citizens to delve deeper.

Only about half of the posts studied here contained links of any kind, 44% for Romney and 51% for Obama.

When there was a link, the vast majority of the time it took users to another part of the campaign’s controlled communications rather than to some independent or verifying source (71% of Obama links and 76% of Romney links). This was the case for both candidates for every single link in a Facebook or YouTube post.

Fewer than 18% of the most prominent links for each candidate went to an outside entity. Links from the Obama campaign went to a slightly broader mix of places than those within Romney campaign posts. But still, the vast majority of Obama campaign links went to campaign or White House outlets.

In other words, for the most part, the digital communications of the campaigns were a closed loop, not a way to access the depth of the web.

One interesting element was how rarely the news media were a source of information or validation for what the campaigns wanted to argue. Just 5% of links for each campaign went to a mainstream news story. Obama also linked a handful of times to non-traditional news entities like ThinkProgress, MaddowBlog or the Huffington Post. One example from the Obama campaign came in a reoccurring blog feature “Did You See?” which lists a number of recent news items.  This one shown below was posted by one of his campaign staffers.

For Romney, the mainstream news media more often than not were a source to criticize the president. Most of the embedded links to news reports from the Romney campaign were to articles critical of Obama, such as the June 14 piece by Congressman Gus Bilirakis (R-FL) that appeared in the Tampa Tribune taking Obama to task for saying that “the private sector is doing fine.”

No links posted during these two weeks went to citizen-produced content.

One distinguishing feature of the web is the opportunity it creates to engage in a direct dialogue with citizens. This can be done several different ways. For this analysis we examined the extent to which each campaign addressed specific voting groups or segments of the population, the extent to which each asked users to take action (and what they asked them to do), and the extent to which citizen response was visible (or incorporated) on the site.

By all three measures, the Obama campaign’s engagement with citizens exceeded that of the Romney campaign. For the most part, however, this was limited to the website, where such engagement was more carefully controlled. Neither campaign created much public dialogue with citizens in their social media channels.

Tailored content:

One of the biggest changes in the Obama website from 2008 to 2012 was the ability to tailor content and news feeds to one’s location.

Geography was one prominent way the Obama campaign allowed users to customize their digital interaction,  but the campaign also offered opportunities to join 18 different constituency groups-such as African-Americans, women, LGBT, Latinos, veterans/military families or young Americans-and receive content targeted to that constituency.

In June, the Romney campaign offered no such targeting by groups. It has since added this feature, though in a different manner. In mid-July the campaign added six voter group pages that users can choose from (they have since added three more). But users do not “join” a group and receive content thereafter. Rather, users are taken to a page with dedicated content. That content, so far, is often not updated very frequently.

This marks a shift from the approach McCain used in 2008. McCain offered 18 target voter groups in 2008 (and Obama 20). Obama’s list of groups is also different than it was four years earlier. Parents are now an Obama target group, for instance. So are small business owners and educators.

Incorporating Citizen Content

Neither campaign engaged heavily in the “social” aspect of the social media-But the Obama campaign filled its news blog with citizen content.

Nearly all of the tweets, posts on Facebook and YouTube videos originated with someone inside the campaign or a well-known supporter. Rarely did either candidate reply to, comment on or retweet something from a citizen.

On Twitter, for example, 16% of Obama’s 404 total tweets were retweets. And just 3% of all tweets (14) were retweets of citizen posts. The most prominent example of this was on June 14, when David Axelrod hosted a live Twitter Q & A consisting of 28 total tweets, 10 of which were re-tweeted questions from citizens, immediately following Obama’s speech on the economy in Ohio. 

Romney produced just a single retweet during these two weeks and it came from his son Josh, who was passing along a photo of a climber holding a Romney campaign banner on Mt. Everest.

Nor did the campaigns use two other social media platforms to project citizen content. On Facebook and  YouTube every post included content directly from the campaign itself.

There was one area where the Obama campaign heavily employed citizen content, but it was one where that engagement was fully controlled: the campaign’s news blog. It is a channel on their website where all content is posted by campaign staffers. Here the Obama team gave high priority to citizen voices.  Four-in-ten posts (42%) were written (or taken, in the case of photos) by citizens. Many of the others were written by staff but included quotes from citizens.

Many of the citizen contributions in the Obama blog spoke of how and why they became involved in the campaign or shared personal stories of how a particular policy of President Obama had changed their life.

Others depicted ways the individual was working to garner more support for Obama, by training to become an organizing fellow, offering canvassing tips or providing advice on how to register voters.

As in the Obama news blog, posts appearing in Mitt Romney’s blog were authored by the candidate himself, members of Team Romney or prominent GOP supporters. State-level posts sometimes included photos from campaign stops but more often consisted of quotes from the candidate, state GOP representatives and a list of endorsements.

The Romney campaign generally chose not to use its news blog as a way of conveying citizen input.

Only two of the Romney blog posts between June 4-17 were written by citizens. In both cases, the author was the chairman or CEO of a business complaining that President Obama failed to understand business and free-market principles.

Initiating Activity

The next tier of interplay with citizens is asking them to take action themselves, to help campaign for a candidate. Here both campaigns engaged citizens more, though to varying degrees depending on the platform and with different emphases on what action the campaign asked for.

Overall about half of each candidate’s posts included a request for some kind of voter follow-up activity. These calls to action were most common on the website blog posts. Every single blog post from the Obama campaign during the time studied included some call to action, as did 91% of his YouTube posts. Most, 81%, of Romney’s homepage content and 40% of his YouTube video posts had calls to action as well.  Twitter was the platform least likely to contain a call to action.

Buttons for sharing posts through any one of the many social media platforms were standard across both websites-but it was not always the first or primary response requested.

For Obama, the primary call to action most often (51% of the time) was a request for some kind of digital-oriented response, such as watch this video, join this list or sign up to be part of a “team.” For Romney the request that appeared first most often (31% of the time,), was to donate money. These tended to appear in the form of a donate button.

Social media sharing was the next most popular request. These were the first action requests in 16% of Obama’s posts and 20% of Romney’s. Two elements to rarely appear as the primary request were to vote or to send feedback.

What does all of this add up to in terms of generating a citizen following? One measure is the number of people following any one channel. Again, Obama has the built-in advantage of having begun these connections during the 2008 campaign. But the Obama team has added substantially to this in four years. His YouTube subscribers have more than doubled, and his Facebook supporters are about 16 times what they were in September 2008.

Overall, looking at all the measures from “followers” to “views” to “likes,” Obama’s numbers surpass Romney’s by a margin of at least 13:1. That includes Twitter, which was not in our study in 2008-and the Romney numbers are in question. In late July, Mitt Romney’s twitter feed suddenly reported a massive spike in followers-adding 141,000 in just two days time, but research into those followers finds that they were mechanically generated rather than real individuals.

Another measure of response or engagement is the number of people who have shared or liked content posted by the campaign. Researchers recorded the number of likes, comments, dislikes (where appropriate), retweets and views for every piece of content up to 48 hours after it was initially posted. Facebook by far generated the most attention for both candidates. Obama posts generated in total more than 1,100,000 likes during this period. Romney posts generated about half that, just about 635,000 likes in total.

The difference in response to Obama’s versus Romney’s social media content was more striking in the number of retweets. Obama’s tweets were retweeted more than 150,000 times during these two weeks. Romney’s tweets were retweeted just under 8,600 times.

On YouTube, Obama videos generated more than 800,000 comments, likes, dislikes and views. Romney’s videos generated about half that attention, just under 400,000 responses.

These differences are not simply a reflection of Obama’s campaign posting content more often than Romney. If one looks at the average response to the campaign posts on each platform, Obama also has a substantial advantage. For instance, Obama’s YouTube videos averaged 466 likes per video versus for 253 for Romney’s.  People commented on his Facebook content an average of 2938 times per post versus 1,941 for Romney’s. [3]


One of the most striking new features of the race for president four years ago was the emergence of YouTube. This became a venue for Obama to post longer videos than conventional political advertising. The Obama campaign later reported that it had a billion viewings of videos on its YouTube channel by the time the race was over.

Four years later, YouTube is the platform with the fewest posts from the candidates of any of those studied. But every video posted was promoted widely across their various platforms. During the time period studied in June, no video became hugely popular or could be considered to have “gone viral.”  In fact, while both campaigns averaged roughly 40,000 views within 48 hours of posting, just a third of each candidate’s videos actually surpassed that viewership. The video that garnered the greatest number of views by far was not exactly campaign-related. It was the first lady’s Father’s Day card to her spouse.   

Finally, the study also delved deeper into the realm of engagement to see which topics generated the most attention from each candidate’s constituents.

While the economy was the biggest topic in terms of content posted, it was not the one that generated the most response. Obama’s economic posts generated an average of 361 shares or retweets per post. Attention was far greater to posts about public servants (892 shares on average per post), veterans (1,005), women’s issues (1,356) and immigration (1,600).

Although Romney touched on fewer non-economic issues, this was also true of attention to his content, with posts on health care and veterans averaging 2,162 and 2,147 shares and retweets, respectively, compared with 1,186 shares and retweets of content on the economy.

While it’s important to not draw any overarching conclusions from these numbers, given that the number of economic-focused posts far outweighed all others for both candidates, they do suggest that content particularly relevant to specific constituencies, e.g. women’s equal pay or immigration policy for Hispanics, may inspire the most social media reaction.

But the biggest single video across all platforms during the time period studied was that of Michelle Obama, Malia and Sasha wishing the President a Happy Father’s Day. That video was shared through Facebook 2,265 times in the first 48 hours and viewed 211,663 times.


1. For the time period studied (June 4, 2012 to June 17, 2012), this report considered those platforms that the candidates’ respective websites listed and linked to.

2. This report studied the candidate’s twitter accounts that the two campaigns linked to from their website homepage. There is another verified Romney campaign Twitter account, @TeamRomney, but it is not listed or linked to from, and thus was not formally associated with the campaign in the same way as other content streams.  Nor did the main Twitter account, @MittRomney, retweet or link to posts from @TeamRomney, at  the time of the study. Researchers did count the tweets from @TeamRomney during the time of the study, June 4 – June 17, 2012. There were a total of 63 tweets which would bring the average activity across the two Romney accounts from one to 6 per day.

3.One other metric specific to Facebook is the People Talking About This (PTAT) score. Produced by Facebook, PTAT measures the “conversation” around a specific Facebook page by tracking the number of different people who have interacted with that page through page likes, post shares, events RSVPs or other actions. PTAT’s are measured over a seven-day range. During the time period studied in this report, Obama’s PTAT score was also far higher than that of Romney’s. For time period ending June 6, 2012, for example, Obama’s PTAT score was four times higher than Romney’s. In the first half of August, the gap between the two candidates’ PTAT scores closed somewhat, though Obama still enjoying a higher score than Romney.