With the election less than two weeks away, Americans are following the presidential campaign more closely on nearly every news platform than they were earlier in the year, including print newspapers. The biggest gains have come on the internet-both to the websites of traditional news sources and those native to the web.

Social media platforms such as Facebook, Twitter and YouTube are growing especially rapidly as a source of political news. The number of Americans who say they regularly go to these destinations to learn about the campaign has doubled since January. Even with that jump, however, these leading social media platforms are still turned to by a relatively limited number of Americans, about 17% in all, when those who mentioned at least one of those platforms are combined.

When asked which sources of campaign news had been "most useful," nearly half of Americans named television in one of its various forms. Cable news was first on that list, named as the most useful source by 24%; a little more than a quarter volunteered various forms of the internet, while a third as many named local or national newspapers (8%) or radio (6%).

These are among the findings of a new survey of Americans about how they are learning about the election conducted October 18-21 among 1,005 adults nationwide by the Pew Research Center's Project for Excellence in Journalism.

The numbers portray a diverse landscape in which no platform dominates as the place for politics, and the vast majority of Americans say they regularly rely on multiple platforms to get political information. Just 6% said they turn regularly to just one platform.

Cable news channels continue to have the furthest reach, but a number of other destinations are close. Currently, 41% of Americans say they regularly learn about the candidates or the campaign from cable news networks, up five percentage points from 36% during the primaries.

But local TV news is almost as popular as a means for learning about the campaign; 38% of Americans regularly use it to learn about the candidates and the election, up six points since the primaries.

That is now nearly matched by the internet, which has seen an increase of 11 points in the number of Americans who say they regularly turn to it for campaign news since the year began. Fully 36% of Americans say they regularly get election news there, up from 25% in January.

Yet nearly as many-31% of Americans-regularly get information about the candidates and the campaign from national evening network newscasts; it was 26% during the primaries.

Local and national newspapers have also seen their audiences grow. Altogether, 30% say they regularly read one of these two types of newspapers for campaign news; 23% regularly turn to their local daily newspapers, up from 20% in January; 13% turn to national newspapers, up from 8% in January.

The Pew Research Center traditionally asks Americans about their sources of campaign news at the beginning of each presidential election. Back in January, those surveys found that with contested primaries in just one party, the long-term declines seen in several traditional sources such as newspapers, local TV and network news had steepened.

The new survey was conducted to examine media consumption behavior about the campaign during the general election phase, both to benchmark that time period for the first time, and to see whether there were patterns in a changing media landscape.

Beyond traditional news sources and digital destinations, other forms of media have also established a role in the political information ecosystem, though some of these so-called alternative forms may have stabilized or even shrunk in popularity.

About one-in-ten Americans, 12%, regularly get news from comedy programs such as The Daily Show, Saturday Night Live or The Tonight Show with Jay Leno. Up from 9% in January, it is now on par with public radio and national newspapers.

Cable talk programs also play a sizable part in the political dialogue. Fully 18% said they regularly got political information there, up from 15% in January. But there is also evidence that role of cable talk might be smaller than it was some years ago. In January 2004, for instance, a combined total of 44% of Americans said they regularly or sometimes watched the shows for campaign information. In January, that number was 34%. Now it is 35%. Meanwhile, the percentage of Americans who said they never watch cable talk shows has risen from 38% in 2004 to 47% today. This possible downward trend in the broad audience for cable talk, moreover, stands in contrast to the trend for cable news in general, which has been among the most stable in audience reach of the older news platforms.

The number of Americans who say they turn to radio talk for campaign information also changed little from January and has declined from several years ago. In all, 16% said they regularly turn to figures such as Rush Limbaugh and Sean Hannity on the radio for election information, unchanged since the primaries. Another 19% do so sometimes, also unchanged from January. But in January 2004, 46% said they regularly or sometimes got campaign information from talk radio. The total now comes to 35%.

Where People Go Online for Election News

The survey then probed more deeply into where on the internet people go for campaign news on a regular basis.  More said they turn to the websites or apps of traditional news organizations than to online-only sites or apps (28% of Americans who are online versus 19%).

And while social media, which draws on many sources for the information, remains relatively small, it is growing rapidly as a means for getting political news. Currently, 12% of those online say they regularly use Facebook to get campaign news, more than double the 6% who said so in January. That number represents 21% of those who use social networks.

YouTube is a regular source for campaign news for 7% of Americans, also more than double the 3% who said so in January, when the campaign involved primaries just in the GOP.

Twitter has also doubled to 4%, but remains the smallest of the three main social media formats. But when those on Twitter are asked about whether they use the platform for campaign news, the numbers become much larger. Fully 25% of those on Twitter use it regularly for campaign news. In January, 17% of those on Twitter used it regularly for political information.

Which Sources Are Most Useful

With such a complex network of platforms and sources to choose from, the nuance of how and when people seek out different places for information about the election becomes much more difficult to understand. The concept of a primary source of news-a gatekeeper that provides most of what a voter might know-seems obsolete.

To get at some sense of value, the survey asked people to name the source, out of those they turned to regularly, which had been "most helpful" to them overall.

The answers showed the continuing power of television in general to let people see events, but also the increasing influence of convenience, breadth and depth of the web.

In all, 49% volunteered some form of television as their most helpful source in providing campaign news, combining local, network and cable together. Cable led the way among these, with fully 26% of Americans naming cable news in general (24%) or cable talk shows in particular (2%). The other two major television sources-network nightly newscasts and local TV news-were both mentioned as most helpful by 11%.

About half as many as named a television source mentioned an internet one (28%).  That puts the internet in a tie with cable news as the most helpful medium for campaign news.

Most of these people volunteered the internet in general as the medium they consider most helpful, but some specified particular destinations; 4% mentioned Facebook, 3% the websites of traditional news organizations, 2% web-only sources, and 1% YouTube videos or Twitter.

Newspapers were named by just 8% of Americans as the most helpful source for learning about the candidates and the election (5% said local papers and 3% said national ones).

And 4% mentioned public radio as the most helpful source for learning about the campaign.

Comedy programs were tied with Twitter and YouTube at 1%.

About the Survey

The analysis in this report is based on telephone interviews conducted October 18-21, 2012, among a national sample of 1,005 adults 18 years of age or older living in the continental United States (601 respondents were interviewed on a landline telephone, and 404 were interviewed on a cell phone, including 224 who had no landline telephone). The survey was conducted by interviewers at Princeton Data Source under the direction of Princeton Survey Research Associates International. A combination of landline and cell phone random digit dial samples were used; both samples were provided by Survey Sampling International. Interviews were conducted in English. Respondents in the landline sample were selected by randomly asking for the youngest adult male or female who is now at home. Interviews in the cell sample were conducted with the person who answered the phone, if that person was an adult 18 years of age or older. For detailed information about our survey methodology, see:  http://people-press.org/methodology/.

The combined landline and cell phone sample are weighted using an iterative technique that matches gender, age, education, race, Hispanic origin and region to parameters from the March 2011 Census Bureau's Current Population Survey and population density to parameters from the Decennial Census. The sample also is weighted to match current patterns of telephone status, based on extrapolations from the 2011 National Health Interview Survey. The weighting procedure also accounts for the fact that respondents with both landline and cell phones have a greater probability of being included in the combined sample and adjusts for household size among respondents with a landline phone. Sampling errors and statistical tests of significance take into account the effect of weighting. The following table shows the unweighted sample sizes and the error attributable to sampling that would be expected at the 95% level of confidence for different groups in the survey:

Sample sizes and sampling errors for other subgroups are available upon request.

In addition to sampling error, one should bear in mind that question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of opinion polls.

See the topline.

© Pew Research Center, 2012

As the candidates for president reintroduced themselves at their conventions and began the last phase of the campaign, they received markedly different treatment in social media than in the mainstream press, a new study finds.

The conversation on Twitter, blogs and Facebook about Mitt Romney and Barack Obama during this key period changed little with events-even during the two candidates' own nominating conventions. The conversation in all of these platforms was also consistently negative, according to the study by the Pew Research Center's Project for Excellence in Journalism.

In the mainstream media, by contrast, both Romney and Obama received a version of the traditional convention bounce, with coverage about them becoming more positive during the week of their party's nationally televised gathering.

The media portrait of this key month in the campaign is consistent with what PEJ has seen in social media throughout the campaign. Whether it would prove true in other campaigns cannot be known. But the differences raise a question about whether social media may make what Americans hear about politics more negative and may make it harder for political actors, particularly those trailing in the polls, to alter the media narrative. 

On Twitter, for instance, the percentage of negative discourse about Romney was identical during the two convention weeks-and made up the majority of assertions. The level of positive discourse, meanwhile, varied by just two points between the two weeks. On Facebook, the numbers barely budged. There was a more noticeable shift in tone during the conventions, though modest, on blogs.

During his convention, Obama received a modest improvement in the tone of the conversation about him on Twitter, and somewhat less on Facebook and blogs, but the tone in all three venues remained consistently negative.

In the mainstream press, by contrast, positive stories about Romney during his convention outnumbered negative ones by about two-to-one. Obama's positive stories outnumbered his negative ones, by a somewhat smaller margin, during the Democratic convention.

The study also finds that generally more of the discourse has been focused on Obama rather than on Romney. The President has been the subject of more Twitter traffic, more YouTube viewings, more blog and Facebook conversation and even more mainstream news coverage. The only time that changed noticeably-and was likely unwanted-was on Twitter the week that a video surfaced in which Romney derided the 47% of Americans who do not pay income tax.

These are among the results of one-month study of the tone of discourse across 52 mainstream media outlets, the full range of conversation on Twitter, a large sampling on Facebook and in blogs, as well as a measure of traffic on YouTube.

The study examined the tone and volume of news coverage and social media conversation from August 27 to September 23 across a wide range of media platforms. Human researchers studied 1,084 stories in 52 mainstream media outlets-from television, print, radio and news websites. And a combination of human coding combined with computer software from the company Crimson Hexagon analyzed more than 18 million posts on Twitter, 323,000 on Facebook, and 690,000 on blogs. In addition, researchers examined the traffic to key videos of the candidates on YouTube during the same period.

Twitter

The week of the GOP convention, from August 27-September 2, fully 59% of the Twitter assertions about Romney were negative while 17% were positive. The following week, when the Democrats held their convention, Romney's numbers were virtually the same, again 59% negative and 19% positive.

These figures for Twitter reinforce what PEJ has seen throughout the campaign season about this growing social media platform. For every single one of the 16 weeks studied since June 4, the tone of the conversation about Romney has been negative-by a differential of at least 34 points.

The settled tone of that dialogue is reflected in the fact that during the week of September 17-23, when the damaging 47% video surfaced, the numbers were nearly the same as the week of Romney's convention, when Republicans were putting out their most positive messages about the candidate. The week of the video, 62% of the Twitter conversation about Romney was negative, and 14% positive.

Obama did enjoy some bounce on Twitter from his convention, but that proved ephemeral. The week of the Democratic convention, from September 3-9, Obama could boast that the gap between positive assertions about him (31%) and negative assertions (42%) narrowed to 11 points, about half of what it had been the week before when he was being rhetorically pummeled at the GOP convention.

Since then the Twitter conversation about Obama has returned to levels typical of what PEJ has seen throughout the general election period, with negative assertions exceeding positive ones about the president by about 20 points.

And with the exception of the week of the Democratic convention, the tone of Twitter conversation about Obama has been more negative than positive each week since June by a differential of at least 19 points.

Blogs

The conversation about the candidates on blogs has also been highly negative, and here it is distinguished by how evenly split it was during the conventions. At the time of the GOP gathering, negative assertions about Romney in the blogs studied outstripped positive ones by 22 points (41% to 19%). That same week, the conversation about Obama tilted to the negative by 30 points (45% negative and 15% positive).

The week of the Democratic convention, the conversation was again harshly negative for both, with a similar easing toward the host party. The conversation about Obama was negative by 24 points (43% to 19%). And the conversation about Romney was negative by 30 points, (46% negative assertions to 16% positive).

Since then, the conversation among bloggers has become worse for Romney. The week after the Democratic convention, the negative to positive gap for Romney grew to 34 points (48% negative and 14% positive). The following week, it increased further, to 41 points (54% negative and 13% positive).

In the case of Obama, the conversation in the blogosphere has remained fairly static since the Democratic convention.

Facebook

Of the three main social media venues in which the tone of conversation can be assessed by PEJ, the treatment of the candidates has been the most unchanging on Facebook. Here the sample is more limited. All that can be analyzed is a random sampling of Facebook posts that users are willing to have seen publicly. Presumably, that is a conversation that is designed to persuade but may not reflect the full range of what people say to just their "friends" on Facebook.

What was in this space during the month was overwhelmingly negative and barely budged.  In each of the four weeks, negative assertions about Romney exceeded positive ones by 43 points-exactly.

There was a little more variation for Obama and a mildly positive trend. During the Republican convention, negative assertions about the president outstripped positive ones by 49 points. During the Democratic convention, that dropped just slightly, to a 42 point negative differential. The last week studied, September 17-23, that had narrowed to 33 points-21% percent positive and 54% negative.

The Volume of Attention on Social Media

Another way of gauging the social media response to the candidates during the convention period is to look at volume, or how much attention was focused on each candidate. Here, there is a clearer distinction between Obama and Romney. For most of the month, most of the attention, whether measured in conversation or in views on YouTube, was focused on the president and Democrats.

On Twitter, Obama was the focus of more than twice as many assertions during the week of the Democratic convention than Romney was during the Republican convention (4.92 million assertions versus 1.98 million). That pattern, more attention to Obama, has held throughout the general election period and all of 2012, even during the period of the GOP primaries.

The two exceptions came just recently. The week of the GOP convention, Romney was the subject of slightly more assertions on Twitter than Obama - 1,975,872 to 1,958,298. The only occasion when Romney was in a significantly larger number was the week of September 17-23 when the 47% video surfaced. That week, there were 2.3 million Twitter assertions about the GOP nominee versus 1.9 million about Obama.

This generally higher focus on Obama is also reflected in the volume on Facebook. During Romney's convention week, for instance, there was more conversation about Obama than Romney by about 18%. And during Obama's convention week, the conversation about the president outstripped that of Romney by about 3-to-1, 166,724 assertions compared to 51,762.

The same basic volume pattern prevailed on blogs. During the GOP convention week, the Obama conversation was more than 17% greater than for Romney. During the Democratic convention week, it was twice as big.

And unlike Twitter, even the emergence of the now famous 47% video did not move the focus of the conversation from Obama toward Romney on blogs and Facebook.

The higher interest in Obama and the Democrats in social media is also reflected in the viewership of videos since the two conventions. For instance, through September 21, 2012, Obama's acceptance speech on various YouTube channels has been viewed nearly five times as often as Romney's (4.9 million to 1.1 million). And contrary to what some observers might speculate, Obama's speech has also been viewed more than former President Bill Clinton's address to the nation, though that speech, in various forms, has been viewed on YouTube nearly four times as often as Romney (3.9 million times to Romney's 1.1 million). The same pattern can be seen in the numbers as they relate to the wives of the candidates. Michelle Obama's speech has been viewed 3.2 million times, about five times as often as the one delivered by Ann Romney (563,000).

The only major Republican figure to generate more attention than his Democratic counterpart on YouTube was vice-presidential nominee Paul Ryan, whose speech has attracted 457,000 views compared with 173,000 for Vice President Joseph Biden, whose speech was not delivered in prime time.

The only other speaker at the Republican convention to generate major YouTube attention was one that Romney might like to forget. That was actor and director Clint Eastwood's "empty chair" speech. Since Eastwood's appearance, that video has been viewed 3.2 million times.

The only GOP video since the convention to be viewed more often than the Eastwood clip was footage of Romney's 47% remarks to fundraisers, which as of September 24, had been viewed 3.4 million times. That is still less, however, than the president's acceptance speech or Clinton's remarks.

Mainstream Media

The one media platform where the tone of the discourse changed markedly during the last month, and where a candidate managed to generate more positive than negative treatment, however briefly, was in the mainstream news media.

Here, in a sample that also includes cable and talk radio hosts, Romney fared somewhat better during his convention week than Obama during his.

The week of the GOP convention, 36% of the stories about Romney studied in the mainstream media outlets was positive compared with 15% negative-a margin of 21 points. The week of the Democratic gathering, 32% of the stories about Obama were positive compared with 22% negative-a gap of 10 points.

Since then, Obama's coverage has turned somewhat negative, but is still far better than Romney's. In the week following the conclusion of the conventions, September 10-16, 20% of the stories about Obama have been positive compared to 24% negative.

For Romney, the majority of stories (53%) that week were negative. Strikingly, of the 130 stories about Romney examined from the mainstream press that week, researchers found none in which positive assertions about Romney outnumbered negative ones by a ratio of 3-2, the threshold used to determine a story as having a clear tone. But 47% of the stories that week were mixed in tone, meaning that the assertions about Romney were fairly evenly divided.

The mainstream press has also given more attention to Obama during this period, even with the negative publicity associated with Romney's video. From August 27 to September 16, Obama was a bigger newsmaker than Romney, the focus of 667 stories compared with 477 for his rival. [1] And while the number of stories about Romney exceeded those about Obama by more than 30% during the Republican convention, Obama was the focus on more than twice as many stories as Romney during the Democratic convention.

---

Footnote

[1] Unlike the social media sample, which included the four weeks from August 27-September 23, the mainstream media sample includes three weeks, from August 27-September 16. This reflects the fact that the human coding of mainstream media outlets takes more time than the computer algorithmic coding of social media.

  

Summary of Findings

In January, 2011 the Pew Research Center's Project for Excellence in Journalism and Internet & American Life Project, in partnership with the Knight Foundation, conducted a nationally representative telephone survey of U.S. adults exploring local news consumption habits. Overall, the survey indicated that most adults follow what is happening in their local communities and that the local news ecosystem is complex.  Rather than relying on one or two main sources of local news, most adults use a wide variety of both traditional and online sources depending on which local topic they are seeking information about.[1] 

This report reexamines those data with an eye toward how local news consumption practices vary by community type.  Specifically, it focuses on the ways residents in large cities, suburbs, small towns and rural areas compare in their levels of interest in local news, the topics they are most interested in, and the sources they rely on to learn about those topics.  

The results indicate that from large urban areas to rural communities, Americans often report similarly high levels of interest in news in general, in local news and information, and in national and international news.  Moreover, similar percentages of adults report following the specific local topics asked about, regardless of the type of community in which they live.

Still, community differences do emerge in the number and variety of local news sources used, as well as the degree of "local news participation" and mobile news consumption.  Many of the differences in local news consumption emerging from these data reflect the varying demographic composition of different community types in the U.S.

Some of the main differences include:

Urban residents: People who live in large cities rely on a wider combination of platforms for information than others and are more likely to get local news and information via a range of digital activities, including internet searches, Twitter, blogs and the websites of local TV stations and newspapers. Urbanites were also those least tied to their communities in terms of how long they lived in the community and how many people they know. They were the least interested of all groups in information about local taxes. At the same time, those who live in large cities, along with suburban residents, are the most likely to be digital "news participators" who email local stories to others, post material on social networking sites, comment on news stories online, or contribute to online discussions on message boards.  Also along with suburbanites, they are more likely to get news via mobile devices.
Additionally, they are the most likely to rely on local TV news for information about breaking news, weather, crime, politics, and traffic.

Suburban residents: Those who live in suburban communities are more likely than others to rely on local radio as a platform (perhaps because of relatively longer commuting times); they are more interested than others in news and information about arts and cultural events; and they are particularly interested in local restaurants, traffic, and taxes. Like urbanites, they are heavy digital participators who comment and share the news. These suburban residents rely mainly on the internet for information about local restaurants, businesses, and jobs. They look to television news for weather and breaking news. 

Small town residents:  Along with rural residents, people who live in smaller towns are more likely to rely on traditional news platforms such as television and newspapers to get local news; newspapers are especially important to them for civic information. Small town Americans prefer the local newspaper for a long list of information-including local weather, crime, community events, schools, arts and culture, taxes, housing, zoning, local government and social services. Residents of smaller towns are also the most likely to worry about what would happen if the local newspaper no longer existed.            

Rural residents: Those who live in rural communities generally are less interested in almost all local topics than those in other communities. The one exception is taxes. They are also more reliant on traditional platforms such as newspapers and TV for most of the topics we queried. And they are less likely than others to say it is easier now to keep up with local information.

It is important to note that the choices about information acquisition are not necessarily the same in all communities. For instance, it might be the case in rural areas that the local newspaper and broadcast outlets are not online or have a very limited online presence and that is a determinant in whether residents get local information online or not. Our survey asked what consumers do in terms of information acquisition and what sources they "rely on." It did not ask what they could do-that is, what information and sources are available in their communities. In many respects- but not all respects- people generally want similar types of news and information. In some communities, they have many choices and are quite deliberate in which platform they use to get which kind of information. In other communities, they have fewer choices.

Across the four community types, residents report similarly high levels of general interest in news, attention to local news, and interest in most specific local topics

The percentage of Americans who indicate they enjoy keeping up with the news 'a lot' ranges from 53% to 60% across the four community types, and similar percentages follow international news closely regardless of what is happening (ranging from 54% to 58%).  More residents in all community types follow local news this closely, with percentages ranging from 68% in large cities to 73% in rural areas.  Interest in national news is highest among suburban residents, with three quarters (74%) following closely regardless of what is going on, compared with two-thirds (67%) of residents of other types of communities.

Across the four community types, residents also report similarly high levels of interest in most of the 16 specific local topics asked about

The survey asked a nationally representative sample of adults whether they ever get news and information about 16 different local topics.

For 11 of 16 local topics that we queried, there are no statistically significant differences in interest level across residents of different community types.  The five local topics for which interest levels differ are arts and culture, restaurants, traffic, taxes, and housing.  Residents of suburban communities show the highest interest level in all five of these topics, while rural residents show the lowest interest level on all but taxes.  In the case of taxes and tax issues, residents of large cities are the least likely to say they follow the topic.

Residents of different community types differ in the sources they rely on for their local news

Residents of large cities, who on the whole skew younger and are more mobile than populations living in other community types, are most likely to stay informed about local topics that interest them through a combination of online and traditional sources. They are particularly likely to get local news through internet searches, Twitter, blogs, and websites of TV and newspapers.  In contrast, small city (31%) and rural (34%) residents are more likely than those in larger cities (21%) and suburbs (16%) to rely solely on "traditional" forms of media for their local news such as local print newspapers and broadcast television. 

Suburban residents are distinct in their higher dependence on local radio (likely due to longer commutes to work), while small city and rural residents stand out in their reliance on word of mouth for some types of local information.

Urban and suburban residents on average use more sources of local news than their small town and rural counterparts and are more likely to consume local news on mobile devices

In an average week, residents of large cities and suburbs use more sources of local information than others.  On average, residents of large cities and suburbs use just under four sources per week (3.63 and 3.72, respectively) compared to those in small cities or towns and rural areas who use closer to three sources per week (3.31 and 3.28, respectively).  In addition, more than half of urban (53%) and suburban (57%) residents get some kind of local news or information via cell phone or tablet computer compared with 45% of small city and 35% of rural residents. 

The most active "local news participators" also tend to reside in suburban and urban communities

Suburban residents are more likely than any of the other groups (53% vs. 45% large city, 36% small city, 32% rural) to actively participate in local news and large city residents are more likely than small city or rural residents to be classified as local news participators, meaning they email local stories to others, post news or information about the local community on social networking sites or Twitter, comment on local stories they read online, contribute to online discussions on message boards about the local community, and the like.  

Rural residents are the least likely to say it is "easier" to keep up with local news and information today than it was five years ago

Residents of large cities (59%), suburbs (60%) and small towns (55%) are more likely than those in rural communities (46%) to say it is "easier" to keep up with local news and information today than it was five years ago.   Yet at the same time, residents of large cities are the least willing to pay for local news content through a paid subscription to a local newspaper (22% vs. 40% suburbs, 33% small towns, 37% rural). 

These are just some of the findings about Americans from different communities as identified in a survey conducted by the Pew Research Center's Project for Excellence in Journalism and Internet & American Life Project, in partnership with the Knight Foundation.  The nationally representative phone survey of 2,251 adults ages 18 and older was conducted January 12-25, 2011 and included 750 cell phone interviews.

Read the complete report.

---

Footnote:

[1] See "How People Learn About Their Local Community," available at http://www.pewinternet.org/Reports/2011/Local-news.aspx.