The final indicator in this section is how much attention people paid to the Senate contest. If the Record‘s public journalism approach stimulated interest by covering the issues people said they wanted to learn about, Record readers would be expected to pay more attention to the campaign than readers of other papers. But again, the data in Table 5 do not offer evidence to support this hypothesis. Record readers paid about the same amount of attention to the Senate race as other newspaper readers in the state.
| How Closely Followed Campaign | Record | Other NJ | |||
| Very closely | 9 | 6 | |||
| Somewhat closely | 33 | 32 | |||
| Not too closely | 31 | 35 | |||
| Not much at all | 27 | 27 | |||
| Total | 100% | 100% | |||
| (n)* | 400 | 458 | |||
| * Columns here are based on the POSNPALL variable. Chi square values for none of the three independent variables were statistically significant. | |||||
KNOWLEDGE AND DECISION MAKING
Objective and Subjective Knowledge
As noted earlier, knowledge about the Senate candidates started almost from ground zero in the fall campaign. On our pretest survey conducted during the summer, 79 percent of Record readers could not name either of the two Senate candidates in response to an open-ended question; 12 percent could name either Torricelli or Zimmer, and just nine percent could name both. Readers of other papers fared even worse: 88 percent could name neither, four percent one and eight percent both candidates. The difference is probably due to Torricelli’s greater name recognition in his home area, but it was clear that citizens did not know his policy views in any detailed way. This difference between Record and non-Record readers held through the end of the campaign. While unaided name recollection may be a tough measure, remarkably 42 percent of Record readers and 55 percent of non-readers could name neither candidate after an election where perhaps 15 million dollars was spent on political advertising. Just 36 percent of Record readers and 32 percent of non-Recordreaders could volunteer the names of both men.
While there is no pre-measurement available on being able to ascribe issue positions to the candidates, Record readers did not end the campaign season appreciably more knowledgeable than readers of other papers. We have presented the figures for each of the three independent “newspaper exposure” variables so that readers can see the small variations. Only among the variable with a floor of exposure three times a week do the differences approach statistical significance. Interestingly, while Recordreaders were not really any better “objectively” informed than were readers of other papers, more of them felt better informed. The data in Table 7 shows Record readers to be about 10 percentage points more likely than readers of other papers to say they think they knew where the candidates stood on issues at the end of the campaign.
| Correct # of issues | POSNPALL | POSNPALL | POSNPPUR | POSNPUR | PRENP3X | POSNP3X |
| RECORD | OTHER NJ | RECORD | OTHER NJ | RECORD | OTHER NJ | |
| None | 23 | 30 | 25 | 33 | 21 | 30 |
| One | 29 | 24 | 31 | 29 | 30 | 23 |
| Two | 24 | 20 | 24 | 19 | 21 | 19 |
| Three to Five | 25 | 25 | 21 | 18 | 28 | 29 |
| Total | 101% | 99% | 101% | 99% | 100% | 101% |
| (n)* | 400 | 459 | 231 | 285 | 301 | 336 |
| * p for POSNPALL = .11, POSNPPUR = .22, POSNP3X=.06. | ||||||
| How Well Do You Know Where Candidates Stand on the Issues | Record | Other NJ |
| Very well | 13 | 7 |
| Somewhat well | 36 | 33 |
| Not too well | 29 | 33 |
| Not well at all | 22 | 27 |
| Total | 100% | 100% |
| (n)* | 400 | 457 |
| * Columns here are based on the POSNPALL variable. Chi square value p =.02; POSNPPUR = .14, POSNP3X = .04. | ||
Opinionation
An early stage of the decision-making process in choosing a candidate to vote for is opinionation–the process of forming opinions of the candidates. As we have seen from earlier data, both candidates entered the general election season largely unknown to the great majority of voters. Did The Record‘s coverage of more voter-defined relevant information aid citizens in this process? Unfortunately, it is not possible to disentangle The Record‘s (or any newspaper’s) effects from the barrage of spot advertisements, which would also contribute to the electorate forming and expressing opinions. Simple opinionation–does one have an opinion or not?– is not a terribly useful indicator when the time span between pre- and posttest is from summer to after the election.
Yet there may be an indicator in the area of opinionation that is useful. The information presented in spot ads is usually brief and one-sided. While citizens may form impressions from the ads, it is also true that these opinions may be quite soft, resting on a precarious informational base. With additional information comes more firmly held attitudes, as they are more solidly grounded (Converse 1966, Zukin 1977). Thus if some citizens received more useful or “better” information, we would expect that a greater proportion of them would have more “solid” opinions–that they would be more strongly held.
| Opinion of Torricelli | Record | Other | Opinion of Zimmer | Record | Other |
| No Opinion | 10 | 19 | No Opinion | 17 | 21 |
| Weak Opinion | 62 | 58 | Weak Opinion | 65 | 54 |
| Firm opinion | 29 | 23 | Strong Opinion | 18 | 25 |
| Total | 100% | 100% | Total | 100% | 100% |
| * Columns here are based on the POSNPALL variable. Chi square values are significant at .00 | |||||
Because the Democratic candidate, Bob Torricelli, was from Bergen County and had greater visibility before the campaign started, the power of our posttest experimental-control group comparisons are limited: the two groups were different on this score to start with. And indeed, a slightly greater percentage of Recordreaders ended the campaign season with a firmer opinion of Torricelli. However, as the data on the right hand of Table 8 indicate, precisely the reverse was true for the Republican candidate, Dick Zimmer. All in all, the data do not support the notion that Record readers had better or different information that made it easier for them to come to firmer judgments about the Senate candidates.
A corollary of this hypothesis is that having better information would have madeRecord readers more certain of their judgments. And this did not prove to be so either. Some 45 percent of Record readers said that when they voted for Torricelli or Zimmer they were “completely sure that he was the right choice,” as did 44 percent of those who read a newspaper other than The Record. Looking back after the campaign, Record readers were split, with half saying they had enough information to decide how to vote and half saying there were other things they would have liked to know. This is not statistically distinct from the 44-56 division of non-Recordreaders, where the majority would have liked additional information.
Record and non-Record readers were also similar in the proportion of them saying they had decided who to vote for in the week or two before the election. Slightly more Record readers said they had decided in the summer, even before the campaign began; slightly more non-Record readers said they decided after Labor Day but before the last two weeks of the campaign.
In sum, the evidence does not support the hypotheses that Record readers were more knowledgeable or employed a “better” decision-making process in coming to a judgment about how to vote in the Senate election.
Participation
The findings here can be summarized succinctly. Record readers were no different from readers of other papers in the extent to which they reported voting in the election and talking about the election. Some 83 percent of Record readers claimed to have voted in the election, compared to 80 percent of those reading some other New Jersey paper.
About one-fifth of each group (Record 17%, non-Record 19%) reported talking about the campaigns “often” outside their families, with another 37 percent of each group saying they did so “sometimes.” Clearly, when compared to the control group, reading The Record did not stimulate conversation or cause people to be more participative on these indicators.
ATTITUDES TOWARDS THE POLITICAL SYSTEM AND TOWARDS THE RECORD
The Political System
We employ two attitudinal items to help us ascertain if there were any effects of Record readership on general attitudes towards the political system. Admittedly, this is a most difficult test. Attitudes towards the political system are generally long-term, stable predispositions, unlikely to change upon something as frail as the provision of “Campaign Central.” Nevertheless, respondents were asked to indicate if they felt that as an individual they could have an impact on what the government does, and to agree or disagree about whether it was worthwhile to vote. The pre-election and post-election results for the first indicator are presented in Table 9.
| Pre-election Survey | Record | Other | Post-election Survey | Record | Other |
| Strongly Disagree | 29 | 30 | Strongly Disagree | 21 | 19 |
| Somewhat Disagree | 30 | 23 | Somewhat Disagree | 30 | 26 |
| Somewhat Agree | 18 | 22 | Somewhat Agree | 20 | 22 |
| Strongly Agree | 24 | 26 | Strongly Agree | 29 | 33 |
| Total | 101% | 101% | Total | 100% | 100% |
| (n)* | 379 | 369 | (n)** | 398 | 453 |
| * Columns here are based on the PRENPALL variable. Chi-square value p = .16 | |||||
| **Columns here are based on the POSNPALL variable. Chi-square value p = .36 | |||||
These results may be quickly summarized: The experimental and control groups were the same prior to the election and were the same at the end of it. Even looking at panel changes from Time 1 to Time 2, so that the respondent base is held constant, those exposed to The Record are not statistically different from those who read some other newspaper. The act of voting and going through the electoral cycle seems to have made citizens somewhat more empowered, as more individuals moved up to higher points on the scale than moved down, but this was true for Record and other readers alike. The findings with the other indicator of wondering whether the act of voting is worthwhile are virtually identical. There was not a statistically significant difference between the two groups.
Evaluations of The Record
A battery of 10 items bearing on the nature of press coverage of the campaign was administered to those who read The Record and those who read other papers. Respondents were asked to think specifically about the paper they named and to rate its coverage on a scale of 1 to 10 depending on how applicable they felt each phrase was. One would mean the phrase was not at all descriptive, while 10 would mean it described the paper’s coverage completely.
These items have been sorted by their mean value and are presented in Table 10. An eyeball’s glance is all that is necessary to see that evaluations of The Record were virtually identical to all the other papers rated in the survey. The mean differences between the “Record” and “Other” pairings are minuscule compared to the standard deviations in all of the 10 cases.
Moreover, there was no change among the Record readers on the panel for eight of the ten items included on both the pretest and posttest surveys. The average (absolute value) change from Time 1 to Time 2 was just .14 on the 10 point scale; the accompanying average standard deviation was 2.58. Clearly, respondents exposed to The Record did not change their opinions of The Record‘s coverage over the course of the campaign, just as The Record was judged to be no different from other newspapers.
Multivariate Analysis
A number of regression equations were run trying to see if any “Record effects” were being masked or if the simple bivariate analysis were incapable of revealing more complex relationships. However, models that attempted to explain the many of dependent variables reported here by education, age, Record readership, general newspaper readership and other independent variables fully confirmed the bivariate analyses presented in this section.
Overall, then there is little in the data to suggest that The Record was successful in its civic journalism enterprise. The final section of this paper takes up why this may have been the case.
| Statement | Mean Record | Mean Other | Mean Diff. | St. Dev. Record | St. Dev. Other |
| Made you feel like it wasnt’t worth voting | 3.98 | 3.69 | .29 | 2.69 | 2.59 |
| Favored the Republicans | 4.61 | 4.53 | .08 | 2.33 | 2.24 |
| Boring | 4.65 | 4.26 | .09 | 2.49 | 2.45 |
| Showed how people like you can make a difference | 5.23 | 5.21 | .02 | 2.48 | 2.42 |
| A watchdog for the people | 5.56 | 5.52 | .14 | 2.17 | 2.21 |
| Favored the Democrats | 5.75 | 5.66 | .09 | 2.49 | 2.42 |
| Gave you the information you wanted to decide how to vote | 5.91 | 5.93 | -.02 | 2.28 | 2.32 |
| Included the issues you cared about most | 6.10 | 6.21 | -.11 | 2.15 | 2.13 |
| Honest and believable | 6.12 | 6.33 | -.21 | 2.21 | 2.21 |
| Written for people like you | 6.27 | 6.30 | -.03 | 2.28 | 2.22 |
| * n’s for the Record range from 360 to 384 (don’t know’s excluded); n’s for Other range from 761 to 811. | |||||
THE LESSONS OF MINIMAL CONSEQUENCES
For the final nine weeks of a hard-fought, high-visibility political campaign, the editors of The Record undertook what they believed was an important public service: They broke away from the conventional wisdom of “the boys on the bus,” from the phantom issues favored by consultants and spin doctors, and covered the campaign as regular citizens saw it. The effort consumed about $100,000 in additional newsprint, and required almost a dozen reporters and editors to be shifted temporarily from their regular assignments. And by all accounts, hardly anything happened because of it.
Record readers were no more interested in the election or knowledgeable about the candidates and the issues than readers of other New Jersey newspapers. They had about the same level of trust in politics as other newspaper readers, and were not significantly more likely to vote or to talk about the election with people outside their family, once demographic differences were controlled. Their opinions of The Record and its political coverage were roughly comparable to the opinions of other New Jersey readers about their local newspaper.
It would be presumptuous to make broad judgments about the potential of public journalism based on the apparent lack of success of one experiment. Still, The Record‘s “Campaign Central” experience raises useful and important questions about the circumstances under which public journalism can make a difference. Our results suggest there are significant limits – at least in the short term – to the ability of journalists to reconnect citizens with democratic institutions solely by altering the way they write about them.
That does not necessarily mean that public journalism is not worth doing. The most powerful arguments in support of public-style coverage are, after all, essentially moral: They contend that journalists have a public responsibility to behave in ways that actively support and enhance the workings of democracy. It may be, however, that some of the tactics identified by the first wave of public journalists are not altogether what citizens want or need.
What happened to public journalism in this experiment? Why was there no significant change in citizen opinion or behavior? There are at least six possible explanations:
- The Record‘s experiment was too short to produce significant results. Just as citizens did not become cynical about politics and detached from political life overnight, one might argue it is unreasonable to suggest that these bonds can be renewed in the space of a few weeks. Public-style coverage may need to be repeated over several election cycles before the electorate becomes reengaged.
There is, indeed, some evidence for this point in a recent review of other public journalism projects. That study, also funded by the Pew Center for Civic Journalism, found that public-style coverage was more likely to produce community change if it focused on a single issue over a long period of time. An “episodic approach,” these authors found, “diffuses both the intensity of the project and its impact” (Thorson et al. 1997: 7).
Yet while The Record‘s project lasted for only nine weeks, it occupied an extraordinary amount of space – 54 full pages, plus an array of four-color teases and other promotional devices. As Shanto Iyengar’s pathbreaking experiments have shown, exposure to a concentrated media message can affect public opinion about politics in a relatively short time (see, for example, Iyengar 1991). If public journalism does have the potential to produce change, it is surprising that it did not noticeably do so in some fashion when given such unusual prominence.
- The Record‘s coverage wasn’t sufficiently “public.” It is conceivable that the impact of “Campaign Central” was blunted by the decision of Record editors to continue running conventional campaign stories elsewhere in the newspaper and to give public-style and conventional articles equal consideration for the front page. Because the horse-race approach has dominated campaign coverage for so long, it may be that even modest amounts of it are enough to blind citizens to a change in focus. That might explain the greater success reported by Miller in the 1992 Charlotte experiment, where editors banished horse-race polls from the paper and gave public-style coverage special display on the front page.
Record editors established the “compromise” between public and conventional coverage because they felt, as a matter of ethics, that they would be censoring the news to leave legitimate information about the campaign out of the paper solely because it did not conform to the public journalism model. There would have been strong objections to the project from several key newsroom managers if conventional campaign stories had been completely barred from the newspaper or given a decidedly secondary position.
Public journalism already faces considerable organizational resistance in many, if not most, news organizations from reporters and editors who question its value (see, for example, Thorson et al. 1997: 23-25). If it cannot succeed in tandem with other approaches to political coverage, it is unlikely to gain many more converts.
- Public journalism was drowned out by competing messages from other media. Even though The Record is the dominant newspaper in its home market, it hardly has a monopoly on its readers’ attentions. More than two-fifths of Record readers – 44 percent – said television was their most important source of campaign news. And according to some researchers, the message coming from network television about the 1996 campaign was very different from the message that The Record sought to convey. While “Campaign Central” was trying to portray the campaign as serious and important, network television was depicting it as dull and irrelevant.
Analysts at the Annenberg School for Communication at the University of Pennsylvania monitored network newscasts from the start of party conventions in midsummer through the fall. They found that the three major networks – ABC, CBS, and NBC – spoke about 50 percent fewer words per newscast about the campaign in 1996 than they had in 1992. Asked to explain the difference, a spokeswoman for CBS said: “You must agree with me that there’s not much of a story this year to cover…. How many different ways can you say the same thing?” (Scarborough 1996).
However, our data offer little circumstantial evidence that watching the news on television discouraged people from voting. Although respondents who watched network newscasts four or more nights a week were somewhat less likely to vote than people who watched the news less often, the relationship turned around once controlled for differences in gender, age, education, and partisanship. With these multivariate controls in place, there was a statistically significant positive relationship between watching television news and voting.
- Public journalism could not overcome negative messages about the democratic process from the candidates and the political parties.Journalists may, as the supporters of public journalism argue, deserve some of the blame for the rise in cynicism about politics. But candidates and parties are hardly without fault – and 1996 was no exception.
In a year marked from coast to coast by negative campaigning, New Jersey’s Senate race was widely agreed to be among the roughest. Many observers felt Zimmer, the Republican, took hard-hitting advertising to a new level with a 60-second spot about Torricelli’s alleged relationship with a Korean businessman and Democratic fund-raiser who fled the country to escape prosecution on a charge of embezzlement. The spot was photographed and edited to look like an excerpt from a local newscast, complete with the words “Breaking News” across the screen at the start. Torricelli replied with commercials that wrongly claimed Zimmer voted to cut Medicare, deny coverage for mammograms, and reduce funding for Head Start.
By the end of the campaign, voters were frustrated and fed up by the barrage of accusations, name-calling, and twisted facts. “They’re all a bunch of bozos,” said one focus-group respondent. Said another: “I think anyone who would be interested would be turned off.” Said yet another: “It was pretty hard if you wanted to focus on issues.”
Once again, The Record‘s public journalism was challenged by the environment around it: While “Campaign Central” was attempting to frame the Senate race as a constructive debate, the candidates were spending more than $17 million on television commercials that made it seem like a showdown in a sandbox. And even in the expensive New York and Philadelphia television markets, $17 million buys a tremendous amount of visibility – especially when the message is tightly concentrated and hammered endlessly from sunup to midnight.
It is difficult to imagine that any sort of journalism, traditional or public, will make people feel much better about the possibilities of democracy so long as the dominant product of political campaigns is scorched earth. Public journalism may have greater success if it devotes less attention to elections and more to other places and processes where citizens are actively connected to civic life rather than turned off by it. The rush of recent work sparked by Robert Putnam’s vision of declining social capital shows that most Americans are not “bowling alone,” but are still engaged in their communities and in their common future – albeit perhaps in different, less confrontational ways than either politicians or journalists expect (Ladd 1996, Pew Research Center 1997). A journalism which seeks to strengthen civic connections will likely find that it is heard more clearly on this less cacophonous stage.
- Despite public journalism’s innovations in style and emphasis, the essential information citizens wanted was still too difficult to find, understand, and act upon. For some journalists, the most troubling result of the “Campaign Central” experience may be that so many of the standard devices which editors use to pull readers through a newspaper appear to have had no effect. “Campaign Central” had color teases on the top of the front page, an eye-pleasing modular graphic design, predictable daily placement at approximately the same position in the first section, and above all, clear, interesting writing with crisp headlines – and still fewer than one in five readers noticed it was there.
The enthusiastic reception for the “Voter’s Guide” in the focus groups suggests one intriguing reason why so many people read past most of The Record‘s public journalism: They may no longer have the time or inclination to absorb as much information in paragraph form. Our focus group respondents clearly wanted to know more about the issues and the candidates; they wanted to make meaningful choicesbased on substance. But they wanted that information presented in a more readily digestible format – as bullet points in a concise, easy-to-read chart. If it were presented as a chart, not only did our respondents say they would be more likely to read and understand it, but they would also be more likely to believe it.
As controversial as public journalism has been so far, this finding may strike even closer to the heart of newsmaking. For journalists are first and foremost word people – proud, competitive lovers of language who judge each other by their ability to spin memorable phrases under incredible pressure. It is one thing to ask journalists to shift the focus of political stories from horse races to issues, and quite something else to ask that they give up sentences and paragraphs.
The next phase of public journalism, then, may be just as contentious in newsrooms as the first – and just as difficult. A challenge for public journalists will be to address the demand for accessible, credible, and compact information about the workings of democracy without reducing it to vapid or simplistic “News McNuggets” – an outcome that would certainly be no better than the press we have today, and that could potentially be much worse.
- Finally, it may be that the public is not sufficiently interested in politics for public journalism to be of service. While perhaps not perfect, The Record‘s efforts were real and well promoted in the newspaper. If citizens wanted better – or at least different – information, it was there for them to find, yet few availed themselves of the opportunity to do so.
One of the most surprising findings of the focus groups, confirmed by the quantitative survey, was that the vast majority of Record readers had no idea that “Campaign Central” existed. Even though it may in itself be a result of the prevailing negativity of politics, campaigns, and traditional journalism, there may at present be no demand on the part of the public for something better. There is no guarantee that if we build it, they will come.
REFERENCES
Broder, David. 1991. “A New Assignment for the Press.” Press-Enterprise Lecture no. 26. Riverside, Calif.: The Press-Enterprise.
Charity, Arthur. 1995. Doing Public Journalism. New York: Guilford Press.
Conte, Chris. 1996. “Civic Journalism.” CQ Researcher 6: 819-837.
Converse, Philip. 1966. “Information Flow and the Stability of Partisan Attitudes.” In Angus Campbell, Philip Converse, Warren Miller, and Donald Stokes, eds.,Elections and the Political Order. New York: Wiley.
Denton, Frank, and Esther Thorson. 1995. Civic Journalism: Does It Work?Washington: Pew Center for Civic Journalism.
Eisner, Jane. 1994. “Should Journalists Abandon Their Detachment to Solve Problems?” Philadelphia Inquirer, October 16, p. E7.
Fallows, James. 1996. Breaking the News: How the Media Undermine American Democracy. New York: Pantheon.
Frederick Schneiders Research. 1996. ” ‘Your Voice, Your Vote’: Results From a Survey of Registered Voters in North Carolina.” Photocopied. Washington: Pew Center for Civic Journalism.
Hume, Ellen. 1996. “The New Paradigm for News.” The Annals 546 (July): 141-153.
Iyengar, Shanto. 1991. Is Anyone Responsible? How Television Frames Political Issues. Chicago: University of Chicago Press.
Jackson, William E. Jr. 1996. “The Press Cops Out.” New York Times, October 7, p. A17.
Jamieson, Kathleen Hall. 1992. Dirty Politics: Deception, Distraction, Democracy. New York: Oxford University Press.
Kelly, Michael. 1996. “Media Culpa.” The New Yorker, November 11, pp. 45-49.
Ladd, Everett Carll. 1996. “The Data Just Don’t Show Erosion of America’s ‘Social Capital.'” The Public Perspective, June/July, p.1.
Merritt, Davis. 1995. Public Journalism and Public Life. Hillsdale, N.J.: Erlbaum.
Miller, Edward D. 1994. The Charlotte Project: Helping Citizens Take Back Democracy. The Poynter Papers, no. 4. St. Petersburg, Fla: The Poynter Institute for Media Studies.
Patterson, Thomas E. 1993. Out of Order. New York: Knopf.
_______. 1996. “Bad News, Bad Governance.” The Annals 546 (July): 97-108.
Pew Research Center for the People and the Press. 1997. Trust and Citizen Engagement in Metropolitan Philadelphia: A Case Study. Washington: Pew Research Center for the People and the Press.
Raines, Howell. 1996. “The Fallows Fallacy: A Plan for Turning Reporters into Lackeys.” New York Times, February 25, section 4, p. 14.
Record, The. 1996. “Demographic Analysis of Bergen/Passaic Adults Showing Major Newspaper Readership Daily and Sunday.” Photocopied. Hackensack, N.J.: The Record.
Rosen, Jay. 1996. Getting the Connections Right: Public Journalism and the Troubles in the Press. New York: Twentieth Century Fund Press.
Scarborough, Rowan. 1996. “Networks’ Election Coverage Way Down.” Washington Times, November 1, p. A14.
Star-Ledger/Eagleton Poll. 1996. “Clinton Leads Dole; Few Know Torricelli and Zimmer.” Online Rutgers University. Available: http://www-rci.rutgers.edu/ ~eaglepol/release96/elect1.htm
Stepp, Carl Sessions. 1996. “Public Journalism: Balancing the Scales.” American Journalism Review, May, p. 40.
Still, Tom, and Dave Iverson. 1995. “Still Learning After All These Years…: Lessons from the ‘We the People/Wisconsin’ Project.” Civic Catalyst, March. (Newsletter of Pew Center for Civic Journalism.)
Thames, Rick. 1995. “Covering Politics Civic Journalism Style.” Civic Catalyst, October.
Thorson, Esther; Lewis A. Friedland; and Peggy Anderson. 1997. Civic Lessons: Report on Four Civic Journalism Projects Funded by The Pew Center for Civic Journalism. Philadelphia: The Pew Charitable Trusts.
Woo, William F. 1995. “Should the Press Be an Observer or an Actor in Public Affairs?” St. Louis Journalism Review, July/August, pp. 10-12.
Zukin, Cliff. 1977. “A Reconsideration of the Effects of Information on Partisan Stability.” Public Opinion Quarterly 41 (Summer): 244-254.
_______ and Larry Hugick. 1997. “Profile of the American News Consumer.” Presented at the annual meeting of the American Association for Public Opinion Research.