It could be the new maxim of 21st century politics: To find voters, look online.
They're there in increasing numbers, in a politically diverse population that's growing, expanding its Internet activities and highly distinctive, with remarkable levels of political and social engagement. It's a group with the size and clout to change the way election politics happen in America.
For the first time in polls since 1996, this ABC News/Facebook survey finds the Internet rivaling newspapers as one of Americans' top two sources of news about the presidential election. It's also the only election news source to show growth, doubling since 2000.
One reason is the Internet's advance overall: Seventy-three percent of adults now go online, the most in polls since the dawn of the Internet age. Forty percent use the Internet specifically for news and information about politics and the election, surpassing the previous high, 35 percent in a 2004 survey.
Television remains predominant; 70 percent say it's one of their top two election news sources. But while still far ahead, that's down by 8 points since 2004 and by 15 points since 1996 in Pew polls. Newspapers follow, named by 26 percent as a top election news source — vastly down from 60 percent in 1996. Catching up with newspapers, 23 percent now cite the Internet as a main source of election news — twice the level seven years ago.
This national survey marks the partnership between ABC News and Facebook, the social networking site, in 2008 election coverage. The two organizations, with WMUR-TV in Manchester, N.H., are sponsoring a pair of debates among the presidential candidates in New Hampshire that air starting at 7 p.m. ET Saturday.
ENGAGEMENT — The four in 10 adults who use the Internet for election information are highly attuned to politics; compared with other adults, they're 22 points more likely to be following the campaign closely, 21 points more apt to plan to vote in an upcoming primary or caucus, 13 points more apt to report having voted 2004 and 10 points more likely to report being registered. That's engagement.
Other measures of political involvement point the same way. Eighty-three percent who go online for political information say they understand what's going on in government, and 78 percent feel they have a say in what it does. These comfort levels drop very sharply among other Americans, to 58 percent and 56 percent, respectively — 25 percentage points and 22 percentage points lower.
Social engagement is higher, as well: Rebutting onetime notions of isolation in cyberspace, 72 percent of people in the online political population report doing volunteer work for a church, charity or community group; volunteerism drops to 52 percent among other adults.
Online political participation, moreover, extends to in-person political discourse. Fifty-eight percent in the online political population regularly discuss or debate political issues with others in a face-to-face setting. Far fewer other Americans say they talk politics, 40 percent.
At the same time, even for the online political population, the Internet is more of a tool for information than for discussion. While nearly six in 10 regularly talk about politics with others face-to-face, far fewer, 6 percent, regularly engage in political discussions online (and just 10 percent ever do). That's one of many areas in which the intersection of the Internet and politics may yet develop further.
YOUNG VOTERS — Differences in engagement are especially striking among young adults — a group that, overall, tends to turn out at the polls in comparatively low numbers. Younger people are disproportionately apt to go online for political news — and those who do so are much more politically involved than others their age.
Among people younger than 30 who go online for political information, 84 percent are following the 2008 campaign closely, including 33 percent "very" closely, compared with just 51 percent and 10 percent, respectively, of people the same age who don't look into politics online. They're also 25 percentage points more likely to say they'll vote, 21 points more apt to feel they have a say in what the government does and 16 points more likely to participate in volunteer work. Particularly in the search for young voters, online's the locale.
ACTIVITIES — By far the top use of the Internet for political information is to look into the candidates' positions on the issues; 72 percent of the online political population does so, sharply up from its level in a 2004 Pew poll, 47 percent.
Given the expansion of broadband access, the steepest growth is in viewing candidate- or election-related video over the Internet. Now the second most common online political information activity, it's nearly doubled, from 30 percent in 2004 to 58 percent now. And among younger than 30 who go online for politics, it's higher still — 70 percent.
Among other top activities of the online political population, 52 percent use the Internet to check the accuracy of claims made by or about the candidates (up from 39 percent in Pew's 2004 poll), 50 percent check the candidates' standings in the polls, 44 percent visit the candidates' Web sites and 43 percent — another sharp increase from 2004 — check information on the candidates' voting records in office.
About three in 10 online political users have looked up candidates' endorsements or ratings by interest groups online, sent or received e-mails about the candidates or campaigns or looked up information on where or when to vote. Just shy of two in 10 have signed up for online alerts about the latest political or election news.
All these represent huge numbers of individuals. With the online political population at about 90 million overall, the number who look at political video clips is about 52 million people; visited candidates' Web sites, about 40 million; signed up for online news alerts, roughly 16 million.
Other activities are less prevalent, yet large enough to be strongly influential — with far more potential. The presidential candidates already have produced enormous fundraising results online, even though just 5 percent of Internet users have donated. Also, relatively few use social networking sites for political information or participate in online discussions or chat groups about the election (both these peak among people younger than 30).
Given all these activities, the Internet's role in decision-making has grown. Among people who go online for political news and information, 64 percent call it an important source of information in their ultimate vote choice — up from 52 percent in 2004. And again there's a difference by age: Among under-30s in the online politics population, far more, 80 percent, say it's important in their decision-making.
MULTISOURCING — The Internet to a large extent is being used as an additional information source, not a replacement one. Among people who go online for political information, many also use other political news sources — television (77 percent), newspapers (63 percent), radio (55 percent) and magazines (35 percent). Use of these sources is at similar levels whether or not people go online.
Still, while multiple sources are used, there are differences in extent or frequency of use. People who go online for political news and information are much more likely to cite this as one of their two main sources of election news (44 percent, rising to 57 percent of those younger than 30), and less apt than others to cite television as a main source (57 percent).
Use of the Internet for political news and information in some cases means using traditional media outlets online — more for television sources than for newspapers. Asked the top few Web sites they use for election news, 37 percent mention cable TV Web sites and an additional 33 percent mention network news Web sites; 10 percent mention newspaper Web sites. News portals, which link to any of an array of primary sources, are cited by 31 percent. (Multiple answers were accepted.)
EASE and TRUST — Using the Internet is about convenience and content alike. Among those who go online for political information, 52 percent say convenience is the chief reason; 34 percent say it's either because other sources don't give them the information they want or because the Internet offers information that's not available elsewhere.
Still, trust and confidence in the traditional news media is as high among the online political population (62 percent) as it is among other Americans (and, contrary to conventional wisdom, actually peaks among under-30s). The difference runs in the other direction: Trust in so-called "new media" sources on the Internet is nearly 20 points lower among people who don't go online for political information (43 percent) than among those who do. It's lower still, 31 percent, among those who are offline entirely.
In an even bigger gap, online users of political information are much more likely than other Americans to say the Internet plays a positive role in election campaigns, 71 percent to 29 percent.
Convenience, meanwhile, shows up in another finding, on usability: Eighty-one percent in the online political population say they generally can find what they're looking for on the Internet, up from 70 percent in 2004.
AGE and ATTITUDES — The online political population, as noted, is comparatively young — 71 percent are younger than 50 (compared with half of other adults) and nearly a third are younger than 30 (compared with 17 percent of other adults). Just 7 percent are older than 65, compared with 26 percent of other adults. (Among the 27 percent of adults who don't use the Internet at all, 43 percent are seniors.)
It follows that online political information users hold some views associated with younger age groups — more likely to support gay civil unions, legal abortion and a legal-status program for illegal immigrants.
Online political information users also are more apt than other Americans to define themselves as liberals, 27 percent vs. 17 percent, and somewhat less apt to be conservatives, 31 percent vs. 38 percent. Conservatism peaks, at 46 percent, among people who don't use the Internet at all.
In other respects, however, similarity is the rule. There are no substantial differences between online adults and the broader public in assessments of the country's direction overall, President Bush's work in office or whether the war in Iraq was worth fighting; in support for the death penalty; in being a self-identified feminist; or, with a few exceptions, in emotional responses to politics today (such as anger, hopefulness or frustration).
Partisan allegiance — the share of Democrats, independents and Republicans — is about the same among the online political population as among others.
Preferences in candidate support reflect the age differential. In the Democratic race, Barack Obama, who's more popular with younger adults, does better with the online political information population (33 percent support) than with others (22 percent). There's almost no such difference for Hillary Clinton (who leads Obama in both groups). And there are no Obama-sized differences in the Republican race; the greatest are that Mitt Romney does 8 points better, and John McCain a scant 5 points worse, with the online political population.
Gaps for Obama again are prominent in hypothetical general-election matchups. His support against either Rudy Giuliani or Mike Huckabee is about 10 points higher in the online political population than it is among other Americans.
GROUPS — There are demographic differences beyond age in the online political information population. A remarkable 44 percent are college graduates, for example, compared with 16 percent of others (and just 7 percent of those who don't use the Internet at all).
Sixty-four percent report household incomes of more than $50,000 a year, compared with 40 percent of other adults (and 26 percent of entirely offline adults). Indeed 25 percent report $100,000-plus incomes, again well over the level for other adults.
People who go online for political information are 9 points more likely than all other Americans to be married, 11 points more apt to have kids younger than 18 at home and 16 points more apt to have a full-time job. By and large these follow age or education, which along with income, are strong predictors of Internet use.
There are some interesting attitudinal differences within subgroups; for instance, women in the online political information group are 15 points more likely than other women to identify themselves as liberals. But some of the most striking differences are among conservatives; those in the online information population are noticeably more in agreement on some issues than are conservatives more broadly.
For example, in the online political population, 71 percent of conservatives approve of Bush's performance, 71 percent say the war in Iraq was worth fighting and 51 percent say the country's going in the right direction overall. Among all other conservatives, not just those who go online for political news and information, these numbers are sharply lower. There's no such difference in attitudes among liberals or moderates.
INTERNET VOTING — Finally, for all their use of the Internet, the online political population is divided evenly — as are all Internet users overall — on whether they'd support allowing people to vote online, assuming such a system could be made secure from fraud. Adults who don't use the Internet at all, meanwhile, oppose the idea by well over a 2-1 margin.
A broad concern in both populations is whether online voting indeed could be made secure. Fewer than two in 10 Americans see that happening any time soon — down from a 1999 poll. Two-thirds think it will take many years before it's possible, and one in 10 think it'll never happen — including equal numbers of online and offline Americans alike.
METHODOLOGY — This ABC News/Facebook poll was conducted by telephone Dec. 16-19, 2007, among a random national sample of 1,142 adults, including an oversample of 18- to 29-year-olds for a total of 274 respondents under age 30 (weighted back to their correct share of the national population). The results have a margin of sampling error of 3 points for the full sample, 4.5 points for the online political information population and 4 points for others. Sampling, data collection and tabulation by TNS of Horsham, Pa.