Issues and Agendas: The Case of Wichita, Kansas Revisited
Introduction
Agenda-setting explains the influence the media have on how individuals
view situations and ideas they encounter through the media. The basis of
this theory is that when the media agenda (what is reported by the news
media) is presented to the public, a strong correlation exists between the
amount and kind of attention paid to a particular issue by the media and the
level of importance assigned to that issue by the public. Agenda-setting
theory implies a relationship between decisions about an issue in the media
and the beliefs the public holds about the issue's significance.
Aside from the pros and cons of academic debate regarding the existence
and power of media agenda- setting, producing news that addresses the
concerns of citizens continues to be a traditional goal among news media
organizations across the country. Furthermore, the goal of a new trend of
news reporting, public journalism, also known as civic or community-based
journalism, is to "reconnect" citizens with their newspapers, their
communities, and the political process (Shepard, 1994, p. 29). The aim is to
bring the public back into journalism. Public journalism requires the
journalist to go to the citizens to get the "people's frame" on issues and
then report the news from the people's vantage point.
This study is a replication with extension of primary research conducted
by Sharon Iorio and Susan Huxman prior to the 1992 elections (Iorio & Huxman,
1994) and of a thesis produced by one of their students involved in the
original study (Culot, 1992). Data for this study were obtained, in part,
by a 1996 election issues study funded by the Kansas Public Finance Center,
The Knight-Ridder Foundation, The Wichita Eagle, and Wichita State
University's Elliott School of Communication. The purpose of this research
is to examine any correlation between residents' concerns about "the most
important problems facing people today" and actual media content during the
period studied and compare the 1996 results to the results of the 1992 study.
The importance of studies on agenda-setting is immense. They examine
the power of the news media and how the issues journalists choose, or do not
choose, to report affect our everyday lives. In 1963, Bernard Cohen wrote
that "the media may not tell us what to think, but they are stunningly
successful in telling us what to think about" (Cohen, 1963, p. 13). Thirty
years later, Maxwell McCombs challenged Cohen's idea by stating that research
now suggests "the media not only tell us what to think about, but also how to
think about it, and consequently, what to think!" (McCombs, 1993, p. 62).
The next chapter will give an overview of research on both agenda-setting and
public journalism since their beginnings.
A Review of Literature
To lay a foundation on which to base this study, a review of literature
on agenda-setting theory and public journalism is necessary. This review
will begin by examining the history of agenda-setting research, show the
relationship between agenda-setting and public journalism, and analyze the
role media plays in agenda-setting. Next this review will examine attempts
at public journalism and point out gaps in research studies that exist
between the two research areas.
Agenda-Setting
Since the 1970s, mass communication research focusing on public issues,
events, and personalities in the media has involved more than just factual
details. This approach, known as agenda-setting, also weighs the relative
importance and prominence of the various elements in each day's news.
"Agenda setting is a theory about the transfer of salience, both the salience
of objects and the salience of their attributes" from the media to the public
(McCombs & Shaw, 1993, p. 62).
Media agenda-setting theory states that the media coverage of an issue
affects how citizens "think about" issues they consider important. The
results of agenda-setting research are mixed, depending on the issues studied
and the research design employed. The work of McCombs and Shaw provides the
empirical foundation of this theory (Lowery & DeFleur, 1995).
Before McCombs and Shaw's 1972 landmark study on agenda-setting, other
researchers pondered the connection between media messages and public
perception. As early as the 1940s research supporting the agenda-setting
effect was being performed. Berelson (1941) found that the mass media
influence voters' judgments about the importance of issues. Evidence of an
agenda-setting function of the press and its relation to cognitive aspects of
attitudes is also found in a 1948 study of Erie County (Lazarsfeld, et al.,
1948).
Lang and Lang (1966) observed that, "the mass media force attention to
certain issues. They build up public images of political figures. They are
constantly presenting objects suggesting what individuals should think about,
know about, have feelings about" (p. 468). In 1969, Arora and Lasswell
recognize that "the mere fact of publication begins to modify current
perspectives . . ." (p. 11). In a Charlotte Observer study of the 1970
election (November 8, 1970) a voter gave additional evidence of agenda-
setting when she said she was worried about "whether the economy is going to
collapse, law and order, and the things that come up in the papers" (Kline &
Tichenor, 1972, p. 183).
Based on Lang and Lang's (1966) research, McCombs and Shaw (1972)
hypothesized that the mass media "set[s] the agenda for each political
campaign, influencing the salience of attitudes toward political issues" (p.
177). This landmark study examined agenda-setting effects during the 1968
presidential election by comparing issues voters felt important with the
media message to which these voters were exposed. A random sample was used
to select 100 Chapel Hill, North Carolina registered voters from five
demographically representative precincts. Among undecided voters high
correlations were found between what voters cited as major issues and the
attention of the mass media to those issues. Although the correlation
between media messages and issues cited was not strong, the respondents'
choices of important election issues mirrored issues covered in the media.
In addition, while the three presidential candidates placed widely divergent
emphases upon different issues, the judgments of those voters seemed to
reflect the composite of the mass media campaign coverage.
Other research (Weaver, Graber, McCombs & Eyal, 1981; Becker & Dunwoody,
1982; Gandy, 1982) has supported the agenda-setting function, but none has
been more notable than that of David Weaver. Weaver (1984) noted that "the
press does not reflect reality, but rather filters and shapes it" (p. 682).
This discovery supported McCombs and Shaw's original finding. Weaver also
found that "concentration by the media over time on relatively few issues
leads to the public perceiving these issues as more salient or more important
than other ones" (Weaver, 1984, p. 682). In his review of a decade of
agenda-setting research, Weaver concluded that the agenda-setting hypothesis
is true. He notes, however, that the results of these studies vary depending
on the design of the study, the type of audience data, and the measurement of
technique.
Public Journalism
The move toward public journalism in Wichita, Kansas, can be traced as
far back as the 1920s, when Victor Murdock, then editor of The Wichita Eagle,
attended Sunday church service at a different congregation each week (Iorio,
1996b).
Murdock sat through the Sunday-morning sermons to hear what preachers had
to say about morality and the issues of the day. After services, he lingered
on church lawns, shook hands and listened to what parishioners had to say
about their personal concerns. The complaints and praise he heard, the
worries people shared--everything he experienced--formed a backdrop for the
daily information flow he supervised (p. 1).
Murdock's goal was to gain the perspective of the people in the
community on issues and then report on those issues with a similar frame of
reference. In this way news was reported in a fashion that related issues to
individuals' concerns. Recently, like Murdock, Jon Roe, an Eagle team leader
and reporter, took the time to sit in on a Sunday school class of young
parents in a West Wichita church. During that half hour class, he came up
with five viable new story ideas (Harwood, 1996). Although visiting churches
on a regular basis is not a common reporting practice anymore, The Eagle has
realized that going beyond the traditional routines of news gathering to get
the type of in-depth coverage that provides a forum for public concerns is
important.
In essence, public journalism maintains that
journalists should think of themselves as being part of a society, sharing
responsibility for the long-term health of culture and democracy, rather than
being strictly detached observers whose duty is only to point out that things
are going wrong (Fallows, 1996, p. 1).
Both Murdock and Roe were participating in public journalism. Both went out
into the community to find out about and report on citizens' concerns.
Today, part of the Eagle's problem, and that of other news organizations
across the country, is that population growth and readership are declining.
"Citizens are shunning public life" and "citizens who don't participate have
little need for news" (Hoyt, 1995, p. 28). The staff at The Wichita Eagle is
trying to keep its finger on the pulse of a metropolitan, Midwestern city and
surrounding communities of more than 400,000 people. Editor Davis "Buzz"
Merritt noted that, "The public concerns that drive issues in a community
normally don't arise at city hall or the courthouse. They bubble up from the
rich and, to most journalists, dark and trackless swamp of public life"
(Harwood, 1995, p.3).
Newspaper readership has dropped nearly 20 percent in the last ten years
(Merritt, 1996). A 1995 Times Mirror poll found that the number of people
who had read a copy of the previous day's newspaper had plummeted 13 percent
in just one year to an all-time low of 45 percent between 1994 and 1995
(Hackley, 1996). With newspaper circulation sliding, television news
struggling to keep viewers, and several media critics saying that journalists
have lost touch with their audiences, a change in reporting style appears
beneficial (Hackley, 1996). Hoyt (1995) believes that "journalism has
contributed greatly to the disconnect between the political system and its
citizens, and, therefore, journalism must change" (p. 28).
According to Merritt, another problem is that journalists have tried to
sell papers to an audience defined only by demographics and psychographics.
This system identifies people as the faceless mass, disconnected from the
reporter. Phil Meyer, Knight Chair in Journalism at the University of North
Carolina, believes it is just in the last two years that journalists have
started to connect troubles in their communities with those in their
newspapers. "They are just now beginning to realize that the decline in
civic involvement is tied together with their decreased readership" (Hackley,
1996, p. 1). Traditionally, "news media has revolved around the principle of
objectivity, reported through balanced reporting of events, press releases,
and quotes from 'informed sources' who are generally political and social
leaders" (Iorio, 1996a, p. 1).
Going to informed sources and social leaders is beneficial for
quotations and expert opinions but, if used on its own, this type of
reporting leaves gaps in news coverage because the view of the average
citizen is not considered. In fact, a 1994 Times Mirror poll revealed that
71 percent of Americans felt the press "gets in the way of society solving
its problems" (Hackley, 1996). James Fallows, formerly the Atlantic Monthly
's Washington editor and now editor of US News and World Reports is one of
civic journalism's staunchest supporters. He stated
It's the media's job to make it easier for the public to come to a
decision--they need to feel reconnected to public life. Journalists have
spent so much time talking to the politicians and the spin-doctors, they have
lost touch with their readers (Hackley, 1996, p. 2).
Public journalism has expanded from a handful of experiments in 1990 to more
than 100 projects nationwide in 1996, and the number is growing constantly
(Friedland, Scotirocvic & Daily, 1996). The Wichita Eagle led the way in the
trend toward public journalism with its People Project in 1991. Other
efforts like The Charlotte Observer's "Your Voice, Your Vote" and "Taking
Back Our Neighborhoods" projects; the Minneapolis's Star Tribune's endeavor,
"Neighborhood Repair Kit"; and the work of Norfolk Virginian-Pilot's "public
life team" have followed close by behind in the use of public journalism.
The goal of this philosophy is to produce journalism that includes citizens'
concerns as a starting point for and enhancement of traditional issues news
coverage. To accomplish this, the focus is on the audience as citizens who
have common concerns. Public opinion research is used to examine citizens'
commonly held concerns through methods that encourage full, in-depth
responses (Iorio, 1996a).
Yet, insufficient research has been directed toward both public
journalism and citizens' concerns. Little is known about the relationship
between the public's concerns and issues reported in the media. Culot (1992)
compared national and local news with individuals' concerns, but no research
has looked at agenda-setting in regard to a newspaper whose stated goal is to
advance public journalism. Neither has any previous research looked at the
concerns of individuals associated with public journalism activities over
time and compared that relationship to agenda-setting effects. The present
study will address this arena by examining the relationship of the
agenda-setting function of the media and citizens of Sedgewick County,
Kansas.
Summary
The study to follow is built on agenda-setting research and the concept
of public journalism. Culot's (1992) research compared national and local
issues, and examined individuals' concerns, but no research has looked at
agenda-setting concerning a newspaper whose stated goal is to advance public
journalism in regard to the agenda-setting effect. Nor has any previous
research looked at the concerns of individuals associated with public
journalism activities over time and that relationship to agenda-setting
effects.
Data for this thesis were gathered, in part, from a personal interview
study funded by the Kansas Public Finance Center, Knight-Ridder Foundation,
and The Wichita Eagle. The study was conducted at Wichita State University
by the Elliott School of Communication. To find out what the concerns of
"the people" were, in-depth interviews were performed. A content analysis of
the data was then performed. The next chapter will introduce the research
questions and hypotheses addressed in this study.
Methodology
The purpose of this study is to examine the agenda setting effect with
regard to citizens' concerns and look at the way in which agenda setting is
viewed by individuals who have experience with public journalism. The
proposed research replicates a study that expands a traditional
agenda-setting research content analysis and introduces a correlation with
in-depth interviews, instead of surveys (Culot, 1992). This proposed study
will also add to the original research by making a comparison of the 1992 and
1996 studies and looking at individuals with a connection to public
journalism and those with no connection.
Research Questions and Hypotheses
Based on the review of agenda-setting and public journalism literature
in the previous chapter and a review of Culot's (1992) study on
agenda-setting and individuals' concerns, the following research questions
and hypotheses were formed.
Research Question 1: Which medium sets the agenda for the concerns of
Sedgewick County, Kansas, residents, television or newspapers?
H1) Agenda-setting effects will be weaker for local coverage than with
national coverage.
H2) Newspapers will exert a stronger agenda-setting effect than television.
H3) In instances of low obtrusiveness, agenda-setting effects of the media
will be strong. However, with highly obtrusive issues, the media will
have
little or no effect on issue salience.
H4) The agenda-setting effect found in 1992 will be positively associated
with the effects found in 1996.
Research Question 2: Is the agenda-setting effect stronger for citizens who
have ties to public journalism, than those with no ties?
H5) The agenda-setting effect will be stronger for individuals involved in
public journalism activities compared to those not involved.
As in Culot's (1992) study, the first hypothesis is based on Palmgreen
and Clarke's (1977) study of agenda-setting effects for local and national
issues. Palmgreen and Clarke based their hypothesis that agenda-setting will
be weaker at the local than the national level on three factors: 1) the more
directly observable nature of local issues, 2) the nature and strength of
local interpersonal political communication networks, and 3) the heavier
media coverage of national issues.
Hypothesis 2 is based on the work of McCombs (1976). McCombs found that
"Newspapers are the prime movers in organizing the public agenda. They
largely set the stage of public concern" (p. 6). He did, however, continue
that "television news is not wholly without influence" (p. 6).
The third hypothesis is based on definitions of issues obtrusiveness.
Winter (1981) defines obtrusiveness as the amount of personal experience
people have with issues, while Demers, Craff, Choi and Pessin (1989) define
issues obtrusiveness as the perceived amount of personal experience a person
has with an issue. In their 1989 study, Demers et al. found that people
depend on the media rather than on their own personal experience to form
their agendas for unobtrusive issues such as international and some national
events. For obtrusive issues, such as local issues, however, people rely
more greatly on personal experiences and observations than on the media.
The first and third hypotheses in the study are the address the same
issues as to two of the three hypotheses used in Culot's 1992 study and are
based on the same premises. The second is a construct replication of her
hypothesis and both the fourth and fifth hypotheses are extensions to Culot's
study. The fourth hypothesis is formulated from the notion that the current
concerns of individuals do not differ greatly from the concerns identified in
the past study.
The fifth hypothesis is not based on a previous study. Public
journalism has yet to be thoroughly examined. Up to now, most of the
articles on public journalism are basically definitions of the movement and
descriptions of projects (Merritt & Rosen, 1995; Merritt 1995; Fallows,
1996). Charity (1995) presents a handbook of how to perform public
journalism, but no studies on the relationship of public journalism,
agenda-setting, and citizens' concerns have been published. The last
hypothesis is designed to address that issue by comparing concerns voiced by
those interviewed both in 1991 and in 1996, and those interviewed for the
first time in 1996.
The 1992 Study
In-depth interviews with 1991 Wichita, Kansas residents were conducted
from October 21, 1991 to December 2, 1991 by 18 trained graduate students
from the Elliott School of Communication at Wichita State University. The
pre-tested interview guide consisted of all open-ended questions, with the
exception of demographic information. The goal of the interviews was to get
respondents to identify personal concerns. The data gathered from these
interviews, were used to identify the ten most salient issues for Wichita
residents. The media agendas of The Wichita Eagle, CBS network news, and
Channel 12, CBS's local affiliate, were measured from October 7, 1991, two
weeks prior to the beginning of the interviews, to December 2, 1991, the last
day of the interviews. The analysis excluded Sundays.
Culot found that "agenda-setting effects are stronger for local issues
compared to those for national issues, that television dominates as agenda
setter with respect to both local and national issues. . ." (Culot, 1992).
She concluded that local television was the dominant agenda setter in
Wichita.
The 1997 Study
The preliminary data for the current study were gathered in Wichita,
Kansas, over a six day period from Thursday, October 3, 1996 through Tuesday,
October 8, 1996, as part of a different study. The data were gathered by
members of a graduate class conducting a study on the development of
political issues. The data compiled from that study were used in the present
study to identify the ten most salient concerns for Wichita residents during
that time period. The interview guide was almost identical to the one used
in 1991. The focus of the interview was to elicit personal concerns about
public issues from the respondent.
Of the 191 respondents interviewed for the 1992 study, 37 were
re-interviewed for this study. Attrition was a problem in the drawing of the
sample of previously interviewed respondents. Fifty-four new respondents who
had no connection to the previous research project, was selected by employees
at The Wichita Eagle's Research Center using random digit dialing.
The original sample for the 1996 study was overdrawn by 20 individuals,
to 110, in case some respondents were unable to complete the interview
process. Of the 90 individuals scheduled for the interviews, 11 did not keep
their appointments. Twelve telephone interviews of individuals from the
overdraw sample were conducted. Since two interview methods were used,
personal interviews and telephone interviews, the study mitigated monomethod
bias. The responses from the telephone interviews were similar to those
obtained in the focused interviews.
A check for possible systematic response bias was made. Interviewers
estimated through observation that the respondents involved in the 79
face-to-face interviews were mainly white and middle class. Other
demographic variables, such as party affiliation, age, and gender, were found
to be similar to the current population of Sedgewick County. Despite the
levels of incongruence, the demographic averages were somewhat similar
(Jones, 1996; Zaher, 1996). Tables 1 through 3 in Appendix A show demograph
ic breakdowns for the in-depth interview respondents as calculated by Zaher
(1996).
To measure the public agenda, respondents were asked, "What are the most
serious concerns facing people today?" After all issues had been explored,
respondents were asked to, "name the problem that is the most important."
The face-to-face interview format allowed interviewers to establish rapport
and obtain detailed responses full of personal experiences. A content
analysis of all the issues mentioned in the interviews and one of "the most
important issues" was conducted.
The interviews were tape-recorded to insure coding accuracy. Coding of
the data for this study was standardized through coding sheets that included
the following data: issues identified; how issues are discussed; level of
concern: personal, neighborhood, community, state, national, international;
and most important issue. A copy of the post-interview, individual
respondent coding sheets are attached as Appendix B.
To overcome threats to internal validity, a new step, differing from the
original study, was taken. The interview period was limited to six days.
Limiting the time frame also helped reduced any variances caused by the
history or maturation effect (Wimmer & Dominick, 1994). As with all research
designs, some limitations are always artifacts of the design and the reader
should approach the work with that in mind.
In a quantitative content analysis of the interviews, all 91 of the
interviews conducted were coded for general issue topics. A copy of the
completed coding sheets is included as Appendix C. To assure intercoder
reliability in the coding sheets, 10 percent were re-coded and equivalency in
coding was found. Intercoder reliability, using Holsti's (1969) measurement,
was equal to .98.
Measuring Media Content
Following the 1992 study's methodology as closely as possible, the same
information sources and parameters were used to assess the media agenda. To
measure the media agenda, a content analysis of the local paper and national
and local television coverage was carried out for 69 days, August 1, 1996
through October 8, 1996. Salwen's (1988) research on the length of time it
takes for public adoption of news media salience found that "the period of
peak or optimal relationship between the public agenda and the media agenda
appears after the accumulation of about eight to ten weeks of media coverage"
(p. 106). For this reason, ten weeks of articles were analyzed. As in
Culot's 1992 study, only stories dealing with the ten "most important"
concerns of citizens were counted. Stories not relating to the top ten
concerns of citizens were coded in the "other" category. Copies of the
coding sheets for the three media sources can be found in Appendices D.
In the analysis of The Wichita Eagle, of the 339 stories, 269 were
coded. Since media salience of issues in the news is often exhibited through
prominence of story treatment and display, only front-page stories were
coded. Budd (1964) was the first to judge the prominence of a story based on
its location in the paper and on the page. McCombs (1975) agrees with this
practice, noting that "front pages seem to have more impact than inside
pages. The agenda of the front page is learned better than the total agenda
of news items in the paper" (Sohn & Sohn, 1982, p. 20).
Coverage of national network news, for the same time frame, was
collected for CBS evening news from the Television News Index and Abstracts
published by Vanderbilt University's Television News Archives. Of the 207
stories, 201 were coded. Local television news coverage was based on the
6:00 p.m. evening newscasts of KWCH TV Channel 12, Wichita's CBS affiliate.
Of the 201 stories, 145 were coded from the topic "run-down" assigned to them
by the six o'clock producer.
In concurrence with the 1992 study, only the first three stories in both
the national and local newscasts were coded. This decision was based on Behr
and Iyengar (1985) who found that all news stories are not equal in forming
public opinion. Some, namely the lead stories, have a stronger influence on
viewers' perceptions. Watt, Mazza, & Snyder (1993) found that the
agenda-setting effect is stronger earlier on in news broadcasts and loses
influence as the program continues. Petty and Cacioppo (1981) agree that the
first argument presented is often more persuasive than those following.
Intercoder reliability was verified using Holsti's (1969) method. For
all media involved in the content analysis the reliability level totaled
.96. Intercoder reliablilty for the completion of the post-interview coding
sheets was .96. In order to further validate data coding, a comparison was
made between the coding of the 1991 data and the 1996 data. Because the
entire data set from 1991 was not available it was impossible to complete a
statistical test of intercoder reliability. However, sample comparisons of
the coding showed equivalency for the 41% that was available.
Findings
The Findings section will look at which media is the dominate
agenda-setter for Sedgewick County citizens. It will compare the strength of
the agenda-setting effect between local and national coverage, and between
the mediums of newspaper and television. Issue obtrusiveness and the
longitudinal effect of agenda-setting also will be examined. In addition,
involvement or lack of involvement in public journalism activities will be
correlated to the agenda-setting effect. The Pearson r Correlation Matrix,
from which most of the tables in this chapter was composed, can be found in
Appendices E.
A comparison of the top ten "most important" concerns citizens cited in
the 1996 interviews and the ten most cited concerns in the interviews are
listed below in Table 4.
Table 4
"Most Important" Concerns and Most Cited Concerns by Citizens in 1996
Ranking
"Most Important" Concerns
Most Cited Concerns
1.
Family/Family Values
Economy
2.
Crime
Crime
3.
Economy
Government
4.
Education
Family/Family Values
5.
Children
Education
6.
Government (tie for 6)
Health Care
7.
Health Care (tie for 6)
Taxes
8.
Taxes (tie for 6)
Children
9.
Drugs
Environment
10.
Environment
Drugs
Spearman rank-order correlation: r =.8177; p<.01
As would be expected, the same ten concerns that were most cited by
citizens also made up the most important concerns. Family/Family Values is
ranked first on the most important list and fourth on the most cited list.
Family/Family Values ranked first on the most important concerns list because
many citizens felt that if there was a stronger emphasis on family and family
values many of the other concerns they mentioned would take care of
themselves. Children are listed as five and eight. The fact that this
concern is ranked higher on the most important concern list than the on the
most cited list also may be attributed to the reasoning that if youth are
raised correctly, other problems would diminish.
Table 5
Top Three Most Important/Cited Concerns and Media Ranking of Issues
The Wichita Eagle
CBS
Channel 12
Most Important (Family/Family Values, Crime, Economy)
-.9988
p>.05
.0517
p>.05
.8973
p>.05
Most Cited
(Economy, Crime, Govt.)
-.9177
p>.05
.4874
p>.05
.6102
p>.05
As Table 5 illustrates, there were no statistical significance between
either the top three "most important" concerns and the media outlets, or the
top three most cited concerns and the media outlets. All ten of the concerns
are necessary to find a statistically significance correlation between these
concerns and the issues presented in the media. To further analyze this
question we will examine our four hypotheses.
The content analyses of news sources provided a final sample of 747
media messages for the period of August 1, 1996 to October 8, 1996.
Three-hundred and thirty-nine stories were coded from The Wichita Eagle, 207
stories from CBS's network evening newscast and 201 stories from the Channel
12 six-o'clock newscast. A content analysis for each of the top ten most
important concerns noted by citizens was preformed for The Wichita Eagle, CBS
network broadcast, and Channel 12 broadcasts. The issues presented in the
media and the citizens' concerns are listed in rank order in Table 6 below.
This data will be used in thee next step of the research which will analyze
the findings to determine whether or not they support the research hypotheses
and how they answer the research questions.
Hypothesis 1: Agenda-setting effects will be weaker for local coverage than
national coverage.
The first hypothesis was not supported. The media exerted stronger
agenda-setting effects for local coverage than for national coverage. Table
6 shows the ranking of issues by Channel 12, the CBS affiliate in Wichita,
closest to the ranking of Sedgewick county citizens' most important
concerns, although there was not a great deal of difference among any of the
media outlets.
Table 6
Ranking of "Most Important" Concerns, Media Issues and Percentage of Coverage
(August 1, 1996 to October 8, 1996)
Ranking
"Most Important" Concerns
The Wichita Eagle
CBS network news
Channel 12 news
1.
Family/Family Values
Government
41.57%
Government
45.77%
Crime
58.7%
2.
Crime
Crime
13.86%
Crime
20.9%
Government
23.91%
3.
Economy
Economy
11.24%
Environment
10.95%
Education
9.42%
4.
Education
Education
9.36%
Economy (tie, 4)
5.97%
Economy
2.9%
5.
Children
Environment
6.37%
Health Care (tie, 4)
5.97%
Family/Values (tie, 5)
1.45%
6.
Government
(tie for 6)
Health Care
5.62%
Drugs
6.47%
Children (tie, 5)
1.45%
7.
Health Care
(tie for 6)
Drugs
4.12%
Education
1.99%
Drugs (tie, 5)
1.45%
8.
Taxes
(tie for 6)
Family/Family Values
3.37%
Taxes
1%
Health Care
.72%
9.
Drugs
Taxes
3%
Children
1%
Taxes (tie, 9)
0%
10.
Environment
Children
1.5%
Family/Family Values
0%
Environment (tie, 9 )
0%
Total
100.01%
100.02%
100.0%
Spearman rank-order correlation
r=.2937
p>.05
r=.2093
p>.05
r=.5553
p>.05
Note: Percentage may not sum to 100.0 because of rounding.
Although the ranking of issues such as Crime, Economy, and Education are
similar, when the Spearman rank-order correlation was conducted no
significant correlations were found. However, a comparison of the rankings
of most cited concerns and media issues as shown in Table 7 does produce a
significant correlation.
While the Spearman rank-correlation shows that the issue agendas of all
three media organizations were significantly related to citizens' most cited
concerns, there were drastically different ranking of issues. Although
citizens' were most concerned by the Economy, it was ranked third (The
Wichita Eagle) and fourth (CBS and Channel 12) by the media. The Wichita
Eagle, a institution devoted to public journalism, only devoted 11.24% of its
coverage to the area, this was, however, the largest coverage percentage
devoted by any of the media examined to the Economy.
Table 7
Ranking of Most Cited Concerns, Media Issues and Percentage of Coverage
(August 1, 1996 to October 8, 1996)
Ranking
Most Cited Concerns
The Wichita Eagle
CBS network news
Channel 12 news
1.
Economy
Government
41.57%
Government
45.77%
Crime
58.7%
2.
Crime
Crime
13.86%
Crime
20.9%
Government
23.91%
3.
Government
Economy
11.24%
Environment
10.95%
Education
9.42%
4.
Family/Family Values
Education
9.36%
Economy (tie, 4)
5.97%
Economy
2.9%
5.
Education
Environment
6.37%
Health Care (tie, 4)
5.97%
Family/Values (tie, 5)
1.45%
6.
Health Care
Health Care
5.62%
Drugs
6.47%
Children (tie, 5)
1.45%
7.
Taxes
Drugs
4.12%
Education
1.99%
Drugs (tie, 5)
1.45%
8.
Children
Family/Family Values
3.37%
Taxes
1%
Health Care
.72%
9.
Environment
Taxes
3%
Children
1%
Taxes (tie, 9)
0%
10.
Drugs
Children
1.5%
Family/Family Values
0%
Environment (tie, 9 )
0%
Total
100.01%
100.02%
100.0%
Spearman rank-order correlation
r=.6621
p<.05
r=.5823
p<.05
r=.7372
p<.01
Note: Percentage may not sum to 100.0 because of rounding.
Hypothesis 1 was not supported in Culot's (1992) study, either. She
also found the media to exert stronger agenda-setting effects for local
coverage than for national coverage. Culot, too, found the ranking of issues
by Channel 12 to be the closest to the ranking of concerns by citizens.
Hypothesis 2: Newspapers will exert a stronger agenda-setting effect than
television.
Hypothesis 2 was not supported. As illustrated in the rankings and in
the Spearman's rank-order correlation statistics in Table 7 above, the
strongest correlation of citizens' most cited concerns was with the Channel
12 television newscasts, (r=.7372; p<.01). Next came The Wichita Eagle,
(r=.6621; p<.05), and then the CBS network news (r=.5823; p<.05). As with
hypothesis 1, the ranking of issues by Channel 12, CBS's local affiliate, is
closest to the ranking of concerns most cited by citizens.
Hypothesis 2 is a construct replication of the second hypothesis used by
Culot (1992). Culot's (1992) hypothesis stated, "Newspapers will exert the
strongest agenda-setting effects on local issues whereas television will
dominate with respect to national issues." Culot did not preform any
statistical analysis of this hypothesis; she subjectively observed
similarities and differences in the rankings of concerns by citizens and the
ranking of issues by the media. Her findings did not support the hypothesis.
Hypothesis 3: In instances of low obtrusiveness, agenda-setting effects of
the media will be strong. However, with highly obtrusive issues, the media
will have little or no effect on issue salience.
The third hypothesis was not supported. In order to assess this
hypothesis there is a need to differentiate between obtrusive and unobtrusive
issues. Demers, Craff, Choi and Pessin (1989) operationally define
obtrusiveness as the perceived amount of personal experience a person has
with an issue. Following this guideline, as well as those set by Culot's
study, the issues of Children, Crime, Drugs, Economy, Family/Family Values,
Health Care, and Taxes were characterized as obtrusive. Environment, Ed
ucation, and Government were classified as unobtrusive issues. The only
significant difference between obtrusive and unobtrusive concerns in 1996,
occurred for obtrusive concerns and The Wichita Eagle.
Hypothesis 3 is not supported. No significant correlation is made
between the unobtrusive concerns and any of the media outlets examined,
however, a correlation of r=.8784; p<.05 is made between obtrusive concerns
and The Wichita Eagle. The fact that little difference is seen by
respondents between obtrusive and unobtrusive issues may be because seven of
the ten concerns were talked about on an obtrusive level. This idea will be
further examined in the Discussion section. Below, Table 8 shows the Pearson
r correlation matrix for obtrusive concerns in 1996.
Table 8
1996 Obtrusive Concerns
Most Cited Concerns
Most Important Concerns
The Wichita Eagle
r=.8784
p<.05
r=.5718
p>.05
CBS
r=.5658
p>.05
r=.3452
p>.05
Channel 12
r=.654
p>.05
r=.5352
p>.05
Since Culot had not made a statistical analysis of the 1991 data, one
was ran. Culot had categorized the issues of Children's Futures, Crime,
Drugs, Economy, Family, Health Care System, and Taxes as obtrusive and the
issues of Abortion, Education, and Government as unobtrusive. Culot had
concluded that there was a partial support for this hypothesis in 1991.
Through a nonstatistical analysis of the rank-order table shown in Appendix F
as Table 9, she found that both Channel 12 and The Wichita Eagle "exert weak
agenda-setting effects for three obtrusive issues" (Culot, 1992, p. 51). Her
analysis continued that "the respondents ranked these three issues in their
hierarchy of issue salience as third, sixth and tenth respectively whereas
The Wichita Eagle placed them as sixth, eighth and fifth respectively, and
Channel 12 ranked them as seventh, ninth and fifth respectively."
However, when a statistical analysis was conducted a correlation to
citizens' most important concerns in 1991 was found with The Wichita Eagle
and Channel 12 for obtrusive concerns and with citizens' most important
concerns and Channel 12 for unobtrusive issues. An almost perfect
correlation occurred for unobtrusive concerns and Channel 12. Tables 10 in
Appendix F shows the levels of correlation. As Culot concluded in 1991,
after a statistical analysis was completed it was found that her hypothesis
is only partially supported.
Hypothesis 4: The agenda-setting effect found in 1991 will be positively
associated with the effects found in 1996.
Hypothesis 4 was not supported. Table 11, below, compares the 1991 and 1996
findings to the question, "What is the most important concern facing people
today?"
Table 11
Ranking of "Most Important" Concerns by Residents in 1991 and 1996
Ranking
1991
1996
1.
Crime/Gang Control/Safety
Family/Family Values
2.
Education
Crime
3.
Taxes
Economy
4.
Economy
Education
5.
Abortion
Children
6.
Family
Government (tie for 6)
7.
Government
Health Care (tie for 6)
8.
Health Care System
Taxes (tie for 6)
9.
Drugs
Drugs
10.
Children's Future
Environment
While eight of the original ten concerns are still in the top ten, a significant
relationship between the 1991 and 1996 most important issues was not found. The
agenda appears to shift as the importance of Family/Family Values rose to the
top of the rankings. Other top issues seemed to stay fairly consistent. Crime,
Economy, and Education still ranked in the top four as "most important" concerns
in both years. Drugs ranked ninth in both 1991 and 1992, and Government was
listed as seventh and sixth, respectively.
Hypothesis 5: The agenda-setting effect will be stronger for individuals
involved in public journalism activities compared to those not involved.
Hypothesis 5 was not supported. Table 12 compares the ranking of concerns
by new respondents, returning respondents, and all respondents to the three
media outlets examined. The only significant correlation found was between the
most important concerns of new respondents and Channel 12's reporting of issues.
Table 12
New, Returning, & All Respondents' Concerns and Media Issues
The Wichita Eagle
CBS National News
Channel 12
New Respondents
r=.3782
p>.05
r=.323
p>.05
r=.6587
p<.05
Returning Respondents
r=.2975
p>.05
r=.2036
p>.05
r=.3417
p>.05
All Respondents
r=.2937
p>.05
r=.2093
p>.05
r=.5553
p>.05
Table 13 displays the ranking of issues by all respondents, new
respondents, and returning respondents. The concerns of the new respondents
were closer than concerns of returning respondents to the concerns of all
respondents. Although the level of significance is extremely close, a
difference of only .0479, the correlation is closer for new respondents. (The
correlation level for new respondents was .9764, while the correlation level for
returning respondents was .9285.) Returning respondents did, however, have the
same top three concerns as all respondents.
Table 13
Ranking of "Most Important" Concerns by Returning and New Respondents
Ranking
All Respondents
New Respondents
Returning Respondents
1.
Family/Family Values
Crime
Family/Family Values
2.
Crime
Family/Family Values
Economy
3.
Economy
Education
Crime
4.
Education
Economy
Education (tie for 4)
5.
Children
Children (tie for 5)
Health Care (tie for 4)
6.
Government (tie for 6)
Government (tie for 5)
Children (tie for 6)
7.
Health Care (tie for 6)
Taxes (tie for 5)
Taxes (tie for 6)
8.
Taxes (tie for 6)
Health Care (tie for 8)
Drugs (tie for 6)
9.
Drugs
Drugs (tie for 8)
Government (tie for 9)
10.
Environment
Environment (tie for 8)
Environment (tie for 9)
Spearman's rank-order correlation
All to New Respondents
r=.9764; p<.01
All to Returning Respondents
r=.9285; p<.01
New to Returning Respondents: r=.894 ; p<.01
Now that the data has been analyzed and the hypotheses have been proven or
disproved, the two research questions can be analyzed.
Research Question 1: Which medium sets the agenda for the concerns of
Sedgewick County, Kansas residents, television or newspapers?
As the analysis of the second hypothesis demonstrates, television,
specifically local television as the primary agenda-setting medium for residents
of Sedgewick County, Kansas. The local newspaper came in second as the primary
agenda setter, with national television following close behind. Table 7
illustrates this. Culot's 1992 study also found local television to be the main
agenda-setting media. Analysis of the third hypothesis did not support it. The
research found The Wichita Eagle to play a significant part in setting citizens'
agendas for obtrusive concerns in 1996. No other significant correlations were
found for obtrusive or unobtrusive concerns. In addition, a statical analysis
of the 1991 data from the second hypothesis, finds Channel 12 to be extremely
highly correlated with the unobtrusive concerns of citizens in 1991. Both
Channel 12 and The Wichita Eagle, however, were statistically correlated with
obtrusive issues. The third hypothesis was not supported.
The findings of both the 1991 and 1996 research support the conclusion of
Iyengar and Kinder (1987) and Brosium and Keppling (1990 and 1992) which found
television as the strongest influencer in the agenda-setting process.
Research Question 2: Is the agenda-setting effect stronger for citizens who
have ties to public journalism, than those with no ties?
The fourth hypothesis finds a higher correlation to the media agenda for
citizens not exposed to public journalism, as opposed to those with prior
exposure. Although the level of significance is within .0479 (.9764 for new
respondents and .9285 for returning respondents; p<.01), the association is
still higher for new respondents.
Discussion
The Discussion section will begin by examining the homogeneity among the
three networks and then discuss the findings described in the previous chapter.
The findings showed that agenda-setting effects are stronger for local coverage
versus national coverage. It found local television to be the prominent
agenda-setter. The Wichita Eagle sets the agenda for obtrusive concerns. There
is no statistical correlation between agenda-setting in 1991 and 1996. And
finally, that there is little difference between respondents involved in public
journalism activities and individuals not involved.
Network Coverage
As earlier discussed, CBS was the television station chosen to measure the
media agenda for both local and national television coverage. The week of
September 22, 1996 was selected to gage the homogeneity of ABC's, CBS's, and
NBC's national network newscasts. As Stempel (1952) found, any period of four
or more days will adequately represent the news pattern in a media system. The
coding and summary sheet used for this sample are located in Tables 14 through
21 in Appendix G. All three networks yielded 12 stories each.
The comparison supported Meadow's (1973) assertion that network newscasts
are homogeneous and any bias is minimal. As Table 21 in Appendices G indicates,
the same three issues in 1996, Economy, Government, and Crime, occupy the top
three rankings for each of the newscasts. The level of difference between the
networks is slight. Spearman's rank-order correlation of ABC to CBS, r=.9707;
p<.01. For ABC to NBS the correlation is r=.9014; p<.01 and for CBS to NBC the
correlation is r=.9038; p<.01. Therefore, it was determined that the findings
of this study were not skewed by the selection of only one network to measure
television coverage of issues.
Hypotheses
When the first hypothesis was analyzed, it was not supported. As with
Culot's (1992) research, it was found that the agenda-setting effect of the
local media was stronger than the agenda-setting effect of the national media.
This conclusion may be explained by the intertwining of personal and wider
issues in the majority of the citizens' concerns. Seven of the ten concerns
cited are deemed obtrusive, personal, in nature. The interview format also may
have contributed to the findings. The people interviewed appeared to enjoy
talking about themselves and focus on things that directly affect them and often
localized nation issues as personal.
This brings up the question of "which direction does the agenda-setting
effect flow in Sedgewick County?" It seems as if the public may be setting the
media agenda. This is a goal of public journalism, but The Wichita Eagle is the
only media outlet examined that claims to be a proponent of the process.
As in Culot's (1992) study, hypothesis two was not supported. It was found
that local television, then national television, then the newspaper, in that
order, exert an agenda-setting relationship. This finding supports research by
Carter and Greenberg (1965) and Jacobson (1969) who concluded that television is
the most used and trusted news medium. "In commanding attention and shaping
opinion, television is now an authority virtually without peers" (Iyengar &
Kinder, 1987, p. 133). When considering this hypothesis, it is important to
note that because of measurement and problem formulation in agenda-setting
studies comparing television and newspapers, conclusive inferences are difficult
to support.
The third hypothesis was not supported. There was a significant
statistical correlation between citizens' most cited obtrusive concerns and The
Wichita Eagle. However, no correlation was found for unobtrusive concerns and
the agenda-setting effects of the mass media. In 1992, Culot concluded that
"some evidence was found for both the obtrusive contingency and the
cognitive-priming models" (p. 60). However, when a statistical analysis was
conducted a correlation was found between Channel 12 and The Wichita Eagle for
obtrusive issues. Citizens' 1991 unobtrusive concerns were found to highly
correlate with Channel 12.
The fourth hypothesis was not supported. There was not a statistically
significant correlation between the agenda-setting effects found in 1991 and the
agenda-setting effects found in 1996. It is interesting to note, that the same
findings for the first three hypothesis (that were either an exact replication
or a construct replication of Culot's 1992 research) the same conclusions were
reached. Although the topics named were the same, the agenda-setting effect was
not significant. Eight of the 1991 ten most important concerns were repeated in
1996. Crime, Drugs, Economy, Education, Family, Government, Health Care and
Taxes all made the top ten list again. Children's Future and Abortion were
replaced by Children and the Environment.
When hypothesis 5 was analyzed, it was not supported. The agenda-setting
effect was found to be stronger for individuals not involved in public
journalism activities compared to those involved. This may be explained by
David Manning White's (1950) concept of the gatekeeper. White concluded that
the selection of news is an extremely subjective task strongly influenced by the
gatekeeper's own attitudes, emotions, and experiences rather than by constraints
set by the media organization (Gaunt, 1990). Individuals involved in previous
public journalism activities may understand this process and therefore be less
influenced by what the media portrays as the most important issues.
The Wichita Eagle's statistical correlation with obtrusive concerns of
citizens in both 1991 and 1996 demonstrates that their public journalism effort
is making a difference in how the public perceives issues. So, if the goal of
public journalism is to foster a discriminate public. It would appear that The
Wichita Eagle is doing its job.
Conclusion
The purpose of this study was to determine which medium, television or
newspaper, sets the agenda for citizens of Sedgewick, Kansas, if this is the
same medium that set the agenda in 1991 and if the public journalism practices
that started in 1991 at The Wichita Eagle play any effect on individuals
involved in these activities. The literature review of agenda-setting research
and Culot's (1992) study suggested that agenda-setting effects could be
identified and a longitudinal comparison could be made. In respect to the
public journalism activities of The Wichita Eagle, it was proposed that those
with previous involvement in an project would be more sensitive than those not
previously involved.
The findings revealed that Channel 12, the local CBS affiliate, set the
agenda for citizens of Sedgewick County, Kansas. The same conclusion was
reached in 1991. Therefore, contrary to the hypotheses, local media was the
stronger agenda setter and television exhibited a stronger influence than
newspaper. An explanation was offered. Because seven of the ten most salient
issues identified, both in 1991 and 1996, were personal and local in nature,
that it would be logical that they would be discussed on a local level versus a
national level. National issues were often personalized and localized.
Culot concluded that the media "unethically" reported on issues important
to the public to gain sales. She compared this to the Social Responsibility
theory that arose in the middle of the 20th century. In opposition to the
libertarian theory which states that everyone should be granted access to the
press, the Social Responsibility theory states that "the press is controlled by
public opinion, consumer action and codes of ethics" (Culot, 1992, p. 63).
Culot continues,
As seen by the Hutchins Commission [in 1947], freedom of the press is a
moral and not a natural right. Therefore, the right can be lost or
forfeited
when abused. Underlying the basic social responsibility theory postulated
by
the Commission are two very fundamental propositions. The first
proposition
holds that "whoever enjoys freedom has certain obligations to society"
(Black
& Whitney, 1983, p. 432). Indeed, the media have the concomitant
responsibility to use their freedom to serve the welfare of society as a
whole. The second proposition states that "society's welfare becomes the
most overriding concern" (p. 432). Personal rights to free expression are
described in terms of public access to the media or the public's right to
know (p. 64).
Black and Whitney (1983) discussed the five basic requirements or
expectations set by the Hutchins Commission of Freedom of Press in 1947. First,
media must provide a "truthful, comprehensive and intelligent account of the
day's events, in a context that give them meaning" (p. 433). Second, media is
to serve as a "forum for the exchange of comment and criticism" (p. 434).
Third, it should project "a representative picture of the constituent groups in
society" (p. 434). Fourth, it must improve its "presentation and clarification
of the goals and values of the society" (p. 435). Finally, the media should
provide to the public "full access to the day's intelligence" (p. 435). Culot
concluded that in a democracy the media should function as "vox populi," the
voice of the people. Therefore, when the media reports on what the public feels
important it is failing in its social responsibility.
Today, however, the push toward public journalism suggests that the media
should report on issues important to the public. "The idea [of public
journalism] is to frame civic issues from the point of view of the citizens
rather than the politicians and talking heads" (Hackley, 1996, p. 2). So the
question remains, is the local media acting in a socially responsible way or is
profit the motive in determining the media agenda? Based on the premise and
goal of public journalism, this researcher believes that the local newspaper is
acting responsibly.
Further Research
As Culot noted in 1992, the main contribution this study makes to
agenda-setting research is a confirmation that television is the dominant agenda
setter, and does so confirming findings of an earlier study. It also concludes
that, at least in the Wichita, Kansas area, local television is the dominant
agenda- setter and its agenda greatly effects the salience of local issues.
Future research could focus on the long term effect public journalism plays
on local citizens. Five years may be too soon to see significant finding in
response to public journalism. Merritt (Merritt & Rosen,1995) notes that the
move away from detachment to values orientation required by public journalism
will take time. Rosen (Merritt & Rosen, 1995) adds that public journalism is ".
. . a set of practices that are spreading (slowly) through the American press.
|
