Content-Type: text/html

Civic Journalism Influence On Local TV News Coverage of the 2000 Elections[1]


By
Amy Reynolds,
Assistant Professor
Indiana University School of Journalism

and

Gary Hicks,
Assistant Professor
Southern Illinois University at Edwardsville


Submitted to the Civic Journalism Interest Group,
Association for Education in Journalism and Mass Communication
April 1, 2001


Please send correspondence to:
Amy Reynolds
IU School of Journalism
Ernie Pyle Hall 200
Bloomington, IN 47405

Phone: (812) 855-1719
E-mail: <a href="/cgi-bin/wa?LOGON=A3%3Dind0109B%26L%3DAEJMC%26E%3D8bit%26P%3D10052886%26B%3D--%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D_337800291%253D%253D_%26T%3Dtext%252Fplain%3B%2520charset%3Diso-8859-1" >[log in to unmask]</a>









Civic Journalism Influence On Local TV News Coverage of the 2000 Elections


        During the first few months of 2000, several corporate owners of local television stations announced that they would incorporate a variety of initiatives into their coverage of the 2000 elections to improve the quality of political journalism on the local news.  Some of the initiatives sounded strikingly similar to political journalism approaches advocated by the civic journalism movement, including the frequent airing of "issue-check" stories, sponsoring town meetings and checking the validity of candidate advertisements through "Adwtches."
At the same time, Best Practices 2000, a new project funded by The Pew Charitable Trusts, the Corporation for Public Broadcasting and PBS, began to help local television stations develop strategies and plans for innovative election coverage of the 2000 elections.  One of Best Practices partners is Wisconsin Public Television, well-known for its innovate, civic-journalism approaches to television coverage of a variety of issues, including elections.  Another of Best Practices partners is the Pew Center for Civic Journalism.
Meanwhile, organizations like the Alliance for Better Campaigns continued to encourage television stations to follow the recommendations of the Gore Commission to voluntarily air five-minutes a night of "candidate-centered discourse" in the 30 days preceding the primary and general elections.  Although civic journalism advocates would argue in favor of giving citizens a greater voice over candidates in deciding what issues should receive election coverage (Rosen, 1993), the addition of candidate-controlled (or "centered") discourse for such a significant amount of time is an example of one new approach to election coverage that shies away from television's traditional methods of coverage.
        This seemed a strong climate in which to study whether or not a variety of principles advocated by the civic journalism movement had actually taken hold in local television news, even if none of the television stations themselves were civic journalism advocates.  This study focuses on answering the following research questions:
1.  How did various ownership and project initiatives impact the framing of election
stories that appeared on local television news stations?
2.  How did various ownership and project initiatives impact the sources used in election
stories that appeared on local television news stations?
3.  If ownership and project initiatives did impact framing and source selection, did that
impact highlight some key characteristics of the civic journalism movement, such as
avoiding horse-race coverage and promoting issues and using citizens as sources?

Ownership Initiatives and Special Projects
Belo, E.W. Scripps Company and Hearst-Argyle Television, Inc. all issued news releases that promoted their plans for improved election coverage during 2000.  Hearst-Argyle, which owns and manages 26 television stations that reach about 17.5 percent of U.S. television households, was the first to publicly declare, via news release, that it would launch an effort called "Commitment 2000" designed "to provide more comprehensive news coverage of national, state and local election campaigns" (Hearst-Argyle News Release, January 5, 2000).
According to the release, posted on the Hearst-Argyle corporate web site, "Commitment 2000" initiatives would include the following (Hearst-Argyle News Release, January 5, 2000):
ù       The airing of "Issue Check" stories in the 30-day period prior to the primary and general elections.  "These stories will address the realities of the offices that candidates are seeking and will supplement routine coverage of scheduled campaign appearances."
ù       Each day, Hearst-Argyle television stations will commit to at least one report on campaign-related subject matter during a prime newscast.
ù       "Ad Watches" will be incorporated into news coverage.  These "Ad Watches" will include advertising in print, broadcast and on the Internet.
ù       Special "Commitment 2000" web sites that would include candidate position statements, election boards and voter registration offices and special links to political parties' and related organizations' web sites.
ù       Stations will staff local town meetings and post transcripts and audio recordings on station web sites.
     At the end of the company's news release, Hearst-Argyle added a disclaimer:  "This news release contains forward-looking statements that are subject to risks and uncertainties. _ The following important factors _ could affect the future results of the Company and could cause those results to differ materially from those expressed in each forward-looking statement" including changes in economic market conditions, future regulatory action or competition from others in the same television markets (Hearst-Argyle News Release, January 5, 2000).
        In September of 2000, the company issued a second news release to announce that it would devote a minimum of five minutes' airtime each night to "political-candidate discourse" at each of its 24 news-carrying stations between 5 p.m. and 11:35 p.m. (Hearst-Argyle News Release, September 5, 2000).
        A week after the Hearst-Argyle announcement, the E.W. Scripps Company launched its election coverage project, "Democracy 2000."  E.W. Scripps television stations, which are concentrated in the top 50 U.S. markets, reach about 10 percent of U.S. households.  Unlike Hearst-Argyle, the Scripps news release directly tied its initiatives to the recommendations of the Gore Commission.  Appointed in 1997 by Vice President Al Gore to develop a set of new public interest obligations for broadcasters as a result of Congress giving them additional channels for broadcasting digital signals (including high definition television), the Gore Commission recommended that broadcasters provide free air time to political candidates during election years (Washington Post, December 16, 1998).  The panel's non-broadcast members wanted to require stations to provide the free airtime as a condition for receiving their new digital licenses, but that effort was blocked by the broadcast industry professional
s who also sat on the panel.  The non-broadcast members also wanted to require station owners to broadcast at least three hours of local news and three hours of locally-produced educational programs per week on the new channels, but that recommendation never appeared in the final commission "consensus" report (Washington Post, December 16, 1998).
        In its "Democracy 2000" program, the E.W. Scripps company noted that each station would be given the discretion of deciding which candidates and races would be eligible for the free airtime.  The company further pledged that the "Democracy 2000" campaign would also feature election-related content on each station's web site that would highlight candidate positions on issues, voter registration information and information about political jurisdictional boundaries as well as links to candidate web sites when possible (E.W. Scripps News Release, January 13, 2000).
        The company's web site provided a link to a second news release posted by the Alliance for Better Campaigns on the same day.  This release noted that the Alliance for Better Campaigns "applauded" Scripps' efforts and added that Alliance Executive Director Paul Taylor called the Democracy 2000 project "a bold step _ to offer viewers a political campaign based on issues and ideas, not money and ads" (Alliance for Better Campaigns, News Release, January 13, 2000).
        Finally, Belo Corporation, which owns 18 stations that reach 14 percent of U.S. television households, was the third large local television ownership group to publicly pledge support for new election coverage guidelines in 2000.  Belo announced that it would concentrate on improving its coverage by doing the following (Belo News Release, March 23, 2000):
ù       To broadcast three issue- or candidate-centered stories a week during the 5 p.m., 6 p.m. or late-night newscast on weekdays about 30 to 45 days prior to the election;
ù       When possible, the issue- or candidate-centered stories would include "Ad Watch" or "Issue Check" analyses of a candidate's political advertisement or position on an issue;
ù       Select Belo web sites would provide space for candidate issue statements and responses to questions and will host "E-town" meetings to allow voters to express their opinions on relevant local issues.  And, all Belo web sites will provide voter-registration information and links to "voting-related sites."
     Belo called its initiative "It's Your Time," which the company claims in 1996 was "the first instance of a commercial television group offering free air time for congressional and gubernatorial candidates" (Belo Press Release, July 28, 2000).  Belo's chairman, president and chief executive officer considered the 2000 version of "It's Your Time" as reiterating "Belo's commitment to providing our viewers with extensive information on the issues and candidates in this critical election year" (Belo News Release, March 23, 2000).  Although Belo and Hearst-Argyle had both adopted the free airtime commitment like Scripps, in accordance with the Gore Commission recommendation, neither station attributed its election project or coverage pledges to Gore Commission objectives.
        In addition to ownership group initiatives, Best Practices 2000 worked with 10 pilot stations in 2000 to develop innovative election coverage strategies.  Listed on its web site (www.BP2K.org), those strategies included sharing production capacity with public television stations, sharing research and reporting through partnerships, providing joint-coverage possibilities between commercial and public television stations and developing joint web partnerships between public and commercial broadcasters.  Some of the pilot participants included stations owned by Belo and Hearst-Argyle.

Broadcast Journalism and Elections
The earliest studies regarding the role of television in the electoral process emerged at roughly the same time that television itself was taking its first steps toward becoming a mass medium.  Television's debut in the political arena occurred, though in a very limited way, during the presidential election of 1948.  It was during the 1952 election that the first television campaign commercials appeared, and the national political conventions were first broadcast on television.  During this time, many of the "classic" studies on voting behavior (Berelson et al., 1954; Campbell et al., 1954,1960; Katz and Lazarsfeld, 1955) concentrated on the media's role in influencing public political opinion.  Television election coverage, though not the focus of any of these studies, was folded into the research mix.  Consequently, the first look at televised election coverage became tied to the effects model of media research.  Not surprisingly, these earliest studies made only minor reference
 to television, and did not discern any special powers belonging to the new medium.
As the reach of television advanced throughout the 1950s and '60s, studies (Kraus, 1962; Greenberg and Parker, 1965; O'Donnell, 1966) began to confirm the growing influence of television on the electoral process.  Some studies looked at how people use television during an election campaign.  Blumler and McQuail (1969) showed that people use television less for guidance (a role earlier voter studies attributed to newspapers) and more for "surveillance" needs to keep up with the main issues and changing news.  It is important to note that even into the 1960s, research into television election coverage still focused on its effects on political opinion and action. Lang and Lang (1968) examined the 1952 nominating conventions and found that television coverage tended to dramatize the conventions.  They also found an incongruity between how people at the convention and those who watched the convention over television perceived the convention proceedings.  McCombs and Shaw (1972) studied the agenda-setting effect of the media in the 1968 presidential election.  Their work has been followed by numerous researchers examining the impact of mediated messages on the public.  Some (Iyengar, Peters, & Kinder, 1982; Hill, 1985) have used the agenda-setting premise to study the specific effect of televised news coverage on political issues.   The content of television election coverage and the processes by which television covers elections were issues yet to be studied.
Criticism over perceived bias in how media covered the Vietnam War and Watergate, together with the increased interest in the people and institutions behind the news brought about by works such as Timothy Crouse's 1973 The Boys on the Bus, ushered in research in the 1970s that delved more specifically into content and processes of television news during election cycles.  Hofstetter (1978) conducted a cross-media study of news bias in coverage of the 1972 presidential election, evaluating the tone of coverage of parties and candidates.  Studies in this period contributed to the development of content analysis techniques specific to television (Frank, 1973).
The last two decades have shown growing scholarly interest in issues regarding influences on television news content, the roles of media workers, and analyses of the images appearing on television newscasts. Shoemaker and Reese (1991) included television in their examination of media content and influences and found that many answers, "lie in such factors as the personal attitudes and orientations of media workers; professionalism; corporate policies; corporate ownership patterns; the economic environment; advertisers; and ideological influences."
Ansolabehere, Behr, and Iyengar (1993) found that the two most newsworthy aspects of political campaigns have become the horse race and the character issue.  The authors develop their theory of how a two-step flow of communication between political actors and the media has actually distorted the behavior of politicians in relation to both media and to the citizenry.
In one study of the images that make up an election newscast, Barnhurst and Steele (1998) examined television news coverage of campaigns from 1968 to 1992 and found that over the years the number of times the correspondent was shown during a report grew more frequent, and the scale of the journalist's image grew significantly larger.
In addition to image, recent scholarship has also been concerned with campaign discourse.  Jamieson and Waldman (1998) looked at television news coverage as part of the Campaign Discourse Mapping Project.  The authors compared coverage with what they deemed commonplace assumptions about campaign coverage and concluded that campaign news has not, contrary to popular opinion, become more negative, nor does campaign news feature more instances of candidate "attacks" on their opponents.
Larson (1999) examined evening television newscasts during the 1996 presidential election campaign to see how public opinion was covered.  The author found that the "horse race" coverage did not extend to "people-on-the-street" features where people had an opportunity to voice opinions on matters of public importance.  Shaw (1999) studied news media favorability in his coverage of candidate events in presidential elections.  Findings included that favorability in media coverage helps to drive changes in voters' preferences.
Dover (2000) argues that what he terms "mediated incumbency" has been a better indicator of the outcome of recent presidential elections than has the long-held notion of the importance of partisanship.  Through an examination of the relationship between the president and television media during the 1996 election, the author came to define mediated incumbency as the relationship between the strength of the incumbent and the way in which television news understands and interprets that strength.

Civic Journalism
        The concept of civic journalism is readily understood today - the movement has been in practice for the past 15 years.  Many proponents of the movement would argue that in recent years civic journalism has flourished (Rosen, 1999).  The central argument put forth by supporters of civic journalism is that it encourages the press to foster public discourse within a democracy, which is part of the role the press was designed to fill under the First Amendment.  In Mixed News, a book that features prominent media scholars arguing both for and against the civic journalism philosophy, Ted Glasser and Stephanie Craft (1997) observe,
Whether it is being celebrated or condemned, there is little disagreement about what, basically, (civic) journalism expects from the press.  It expects the press to recognize its role in fostering public participation and public debate.  It expects the press to acknowledge the decay of democracy, usually measured as a decline in voter turnout, requires a commitment from the press to improve the conditions for self-governance.  It expects the press to abandon the traditional and still dominant view of journalism, which holds that the newsroom must stand detached from, and disinterested in, the affairs of the community (p. 121).

        Hundreds of civic journalism projects throughout the past decade have attempted to provide examples of what civic journalism in practice might look like (Rosen, 1999).  Most of these projects have appeared in newspapers across the country and initially they focused on elections and political reporting.  Today, the civic journalism projects that started in print have become collaborative efforts with local television stations, both public and commercial, and often with radio stations as well.  Topics moved well beyond the scope of politics, although civic journalism is still strongest in the arena of election coverage.
In arguing for the application of the practice of civic journalism in What are Journalists For?, Jay Rosen notes that network television news has missed many opportunities to reimagine the audience as citizens.  He points out that in 1996, Ted Koppel and Nightline packed up and left the Republican National Convention because the event was devoid of any real news.  "Here was Ted Koppel, perhaps the most gifted television journalist of his time, host of one of the most intelligent and innovative news programs in the mediums history _ declaring that neither he or his staff could think of anything useful to say or do as the Republican Party gathered to make its statement to America" (p. 112).  Rosen adds, "If nothing surprising was anticipated _ all the more reason to surprise the audience with a fresh approach.  _ To explore (this), Koppel would have had to imagine his audience differently: not as people awaiting "surprise" from a convention that was designed to be unsurprising, but
as citizens hoping for an intelligible politics - one that spoke to their desire to understand _" (p. 116).  Rosen has suggested that any media - including network and local television news -- could easily support the civic journalism movement by simply changing their mode of thinking (Rosen, 1999).
Critics of civic journalism abound at the network television level, arguing that the movement disavows objectivity when it suggests that it is impossible for journalists to be detached and objective about public life.  They add that a loss of objectivity would encourage an active partisan bias on the part of journalists (Fallows, 1996).  More often than not, local commercial television news has remained indifferent to the civic journalism movement with the exception of the occasional partnership projects and with the exception of some PBS stations (like Wisconsin Public Television).
One example of a successful civic journalism project on local television (without a partnership) is found in a study of the 1996 National Issues Convention[2] in Austin, Texas.  Reynolds (1999) found that one local station's attempt to employ civic journalism practices resulted in more issue-oriented stories, the elimination of horse-race polls, more positive political stories and a greater focus on providing voter information.

Method
  Although civic journalism has traditionally been project-driven, many news outlets that support the philosophy have been working to incorporate civic journalism ideals into everyday coverage of the news.  And although local television news has not been at the forefront of this effort, the ownership and project initiatives outlined earlier seemed like a good context in which to study civic journalism's general influence on election content on local television news.  As already noted, the goal of this study is to explore whether or not a variety of principles advocated by the civic journalism movement had actually taken hold in local television news, even if none of the television stations themselves were civic journalism advocates.  In one sense, the stations studied here are advocating special election projects.  But, in another sense these projects are really simply guides to producing better election coverage and political journalism on television overall.  This would seem to
fit the civic journalism philosophy if audiences are treated like citizens and stories become more issue-focused.
For this study, 15 local television stations' prime newscasts in 13 different markets (either 5:30, 6 or 6:30 p.m. Monday through Friday or 10 or 11 p.m. on weekends) were recorded 21-days prior to the 2000 general election.  Stations were selected based on ownership (a mix of Belo, E.W. Scripps, Hearst-Argyle and ownership groups making no pledge to change coverage) and the ability to find a volunteer newscast recorder in a given market.  Geographic diversity was another consideration.  The entire sample comprised television stations in the following markets - Austin, Texas; Cleveland; Houston; Indianapolis; Louisville; Oklahoma City; Orlando; Philadelphia; Phoenix; Seattle; St. Louis; Tampa; and Tulsa.  Because such a substantial amount of data was collected, for this paper only a portion of the sample was analyzed.  The analysis that follows included nearly one-third of the complete sample of newscasts with nearly equal representation of ownership as well as participation in th
e Best Practices 2000 initiative.  Stations included in this preliminary analysis include KHOU in Houston (Belo), WLKY in Louisville (Hearst-Argyle), WESH in Orlando (Hearst-Argyle), WTXF in Philadelpia (News Corporation/Best Practices), KNXV in Phoenix (Scripps) and KING in Seattle (Belo and Best Practices).  In addition, only one week of newscasts, Oct. 31, 2000 through Nov. 6, 2000 were coded.  The unit of analysis was any story about the 2000 election.  Coders included all issue-related election stories, candidate focused-election stories, local, state and congressional election stories as well as stories about the presidential election.  Any story that mentioned the 2000 election or something clearly tied to it was included in coding.
        To answer the first research question, "How did various ownership and project initiatives impact the framing of election stories that appeared on local television news stations?" coders examined story frames.  Entman (1993) writes that framing essentially involves selection and salience.  He says, "to frame is to select some aspects of a perceived reality and make them more salient in a communicating text (p. 52)."  Across numerous mass media studies, frames are operationalized in a variety of ways.  Gamson (1988) finds frames to be the focus of concern that "suggests what the issue is about." In essence, a frame can be considered a story's dominant focus.  Gamson & Modigliani (1989) describe the use of exemplars, catch phrases, and depiction as devices, which identify the frame or the "central organizing idea or story that provides meaning."  After an exhaustive review of the oft-cited literature on framing, Tankard et. al. (1991) conceptually defined a frame as "a central organizing idea for news content that supplies a context and suggests what the issue is through the use of selection, emphasis, exclusion and elaboration."
        The category system of possible frames for this content analysis was based largely on an extensive examination of the existing literature about how to categorize coverage of presidential elections (Patterson, 1980; Patterson & McClure, 1976; Robinson & Sheehan, 1983; Shaw & McCombs, 1977; Semetko, Blumler, Gurevitch & Weaver, 1991; and, Weaver, Graber, McCombs & Eyal, 1981) and included the following:
        ù  Partisan/campaign activities:  This included stories about specific campaign activities or aspects of the campaign such as endorsements, advertisements, campaign tactics, debates, plans for debates, rallies, fundraising, the primaries or caucuses as events (but not strategic implications).
        ù Political consequences/horserace:  This included stories with a focus on the analysis of political parties and candidates, the political consequences of events, decisions, the primaries, caucuses, etc.  Focus is on the horserace aspect of the campaign and often includes opinion polls as sources.
        ù  Candidate image:  This included stories with a focus on the discussion of candidate qualifications, leadership, style, background (personal and professional), age, etc.  Also included here were discussions of a candidate's family or friends and their impact on the candidate's image.
        ù  Policy issues:  This included stories about discussions of specific or general issues that are a focus of the campaign (for example, the environment) or of candidates' stands on specific issues.
        ù  Voter issues:  This included stories that focused on get-out-the-vote efforts, voter turnout, voter registration, early voting participation and/or issues related specifically to voting (training volunteer workers, counting ballots, etc.).
        ù  Adwatch:  This included stories that provide a validity check on the accuracy of a candidate's advertising.  The focus could be on one candidate's ad, multiple candidate ads or one party's ads.  It did not include what many of the ownership news releases called "Issue Checks."  These stories were coded as policy issue stories.
        To answer the second research question, "How did various ownership and project initiatives impact the sources used in election stories that appeared on local television news stations?" coders conducted a source analysis.  The source analysis included any story with a sound bite but did not include reporter standups or live shots.  Source categories included the following:
ù       Candidates -- national, state or local;
ù       Staff/surrogates -- individuals who are clearly speaking for the candidate or his/her campaign;
ù       Party officials -- individuals speaking for a particular party as a whole, but do not have the authority to speak directly for any individual candidate;
ù       Media -- any sound bite from a member of the media, but this does not include reporter standups and live shots;
ù       Event coordinators -- individuals directly tied to organizing a political event;
ù       Partisan participants -- someone who is clearly identified as a Republican, Democrat, Green Party or other party supporter.  He or she has no other title than that of partisan participant or partisan voter and the sound bite does not come as a result of a random stop on the street;
ù       Person in street -- a sound bite from a random person in the street or at the mall, etc.;
ù       Non-party officials -- a sound bite from an official who is not speaking or acting in any official capacity specifically to represent a particular party;
ù       Experts -- someone who is considered knowledgeable in a particular area and is not tied to the campaigns in some way;
ù        Special interests -- someone who is considered knowledgeable in a particular area but has a clear agenda to promote;
ù       Citizens -- someone who is considered a knowledgeable voter and who might be engaged in a town meeting or some other organized political function for voters.
       To complete this analysis, coders also examined whether or not polls were used as sources in election stories (and the frame or context of the poll information), whether or not stations actively promoted and referred viewers to their web election coverage, story tone (positive, negative or neutral) and whose voice dominated in reporter-produced packages (the reporter, sources or equal).  Intercoder reliability is 93.4% (Holsti).

Findings
        Coders examined 112 stories from the six television stations during the week of Oct. 31, 2000 through Nov. 6, 2000.  In terms of number of stories, KHOU in Houston aired 18 stories (16.1% of the sample); WLKY in Louisville aired 14 stories (12.5% of the sample); WESH in Orlando aired 21 stories (18.8% of the sample)[3]; WTXF in Philadelphia aired 19 stories (17% of the sample); KNVX in Phoenix aired 18 stories (16.1% of the sample); and, KING in Seattle aired 22 stories (19.6% of the sample).  Within these 112 stories, the most frequently viewed frame was partisan activities (44 stories/39.3%), followed by the voter issues frame (25 stories/22.3%).  The third most common frame was the political consequences/horserace frame (19 stories/17%).  In all, these three frames accounted for 88 of the 112 total stories, or 78.5% of the sample.  Table 1 shows the results of a crosstabulation between station and primary frame.  The results are significant using the p<.10 standard.
Table 1:  Frame Use by Station
Station                                 Frames
                        Partisan Activities     Horserace     Image     Issues     Voter Issues
KHOU (Houston)                  8               3         0        1            6
WLKY (Louisville)               7               0         1        4            2
WESH (Orlando)                  7               7         1        2           4
WTXF (Philadelphia)             9               2         2        2           4
KNXV (Phoenix)                  3               1         0        6           7
KING (Seattle)                  10              6         2        1           2
Total (N=112)                                  44               19        6       16          25


(Chi-Square = 34.633; d.f. =25; p<.09)


After the framing comparison with all of the stations, each ownership or project group was evaluated separately for significant use of a particular frame.  The results were not significant for Belo, Scripps, Hearst-Argyle or Best Practices.
        The source analysis showed similar results.  Of the 112 total stories included in the analysis, only 69 contained a sound bite, which was subsequently coded for the source variable.  In terms of frequency, the most commonly used sources were the candidates themselves.  They accounted for 49.3% of the sources used in the election stories with sound bites.  The second most common sources were random people on the street.  They accounted for 17.4% of the sources used in election stories. Table 2 shows the results of a crosstabulation between station and sources.  The results are significant using the stronger p<.05 standard.
Table 2:  Source Use by Station
Station                                 Sources
                                     Candidates  Staff   Party Officials  Partisans  Person in Street  Experts  Special Interest  Other
KHOU (Houston)          2         1             0          1            6               0       0            0
WLKY (Louisville)       4         0             0           1           1               1       4            0
WESH (Orlando)          5         0             0          1            3               0       1            1
WTXF (Philadelphia)     10       0              1          0                  0         1       1            2
KNXV (Phoenix)          3         1             0          2             0              0       1            2
KING (Seattle)          10       1              0          0             2              0       0            0
Total (N=69)            34        3             1          5            12              2       7            5


(Chi-Square = 52.980, df=35, p<.02)

Like with the framing analysis, the source analysis was broken down to analyze each ownership or project group.  They were evaluated separately for significant use of a particular source.  Again the results were not significant for Belo, Scripps, Hearst-Argyle or Best Practices, although Belo (which includes KING in Seattle and KHOU in Houston) approached significance with its pattern of relying most heavily on candidates and people in the street as sound bite sources (p was <.12, but not statistically significant).
        Finally, each ownership group was analyzed separately with several different variables, including whether or not each station promoted their election coverage and projects on their web sites (or sent viewers there for more information), whose voice dominated in reporter stories, the tone of election stories and whether or not polls were used as sources.
        Although a combined analysis shows that the stations used polls as sources in only about 15% of stories, both a collective and individual ownership group analysis did not show significant results.  Only the Best Practices stations showed significance when analyzing whose voice dominated in reporter-produced packages.  These two stations (KING in Seattle and WTXF in Philadelphia) aired 41 reporter packages, 17 of which relied most heavily on the journalists' voice.  This would fit the traditional and not civic journalism model and was significant at p<.01 level.  No significant results were found for story tone.  Neutral stories were dominant across all stations.
        Finally, Belo stations did promote and support their web site election coverage on a regular basis (p<.01 for Belo).  Only 25% of all stories in the analysis promoted a station's web coverage.  Of the 40 stories that aired on Belo stations (KHOU in Houston and KING in Seattle), more than 50% promoted the station's web coverage.  On the flip side, of the 35 stories that aired on Hearst-Argyle stations (WLKY in Louisville and WESH in Orlando), only 4 promoted web coverage (this lack of promotion was significant at p<.08).

Discussion
        The results of the preliminary sample analysis are mixed.  Research question one asks how various ownership and project initiatives impacted the framing of election stories that appeared on local television news stations.  The findings show that collectively stories were framed predominantly to feature partisan activities.  This is not a departure from traditional local television news election coverage.  Taken separately, no ownership group showed any significant difference in the ways that they framed election stories.  The answer to research question one then seems to be "no" -- the ownership initiatives did not substantially alter election coverage.
        Belo had pledged to broadcast three issue- or candidate-centered stories a week during the 5 p.m., 6 p.m. or late-night newscast on weekdays.  It is certainly likely that these stations broadcast additional stories on newscasts not analyzed here.  That could impact these findings.  It is also worth noting that candidate-centered stories, depending on how one defines "candidate-centered," could include stories about a candidate's election stops in a particular town.  For this analysis, such stories would fall into the partisan activities frame.  But, it is possible that in the minds of television news directors or producers, a candidate-centered story might include a simple stop on the campaign trail.  This requires further investigation.
        Hearst-Argyle pledged to air "Issue Check" stories that would "address the realities of the offices that candidates are seeking and will supplement routine coverage of scheduled campaign appearances."  The company clearly stuck to the routine coverage of scheduled campaign appearances.  However, the appearance of "Issue Check" stories  (Only 2 of the 19 stories that aired on WESH fell into the issues category and only 4 of the 14 stories that aired on WLKY were coded as issue-focused) was only supplemental to a small degree (and a non-significant degree at that).
Research question two asked how various ownership and project initiatives impacted the sources used in election stories that appeared on local television news stations.  Again, the findings show that collectively, stories that contained sound bites (69 of 112) relied most heavily on the candidates themselves as sources, followed by people on the street.  This is also not a departure from traditional local television news coverage.  None of the stations pledged any special treatment of sources (besides offering candidates free air time, which was not included in this analysis), but the general idea that the stations would "improve" election coverage is what prompted this research question.  As previously noted, Larson (1999) found that when people on the street were used as sources, they typically voiced opinions on important issues rather than on the horserace aspect of the campaign.  So, despite the fact that candidates dominated election discourse across all of the stations, the fact that the station's collectively pursued people on the street as the second most used category of sources may suggest some minimal effort to connect the candidates and issues to citizens and voters.
Finally, the third research question attempts to find out if civic journalism ideals are creeping into some local television newsrooms, specifically those where ownership groups are showing some awareness for the need to improve and change existing election coverage.  It asks, "If ownership and project initiatives did impact framing and source selection, did that impact highlight some key characteristics of the civic journalism movement, such as
avoiding horse-race coverage and promoting issues and using citizens as sources?"  As already discussed, the answer in terms of framing and sources is "no."  Even the addition to the analysis of use of poll data, story tone, dominant voice in reporter packages and whether or not stations successfully used the web as an additional resource lean strongly toward suggesting that these stations, despite the promotion of better election coverage, stuck mostly to traditional models and did not incorporate civic journalism ideals into their daily election coverage.

Conclusion
        Although this preliminary analysis does not show much support for the notion that civic journalism ideas have impacted some local television newsrooms, additional research is needed.  Not only will this preliminary analysis expand to include all of the stations in the original sample as well as the complete 21-day period prior to the election, but it will also include several other analyses that will provide more concrete answers to these questions (and many other questions as well).  Interviews with all of the station news directors is planned to find out how they conceptualized their election coverage.  This includes areas like defining terms (for example, candidate-centered stories) and asking whether or not these newsroom managers are aware of civic journalism and its approaches to election coverage.  Through discussions with news directors and through requests for producing logs, it might also be possible to find out what kinds of stories were exempted from the analysis of o
nly one prime newscast per night.
In addition, during the 21-day period prior to the election, web site information, links and stories for all of the stations included in the analysis was collected.  It is hoped that by combining the analysis of the broadcast stories and station initiatives with content produced for the web sites, a more complete picture of coverage will be gained.
        The civic journalism movement has gained respect and momentum during the past decade.  Although it is most widely accepted as practice in the printed world, local television news stations are starting to show some interest in the movement.  It is valuable to study individual projects and the effects of civic journalism, no matter what medium is used.  But, the goal of this preliminary study as well as the expanded study is to find ways that civic journalism may be or could be impacting daily coverage of important issues that impact communities across the country.  Once researchers can establish a connection between civic journalism ideals and content, they can next begin to look at the impact this may have on voters and citizens.

 References

Alliance for Better Campaigns.  "Alliance's Political Air Time Proposal Gets a  Foothold:
Scripps Becomes First Major Group to Commit to Air Nightly Candidate Segments in
Month Before 2000 Election."  News Release, Jan. 13, 2000.

Ansolabehere, S. et. al.  The Media Game:  American Politics in the Age of Television.  New
York:  Macmillan, 1993.

Barnhurst, K.G. and Steele, C.A.  "Image-Bite News:  The Visual Coverage of Elections on U.S.
Television, 1968-1992."  Harvard International Journal of Press/Politics, 2, 1997.

Belo Corporation.  "Belo Offers Free Airtime to Political Candidates for the Third Election Year and Expands Political Coverage."  News Release, March 23, 2000.

Belo Corporation.  "Belo Extends Coverage of 2000 Elections With Online Webcast Delivered
by Yahoo."  News Release, July 28, 2000.

Berelson, B. et al.  Voting:  A Study of Opinion Formation in  a Presidential Campaign.
Chicago:  University of Chicago Press, 1954.

Blumler, J. G. and McQuail, D.  Television in Politics:  Its Uses and Influence.  Chicago:
University of Chicago Press, 1969.

Campbell, A. et al.  The Voter Decides.  Evanston, Ill.: Row, Peterson, 1954.

Campbell, A. et al.  The American Voter .  New York:  Wiley, 1960.

Dover, E. D. The Presidential Election of 1996:  Clinton's Incumbency and Television.
Westport, CT:  Praeger, 1998.

Entman, R. "Framing:  Toward Clarification of a Fractured Paradigm."  Journal of Communication, 43, p. 51-58, 1993.


E.W. Scripps Company.  "Scripps Making Free Airtime Available to Candidates."  News
Release, Jan. 13, 2000.

Fallows, J. Breaking the News:  How the Media Undermine American Democracy.  New York:
        Pantheon, 1996.

Farhi, P.  "Panel Disagrees on Air Time Requirement for a Digital TV License."  Washington
Post, Dec. 16, 1998, p.A14.

Frank, R.S.  Message Dimensions of Television News.  Lexington, Mass.:  Lexington Books,
D.C. Heath, 1973.

Gamson, W., & Modigliani, A. (1989). "Media Discourse and Public Opinion on Nuclear Power: A Constructionist
        Approach." American Journal of Sociology, 95, 1-37, 1989.

Gamson, W. "A Constructionist Approach to Mass Media and Public Opinion." Symbolic Interaction, 11, 161-174,
        1988.

Glasser, T. & Craft, S. "Public journalism and the prospects for press accountability," in J. Black
(Ed.), Mixed news: The Public/Civic/Communitarian Journalism Debate.  Mahwah, N.J.:
        Lawrence Erlbaum Associates, 1997.

Greenberg, B.S. and Parker, E.B., eds. The Kennedy Assassination and the American Public.
Stanford:  Stanford University Press, 1965.

Hill, D.B.  "Viewer Characteristics and Agenda-Setting by Television News."  Public Opinion
Quarterly, 49, 1985.

Hearst-Argyle Television, Inc.  "Hearst-Argyle Launches Commitment 2000, A Group-Wide
Effort to Strengthen Political News Coverage."  News Release, Jan. 5, 2000.

Hearst-Argyle Television, Inc.  "Hearst-Argyle Television Stations to Devote 5 Minutes Airtime
Daily to Candidate Discourse as Part of its 'Commitment 2000.'"  News Release, Sept. 5,
2000.

Hofstetter, C.R.  "News Bias in the 1972 Campaign:  A Cross-Media Comparison."  Journalism Monographs, 58, 1978.

Iyengar, S. et al.  "Experimental Demonstrations of the "Not-So-Minimal" Consequences of
Television News Programs."  American Political Science Review, 76, 1982.

Jamieson, K.H. and Waldman, P. "Mapping Campaign Discourse:  An Introduction."  American
Behavioral Scientist, 40, 1997.

Katz, E. and Lazarsfeld, P. Personal Influence.  New York:  Free Press, 1955.

Kraus, S. ed. The Great Debates -- Background, Perspective, Effects.  Bloomington, Ind.:
Indiana University Press, 1962.

Lang, K. and Lang, G. E.  Politics and Television.  Chicago: Quadrangle Books, 1968.

Larson, S. G. "Public Opinion in Television Election News:  Beyond Polls."  Political
Communication 16, 1999.

McCombs, M.E. and Shaw, D.L. "The Agenda Setting Function of Mass Media."  Public
Opinion Quarterly, 36, 1972.

McCombs, M. & Reynolds, A. (Eds.).  The Poll with a Human Face:  The National Issues
Convention Experiment in Political Communication.  Mahwah, N.J.: Lawrence Erlbaum
Associates, 1999.

O'Donnell, K.P. "TV in the Political Campaign."  Television Quarterly, Winter, 1966.

Patterson, T.E. & McClure, R.D. The unseeing eye:  The Myth of Television Power in National Politics.
        New York:  Putnam, 1976.

Patterson, T.E. The Mass Media Election:  How Americans Choose Their President.  New York:  Praeger, 1980.

Reynolds, A.  "Local Television Coverage of the N.I.C," in McCombs, M. & Reynolds, A.
        (Eds.).  The Poll with a Human Face:  The National Issues Convention Experiment in
        Political Communication.  Mahwah, N.J.: Lawrence Erlbaum Associates, 1999, p. 113-
        131.

Robinson, M.J. & Sheehan, M.A.  Over the Wire and on TV:  CBS and UPI in Campaign '80.  New York:
        Russell Sage Foundation, 1983.

Rosen, J.  "Community connectedness passwords for public journalism."  Poynter Report.  St.
        Petersburg: The Poynter Institute for Media Studies, 1993.

Rosen, J.  What Are Journalists For?  New Haven, Conn.: Yale University Press, 1999.

Semetko, H.A., Blumler, J.G., Gurevitch, M. & Weaver, D.H.  The Formation of Campaign Agendas:  A
        Comparative Analysis of Party and Media Roles in Recent American and British Elections.  Hillsdale, N.J.:
        Lawrence Erlbaum Associates, 1991.

Shaw, D.R.  "The Impact of News Media Favorability and Candidate Events in Presidential
Campaigns."  Political Communication, 16,       1999.

Shaw, D.L. & McCombs, M.E.  The Emergence of American Political Issues:  The Agenda-setting Function
        of the Press.  St. Paul, MN:  West, 1977.

Shoemaker, P.J. and Reese, S. D.  Mediating the Message:  Theories of Influences on Mass
Media Content.  White Plains, N.Y.:  Longman, 1991.

Tankard, J.W. et. al. Media Frames:  Approaches to Conceptualization and Measurement.  Association for
        Education in Journalism and Mass Communication.  Boston, 1991.

Weaver, D., Graber D., McCombs, M. & Eyal, C. Media Agenda-Setting in a Presidential Election:
        Issues, Images and Interest.  New York:  Praeger, 1981.

















[1]  The authors wish to thank the Indiana University School of Journalism, whose grant to promote research, funded this study. For assistance with coding, the authors thank Rasha Abdel-Hami Kamhawi, Courtney Irwin, Lesley Meier-Shore and Brian Schultz.  For assistance in recording newscasts, the authors thank Amy Franzini, Mark Hicks, Tracy Huber, Jennifer Jones, Hilary Karasz-Dominguez and Kari Kron.
[2] Political scientist Jim Fishkin organized a deliberative poll in the United States that became known as the National Issues Convention (N.I.C.).  The event brought 459 randomly selected delegates to Austin, Texas, to discuss issues with each other, with experts and with politicians.  In November 1995, the National Opinion Research Center at the University of Chicago conducted field interviews with 600 randomly selected American adults and invited them to participate in the deliberative aspect of the experiment in January 1996.  After the first poll, those who accepted the invitation to Austin received non-partisan, specially prepared briefing materials about three specific issues -- the family, the economy and foreign affairs.  "Instead of a top of the head impression from shrinking sound bites and headlines, people will really have a good chance to get good information, hear opposing points of view and come to a considered judgment," Fishkin told the Public Broadcasting Servi
ce.  "So this represents what the country would think if it were really engaged in the issues." As Jim Lehrer, the U.S. moderator of the televised portions of the N.I.C. said during the final broadcast, "It was an experiment aimed at exploring the connection between information and deliberation and the way voters think and vote."  For a detailed account of the National Issues Convention, see McCombs, M. & Reynolds, A. (Eds.).  (1999) The Poll with a Human Face:  The National Issues Convention Experiment in Political Communication.  Mahwah, N.J.: Lawrence Erlbaum Associates.

[3]  Due to newscast recorder error, no newscast was recorded for WESH on Nov. 4 or 5.  The total stories aired for this station would have certainly been higher with the inclusion of the Nov. 4 and 5 broadcasts.