Missing the Link: Citizen-Based Journalism Intent
Rather than Election Coverage Content
Affects Public Trust in Media
by
Eric J. Rhodenbaugh
Park Doctoral Fellow
School of Journalism and Mass Communication
University of North Carolina at Chapel Hill
Eric Rhodenbaugh, Park Doctoral Fellow
School of Journalism and Mass Communication
CB#3365, UNC
Chapel, Hill, NC 27599
Phone (910)-462-2495, Office (919)-962-4091
email [log in to unmask]
Paper submitted for consideration in the Civic Journalism Interest Group.
Address all correspondence to the author. Missing the Link: Citizen-Based
Journalism Intent Rather than Election Coverage Content Affects Public Trust in
Media
ABSTRACT
Effects of citizen-based journalism content on public trust may be overshadowed
by the attitude change gained by involving the public in the news process. This
study explores the persuasive effects of a newspaper's intent to conduct
citizen-based journalism. In a 1996 election study of 20 markets, intent to
conduct citizen-based journalism, newspaper content, and trust were measured.
Long-term results showed that intent to conduct citizen-based journalism
increased trust in media independently of newspaper content.
Missing the Link: Citizen-Based Journalism Intent Rather than Election Coverage
Content Affects Public Trust in Media
Citizen-based journalism stormed into newsrooms in the early 1990s as a
revolutionary philosophy of newsroom management. The goals of this revolution
were to reconnect newspapers with the communities they serve and invigorate
public involvement in the democratic process. The concept itself is still in
the developmental stage, remains poorly defined, and differs widely in practice
among the 200 U.S. news organization currently practicing citizen-based
journalism.[1] Citizen-based journalism [2] has its critics, who, among other
things, criticize it as a gimmick to increase circulation.[3] To be sure,
increasing citizen involvement in the community should result in more citizens
reading the newspaper.[4] However, there is much more to this revolutionary
philosophy than building circulation. It involves a change in newsroom
attitudes toward the community, the news gathering process, and a shift away
from cynicism toward cooperation. It is, above all, a sincere attempt on the
part of its practitioners to rebuild deteriorating communities, and shore up a
fading participatory democracy.
Newspapers have been faced with declining readership, ad revenues, and public
trust since the 1950s.[5] Davis "Buzz" Merritt, Jr., considered a pioneer in
newsroom application of citizen- based journalism, was critical of the cynicism
and confrontational nature of traditional journalism. To save journalism from
itself, Merritt believed it needed a new philosophy emphasizing the newspaper as
a tool to aid in the betterment of the community through greater citizen
involvement.[6] According to this philosophy, newspapers should assume a major
role in persuading citizens to actively participate in their community and in
the democratic process. Merritt said that newspapers were obliged to "do our
journalism in ways that are calculated to help public life go well by engaging
people in it."[7]
Citizen-based journalism also received support from the academic community.
Jay Rosen, who pioneered public journalism from the academic side, echoed
Merritt's sentiments. Rosen stressed the media's role as the support system for
citizen involvement in the democratic process, referring to it as "an antidote
to disconnected communities and the publics' growing disgust with news people
and politicians."[8]
The philosophical changes espoused by Merritt and Rosen are rooted in the
concerns expressed by Daniel Yankelovich that the political process was becoming
a playing field of the elite. Traditional journalism relies heavily on experts,
spokespersons, and political elites in its reporting on politics. According to
Yankelovich, this emphasis can lead to a rift between the general public, the
media, and the political elites. Citizens disconnect from the democratic
process as they begin, more and more, to feel as though they are no longer part
of the system, and their opinions do not matter. The media, adept at bringing
issues to the public's attention, widen the gap by flitting from issue to issue
without pausing to consider the importance placed on any given issue by the
public. News media create further public cynicism by failing to provide the
amount and duration of coverage that would allow citizens to fully understand
and work through the issues, eventually reaching public judgement.[9]
Consequences of public disconnect and increased cynicism include public apathy
toward elections, less community involvement, loss of trust in institutions, and
loss of social capital. These symptoms are increasingly evident in recent
years. Election participation has declined steadily over the past two decades.
Participation in Presidential elections dropped to 50 percent in 1988, and
dipped below 50 percent in the 1990s. This drop was from a high of 63 percent
in 1960. Voting in U.S. Congressional elections fell to 33 percent in the late
1980s.[10] On the other hand, Ladd contends that voter turnout in Presidential
elections has remained fairly constant since 1932. Using United States Bureau
of the Census data, Ladd reports that in 1992, about 56 percent of those
Americans eligible turned out to vote in the Presidential election. This
represents a decline of seven percentage points from a high mark of 63 percent
in 1960, and Ladd argues that the numbers do not support the notion of
widespread decline in political participation.[11] Nevertheless, it is commonly
held that political participation, which is not necessarily reflected by voting
data, is on the decline.[12]
A second symptom of civic disconnect is loss of social capital. Putnam
analyzed Roper Center poll data and found that in 1993 less than 15 percent of
Americans participated in civic meetings. In addition, Putnam noted declines in
memberships in churchs, fraternal organizations, and other social groups,
including bowling leagues.[13] Loss of social capital is further reflected by a
drop in the number of people polled who felt that most people could be trusted.
Analyzing data from the General Social Survey, Putnam reported a 19 percent
drop from 1960 to 1993 in the percentage of Americans that felt most people
could be trusted.[14]
Public disconnect with the political process is further shown by an increasing
level of cynicism toward government. According to The Harris Poll, a national
poll conducted annually since 1966, in 1998 12 percent of Americans reported a
great deal of confidence in Congress and 17 percent reported a great deal of
confidence in the executive branch of government. These figures represent a
drop of 30 and 24 percentage points, respectively, since 1966. Similar results
were obtained by Putnam, who reported Roper Center poll data showing that in the
1990s, 75 percent of Americans stated that they do not trust the government.[15]
Trust in media has also backpedaled. The Harris Poll results for 1998 showed a
drop of 15 from 1966 in the percentage of Americans expressing a great deal of
confidence in the Media (14 percent in 1998 compared to 29 percent in 1966).
Similarly, a 1994 Times Mirror study showed that 71 percent of those polled
believed that the news media were an obstacle to issue resolution in
America.[16]
The above data show that Americans have lost interest in the political process,
lost trust in each other, in government, and in the media. Certainly, the media
cannot be the whipping boy and accept all the blame. However, traditional
political reporting, with its focus on conflict, horse races, and dependence on
experts and political elites, has contributed to Americans' increasingly
cynical outlook toward government and the news media. What then, could public
journalism offer as an alternative to traditional reporting in the hopes of
reversing the trend?
Merritt wanted to form a journalism whose content avoided traditional focus on
horse race polls rather than issues; emphasis on candidates' criticisms and
negative campaigning, which he thought deterred the voting public from
understanding the issues.[17] Rosen has also called for a shift away from
traditional journalism content. Rosen stated that conventional journalism, in
addition to emphasizing conflict and polarizing issues, tended to "exalt experts
and public opinion over citizens and public judgment."[18] Fouhy and Schaffer
say that one of the primary initiatives of citizen-based journalism should be to
help citizens listen and talk to each other.[19] Meyer listed six goals of
public journalism, one of which was to foster deliberation, meaning that news
media should encourage community members to "make that earnest attempt at
reciprocal understanding" of each others viewpoints.[20] It would follow, then,
that newspapers following the tenets of citizen-based journalism should strive
to provide in-depth coverage of issues, move the focus away from horse race
polls, and increase the number and prominence of citizen references. On the
non-content side, citizen-based journalism should strive to involve citizens in
the news gathering process, in the political process, and in the community.
Early efforts at applying public journalism led to some promising results.
Following the People Project[21], conducted in 1990 by the Wichita Eagle,
Wichita State University, KSNW-TV, and KNSS radio, a survey indicated that
reader satisfaction with the Eagle increased by 12 percent.[22] Meyer and
Potter, in their 1996 election study conducted across 20 media markets, showed
that newspaper intent to conduct citizen-based journalism resulted in more
coverage of hard issues and less reporting on horse race polls. In addition,
they showed that intent to practice citizen-based journalism was related to long
term trust in media. However, Meyer and Potter were unable to connect newspaper
election coverage content with changes in citizen attitudes toward the media,
the government, or each other.[23] Content of the 1996 election study was
defined by coverage of hard issues minus coverage of horse race polls. Further
study of the data analyzed by Meyer and Potter has failed to link other content
variables with citizen attitudes.[24]
An important aspect of the People Project, and also involved in other
citizen-based journalism projects, is direct citizen involvement. In the Voter
Project, which was the forerunner of the People Project, citizens were surveyed
and focus groups conducted in order to identify issues readers deemed most
important. The People Project relied on in-depth interviews and public input
through phone calls, faxes, and letters to identify the critical issues.[25] In
the most recent Your Voice Your Vote campaign, the Raleigh (N.C.) News &
Observer used surveys to identify issues readers wanted the newspaper to cover
in reporting of the 1998 election. The Charlotte (N.C.) Observer in 1992 not
only concentrated coverage on voter-identified issues, but was watched over by a
citizen panel, and had reporters ask questions at news conferences that were
sent in by citizens.[26] News councils, which allow direct public feedback
regarding newspaper coverage of an issue, are another example of methods used by
newspapers to get direct public input into the news process.[27]
An unexpected fallout of this heavy dose of citizen participation in news
gathering might be a variation of the Hawthorne effect. This effect could also
explain why content effects on citizen attitudes appear to be a missing link.
The Hawthorne effect [28] is named after a series of experiments conducted in
1927 at the Hawthorne works of the Western Electric Company. The experiments
were designed to measure worker response, in the form of increased production,
to various changes in the work place, such as rest periods, shorter days, and
lunch breaks.
Throughout the experiment the six workers were consulted continually by
experimenters about
the changes. Their comments were requested, and some changes that did not meet
their approval were abandoned.
A serendipitous finding had occurred. The response of the workers was not
related to the change in working conditions, but to the mere fact that they knew
they were part of the experiment. Productivity increased throughout the length
of the experiment, even when conditions were returned to the original. This
increase was not the result of changes in working condition, but the result of
the fact that the workers had been consulted about the experiment, felt it was
important, and knew the company was interested in the research. Productivity
increases may have resulted from an attitude change of the workers based on the
interest the company and the researchers showed toward them.
This form of persuasion can be adapted to explain changes in public attitudes
toward newspapers practicing citizen-based journalism. Most public journalism
campaigns involve direct public input as to what the newspaper should cover.
Surveys, focus groups, roundtable discussions with the editor, letters, email,
and similar tools are used to get the public to tell the newspaper what issues
to follow. These processes may have an unexpected side effect, which resembles
the Hawthorne effect, in that the mere action of seeking citizen input may
persuade the public that the newspaper cares about them and is interested in
their opinions. This in and of itself may result in changes in attitudes toward
the media, regardless of well the actual output of the newspaper follows
citizen-based journalism principles.
Persuasion involves changing audience attitudes, resulting in a desired change
in behavior. Attitudes are "orientations of the mind" that exert influence on
behavior.[29] Citizen-based journalism as a practice may be a form of
persuasion. One of the goals of citizen-based journalism is to increase public
involvement in the community, which is the desired change of behavior. This
change can be achieved by increasing public trust in government and in the
media, which is the desired change in attitude. The question, then, is whether
changes in attitude are wrought through changes in newspaper content, which
would be akin to a persuasive message, or through intent to practice
citizen-based journalism, which changes attitude by offering the public the
opportunity to be involved in the process. Perhaps this point is best
illustrated with a hypothetical example.
A newspaper professing to practice public journalism could fulfill its
obligation to encourage voter participation by taking out a full page ad urging
citizens to vote, colorfully detailing the importance of participation in the
democratic process, and listing registration and voting sites for the upcoming
election. A second public journalism newspaper conducts surveys, sponsors open
forums, and otherwise actively seeks the community viewpoint as to what the
pressing issues are in the election. Based on the Hawthorne effect, one would
expect the citizens served by the second newspaper to show greater trust in the
media than the citizens served by the first newspaper, regardless of how well
either followed citizen-based journalism tenets regarding election coverage
content.
It is the contention of this paper that the process of carrying out
citizen-based journalism produces a stronger effect on public trust in the media
than the strength of citizen-based journalism content. Content effects on
public attitudes, though possibly present, are overshadowed by effects resulting
from intent to do public journalism. This paper will study the effects of
citizen-based journalism intent versus citizen-based journalism content on short
term changes in media trust and on long term attitudes of the public toward the
media. The following hypotheses will be tested:
(1) Citizens in markets with strong citizen-based journalism intent, but weak
citizen-based journalism content will show similar levels of trust in media as
citizens in markets with both strong citizen-based journalism intent and
content.
(2) Citizens in markets strong in citizen-based journalism intent, regardless of
content, will show greater levels of trust in media than citizens in markets
choosing not to participate in citizen-based journalism.
(3) Citizens in markets strong in citizen-based journalism content, regardless
of intent, will show similar levels of trust in media as citizens in markets
weak in citizen-based content, regardless of stated intent.
(4) Citizens in markets strong in citizen-based journalism intent will show
greater levels of trust in media than citizens in markets with weak intent to
conduct citizen-based journalism, regardless of strength of citizen-based
journalism content.
RESEARCH METHODS
Data were collected by Meyer and Potter for their study of the effects of
citizen-based journalism during the 1996 election.[30] For this paper, only the
data used will be described. In each of 20 markets (see Table 1 for a listing
of the markets), five newsroom workers were surveyed regarding the intent of the
newspaper to conduct citizen-based journalism. The sample was purposive in that
it sampled a known range of media markets in order to get variance in
traditional versus public journalism practices.[31] The five workers selected
were those most likely to cover the 1996 campaign. Response rate was 69.5
percent. In each case, the original survey was mailed June 10, 1996. The
survey was followed by a post-card reminder, a second questionnaire to
non-respondents, and finally a third questionnaire via certified mail to those
that failed to respond to earlier mailings.
Intent of the newspaper to conduct citizen-based journalism (CBJ intent) was
calculated from seven items, listed below. Respondents rated their intention to
do each item on a three-point scale, with 3 indicating their organization was
very likely to do it; 2 indicating that it had not been decided; and 1
indicating a low likelihood of doing it. The average score was then calculated
for each newspaper. Actual CBJ intent results and rankings, relative to other
markets, are in Table 1.
The seven items were:
(1) Sponsor one or more public forums on issues
(2) Use polls to establish the issues coverage will focus on
(3) Conduct focus groups as a means to establish voters' concerns
(4) Form citizen panels to consult at different stages of the campaign
(5) Seek questions from readers for use when interviewing candidates
(6) Base reporting largely on issues developed through citizen contact
(7) Provide information to help citizens get involved in the political process
other than voting.
It is worth noting that five of the seven items involve some form of actual
involvement on the part of citizens, the remaining two items involve newspaper
content.
Based on philosophies and goals reviewed in the public journalism literature,
CBJ content was calculated by adding the percent of stories mainly about hard
issues with percent of stories with any mention of citizen references, and then
subtracting the percent of stories substantially or mainly about horse race
polls. About 44 percent of the variance in content was explained by intent to
do citizen-based journalism. Results of the calculation and rank, relative to
other markets, is included in Table 1. Table 1 - Scores and rankings for
intent to conduct citizen-based journalism, and scores and rankings for
citizen-based journalism content. Groupings are nonparticipants (nonpar),
nonperformers (nonperf), promise keepers (promkeep), and those in denial
(denial). Formation of the groupings is discussed in detail in the research
methods. Ranking of 1 is lowest, 20 is highest.
CBJ Intent CBJ Content
___________________ ___________________
Market Score Ranking Score Ranking Group
______________________________________________________________________________
Birmingham 1.79 5 11.18 8 denial
Little Rock 1.20 2 2.39 1 nonpar
Atlanta 2.03 7 11.53 11 denial
Rockford 2.62 13 6.15 3 nonperf
Chicago 2.31 10 13.08 14 denial
Des Moines 2.18 9 7.84 4 nonperf
Wichita 2.70 16 18.68 17 promkeep
New Orleans 1.71 4 5.28 2 nonpar
Port. (ME) 2.72 18 12.84 13 nonperf
Boston 2.64 14 9.16 6 nonperf
Grand Rapids 1.00 1 11.21 9 denial
Minneapolis 2.67 15 20.82 18 promkeep
Raleigh 2.51 12 12.49 12 promkeep
Charlotte 3.00 20 25.61 20 promkeep
Port. (OR) 2.72 18 16.88 16 promkeep
Columbia 2.37 11 13.26 15 denial
Austin 1.93 6 11.51 10 denial
Houston 1.44 3 9.04 5 nonpar
Norfolk 2.89 19 24.48 19 promkeep
Richmond 2.14 8 9.93 7 nonperf
______________________________________________________________________________
The number of nonparticipants = 3; nonperformers = 5; promise keepers = 6; in
denial = 6.
Actual citizen-based journalism content (CBJ content) was calculated based on
content analysis of campaign coverage over a seven-week period prior to the
election. A sample of 10 issues was collected for each market on random days
during the seven-week period. There were a total of 1,873 stories analyzed.
Stories were ranked from 1-4 for each of the content variables. A ranking of 1
= no content, 2 = minimal, 3 = substantial, and 4 = mainly.
Trust in media is based on pre-election (August 1-11) and post-election
(November 6-17) surveys of at least 50 citizens in each of the 20
markets--defined as living in the home counties of each newspaper as listed in
Editor & Publisher. A total of 1,012 citizens were interviewed in August, 1,030
in November. The November survey consisted of 623 citizens that had been
interviewed in August and 407 new interviews, which served as a control for
interview effects. Media trust was calculated by indexing answers to two of the
questions, listed below.
14g--I'm going to mention some people or things that could be blamed for the way
the political process works. I'd like you to tell me how much blame each of the
following deserves: a lot, a little, or none at all. How about...(g=the media).
38--Would you say the news media are pretty much run by a few big interests
looking out for themselves or that they are run for the benefit of all the
people.
One point was given for an answer of a little, and two points given for an
answer of none at all for question 14g; one point was given for an answer of for
the benefit of all the people for question 38. The possible range of the trust
in media variable was 0-3. The two questions were significantly correlated
(correlation coefficient=.288, p < 0.01).
Index results from August were subtracted from November in order to study short
term shifts in media trust. The shift variable was collapsed into three
categories, "gained trust" had any positive shift value, "no change" had a shift
value of 0, "lost trust" had any negative shift value. Long term results were
based on analysis of categorized index values in August and November. Data
were collapsed into two categories--high trust included values of 2 and 3, low
trust included values of 0 and 1. All trust results were based on panel data
(623 respondents surveyed both times). Frequencies for shift in trust, August,
and November index values are in Appendix 1.
Newspaper markets were divided into four groups in order to test the
hypotheses. The markets in the lower left section of Figure 1 are
nonparticipants. Those in the upper right are promise keepers. Markets in
between are either nonperformers or in denial, based on whether their content
was ranked lower than their intent (nonperformers) or whether content was ranked
higher than intent (in denial). Actual group designations are in Table 1.
Groups were calculated by comparison of intent and content rankings. Rankings
were split into three categories--1-7, 8-13, and 14-20. Nonparticipants
(nonpar), were those ranking in category 1- 7 for both intent and content.
Nonperformers (nonperf), were those ranking in one of the higher categories,
8-13 or 14-20, in intent, but in a lower category in content. Those in the
upper two ranking categories that stayed within their category in both intent
and content were classified as promise keepers (promkeep). If a market's
categorical ranking in content was ahead of its categorical ranking in intent,
it was considered to be in denial (denial) of its intent to conduct
citizen-based journalism.
Four comparisons of these groups were used to address the hypotheses. These
comparisons were done first for the shift in media trust variable, to address
short-term effects, and then carried out on the August and November categorized
media trust variable, to address long-term effects. Comparisons were:
(1) nonperformers versus promise keepers.
(2) nonparticipants versus high intent (mean of nonperformers and promise
keepers)
(3) low content versus high content (low = nonparticipants and nonperformers;
high = promise keepers and in denial)
(4) low intent versus high intent (low = nonparticipants and in denial; high =
promise keepers and nonperformers.
Comparisons involved analysis of percentages across categories, therefore
crosstabs with Chi
square measure of significance was used to test results.
Figure 1 - Comparison of ranking of intent to conduct citizen-based journalism
versus rank of content that reflects citizen-based journalism during coverage of
the 1996 election. Calculations of intent and content are described in the
methods. RESULTS AND DISCUSSION
Short-term results show support for the hypothesis that promise keepers and
nonperformers would show similar levels of gain or loss of media trust. Results
also show that there is no difference in the short-term change in media trust
based on content. However, the results fail to support the contention that high
CBJ intent would show gains in public trust of media while nonparticipants or
low CBJ intent would either remain the same or show significantly less gain in
public trust in media.
There was no significant difference in trust gained or trust lost from August
to November between the promise keepers and the nonperformers (Table 2). Recall
that the nonperformers and promise keepers both had high CBJ intent rankings,
but the nonperformers failed to follow through on that intent with strong CBJ
content. Results from Table 2 tend to support the hypothesis that it is the
intent to practice public journalism, as reflected in the high level of items
relating to actual citizen input discussed in the methods section, rather than
actual coverage content that increased citizen trust in media. It is
interesting to note that for both categories, the same percentage of citizens
lost trust in media as gained trust (Table 2). Further exploration of the
short-term data, though, fails to support the contention that intent, and not
content, affect public trust in the media. Results in Table 3 show that four
percentage points more citizens gained trust, and three percentage points fewer
lost trust in the high intent markets versus the nonparticipants; however, these
results were not significant.
Table 2 - Comparison of nonperformers (Nonperf) with promise keepers (Promkeep)
for shift in media trust from August to November (Shift).
Shift Nonperf Promkeep
_______________________________________________________________
Gained trust 24 22
Lost trust 24 22
No change 52 56
_______________________________________________________________
Values in the Nonperf and Promkeep categories are percentages. Percentages
within a column not adding to 100 are the result of rounding.
Table 3 - Comparison of nonparticipants (Nonpar) with the mean of nonperformers
and promise keepers (High intent) for shift in media trust from August to
November (Shift).
Shift Nonpar High intent
_______________________________________________________________
Gained trust 19 23
Lost trust 20 23
No change 62 54
_______________________________________________________________
Values in the Nonpar and High Intent categories are percentages. Percentages
within a column not adding to 100 are the result of rounding.
Table 4 addresses the effects of citizen-based journalism content, independent
of citizen-based journalism intent, on changes in public attitude toward the
media resulting from the seven weeks of election coverage. There was no
significant difference in those that either gained or lost trust between the low
content and high content markets (Table 4). The effects of citizen-based
journalism intent, independent of actual content produced, are presented in
Table 5. There was a five percentage point difference in those that gained
trust in the high intent markets, but this result was not significant (Table 5).
Table 4 - Comparison of markets that reflected a high content of citizen-based
journalism articles (High content) with markets reflecting low citizen-based
journalism content (Low content) for shift in media trust from August to
November (Shift). High content is the mean of promise keepers and those in
denial. Low content is the mean of nonperformers and nonparticipants.
Shift Low content High content
_______________________________________________________________
Gained trust 22 20
Lost trust 23 24
No change 56 57
_______________________________________________________________
Values in the Low content and High content categories are percentages.
Percentages
within a column not adding to 100 are the result of rounding.
Table 5 - Comparison of markets with strong intention to conduct citizen-based
journalism (High intent) with markets that had weak intentions to conduct
citizen-based journalism (Low intent) for shift in media trust from August to
November (Shift). High intent is the mean of the promise keepers and
nonperformers. Low intent is the mean of the nonparticipants and those in
denial.
Shift Low intent High intent
_______________________________________________________________
Gained trust 18 23
Lost trust 24 23
No change 59 54
_______________________________________________________________
Values in the Low intent and High intent categories are percentages.
Percentages
within a column not adding to 100 are the result of rounding.
Media trust did not significantly shift as a result of seven weeks of coverage
of the 1996 election. Intent to conduct citizen-based journalism in coverage of
the election also failed to produce any effect on citizen trust in the media.
This is not an entirely surprising result. The premise of this research is
based on persuasion theory, specifically the persuasive effect of seeking public
involvement as opposed to relying on persuasive messages to bring about a change
in attitude. Research in persuasion has indicated that attitudes are enduring,
that is they can be changed, but only over a long period.[32] Cynical
attitudes toward media developed by exposure to decades of cynical,
conflict-based traditional journalism are unlikely to change as a result of
exposure to seven weeks of election coverage. An interesting area for future
research might be to compare attitudes toward the media of readers that began
their newspaper habit with a public journalism newspaper versus readers in the
same market that have been reared on traditional journalism, but currently are
exposed to citizen-based journalism. Perhaps cynicism toward media is not the
result of age, but the result of the philosophy of the newspaper one grows up
with.
Long-term results provide support for all three hypotheses used to test the
idea that involvement of citizens in the process, reflected by intent to conduct
citizen-based journalism, resulted in improved attitudes toward the media
independently of whether the newspaper's content reflected the ideals of
citizen-based journalism. Comparing nonperformers with promise keepers and
comparing low content markets with high content markets showed no difference in
citizens level of trust in the media (Tables 6 and 7). Comparing
nonparticipants with high intent markets and comparing low intent markets with
high intent markets did produce significant different levels of trust in media
(Tables 7 and 8).
Citizen's trust in media in August and November did not significantly differ
between promise-keeper markets and nonperformer markets (Table 6). Actually,
the nonperformers showed somewhat greater levels of high media trust than the
promise keepers, 26 percent versus 21 percent, but this result was not
significant. Table 6 results offer proof of the persuasion by involvement
hypothesis. Lack of a content effect is supported by results shown in Table 7.
There was significantly (p < 0.05) greater trust reported in August in the high
intent markets versus nonparticipant markets. High intent markets had 23
percent of their citizens in the high trust category while nonparticipant
markets showed 13 percent. November results showed a similar trend with 22
percent of the citizens in the high trust category for high intent markets, and
15 percent in the nonparticipant markets, but these results were not
significant.
Table 6 - Media trust in August and November for nonperformers and promise
keepers. Trust categories are Low (index values of 0 and 1) and High (index
values of 2 and 3). Values are percentages.
Media Trust
Month Category Nonperf Promkeep
___________________________________________________
August High 26 21
Low 74 79
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - -
November High 23 22
Low 77 78
___________________________________________________
Table 7 - Media trust in August and November for nonparticipants (Nonpar) and
the mean of promise keepers and nonperformers (High Intent). Trust categories
are Low (index values of 0 and 1) and High (index values of 2 and 3). Values
are percentages.
Media Trust
Month Category Nonpar High Intent
___________________________________________________
August** High 13 23
Low 87 77
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - -
November High 15 22
Low 85 78
___________________________________________________
**Chi square significant at the 0.05 probability level.
Results from Tables 8 and 9 provide further support for the persuasion through
public involvement hypothesis. In Table 8, trust in media is compared between
markets that scored low in public journalism content and those that scored high,
regardless of the market's actual intent. There is no significant difference in
the level of public trust in media between these two categories in either August
or November. On the other hand, when low intent is compared with high intent,
regardless of actual content, there is a difference in public trust in media.
In August, there was a difference of five percentage points between the low and
high intent categories in those expressing high levels of trust in the media.
This result was significant at the 0.10 probability level. November results
showed a seven percentage point difference for those with high levels of trust
in the media between the two categories, and this result was significant at the
0.05 level.
Table 8 - Media trust in August and November for those in markets that had
content strongly reflecting citizen-based journalism (High Content) and those in
markets with weak citizen-based journalism content (Low Content) independent of
citizen-based journalism intent. Trust categories are Low (index values of 0
and 1) and High (index values of 2 and 3). Values are percentages.
Media Trust
Month Category Low Content High Content
___________________________________________________
August High 21 21
Low 79 80
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - -
November High 20 19
Low 80 81
___________________________________________________
Percentages within a column, with each month not adding to 100
are the result of rounding.
Table 9 - Media trust in August and November for those in markets that reported
strong desire to conduct citizen-based journalism (High Intent) and those in
markets with weak desire to conduct citizen-based journalism (Low Intent)
independent of citizen-based journalism content. Trust categories are Low
(index values of 0 and 1) and High (index values of 2 and 3). Values are
percentages.
Media Trust
Month Category Low Intent High Intent
___________________________________________________
August* High 18 23
Low 82 77
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - -
November** High 15 22
Low 85 78
___________________________________________________
*Chi square significant at the 0.10 probability level.
**Chi square significant at the 0.05 probability level.
Long-term results show what short-term results didn't. There was a positive
effect on media trust resulting from intent to involve citizens in the coverage
of the 1996 election. There was no effect on media trust resulting from actual
coverage of the election. Results found in this study, though, are not strong
enough to suggest that the persuasive effects of actual citizen involvement is
the end-all explanation for differing levels of media trust.
The results do indicate, however, that there is some persuasive effect
resulting from heavy citizen involvement in setting the agenda for election
coverage, which is reflected by intent to conduct citizen based journalism. The
data further show a lack of content effects. It is feasible, then, that content
effects are indeed overwhelmed by persuasion through involvement effects. Meyer
and Potter also reported a significant relationship between increased long-term
trust in media and citizen-based journalism intent, and they too failed to find
the link between content and public attitude toward media.[33]
Lack of content effects have other explanations as well. Public attitude
toward the media may change as the result of attitude changes by newsroom staff
toward the community, the news reporting, and the news gathering process. Bare
found that management's commitment to citizen-based journalism did, in fact,
result in attitude changes in the newsroom staff regarding their outlook on news
values and the place of the newspaper in the community.[34] Changes in
management and newsroom philosophy would not necessarily be reflected in
newspaper election coverage content, nor would they necessarily be reflected by
citizen-based journalism intent.
Pre-existing community attitudes and values may hold the greatest promise in
linking citizen-based journalism practice to changing public attitudes. It is
likely, given the enduring nature of public attitudes, that the level of trust
in the media, the government, and social capital were well established before
the news media began a public journalism program. In fact, the presence of
these attitudes in the community may have inspired the media to adopt
citizen-based journalism in the first place. The effect sought, then, becomes
the cause.
Putnam showed that in one region of Italy, the northern half, social capital
and trust in institutions pre-existed at greater levels than in the southern
regions. The pre-existing conditions made for a smoother, more prosperous
transition to democracy for the northern region.[35] Similarly, greater
pre-existing levels of trust in media would provide a firm foundation for a
transition to citizen-based journalism, which may provide an incentive for
newspapers to start a public journalism program. Greater backing by the public
should result in a more successful public journalism program. Success to the
newspaper would be increased circulation. Increased community involvement by
the public is one of the keys to increasing newspaper readership.[36] Success
for the public would be a newspaper that becomes a part of the community,
reflecting community values, helping the public work through issues, and
providing a forum for public expression.
Results should not be interpreted to suggest that a newspaper can simply
proclaim its intent to conduct citizen-based journalism and that will make the
public trust it more. It is probable that newspapers scoring higher in intent
had to actually carry out some of the projects mentioned. Whether or not the
promise keepers and nonperformers actually did any of the seven items listed in
the methods was not measured in this study. The nature of the items themselves,
though, do lend credence to the persuasion through participation idea. Five of
the seven involved actual citizen involvement, while only two involved content.
Practices that engage citizens in the news gathering process will not
necessarily be reflected in content, but would be reflected in intent. Perhaps
a future study could specifically address the persuasive effects of citizen
involvement tactics used in public journalism programs. By its definition,
citizen-based journalism should involve a heavy dose of community involvement
beyond getting public input in the news process. It would also be interesting
to study the persuasive effects of a newspaper's involvement in the community.
For instance, would public trust in the media improve as a result of news media
sponsoring community events, or initiating a crime prevention program? These
programs, like practices that include public input into coverage decisions,
would not necessarily be reflected in newspaper content.
It is also likely that those markets with strong intent to practice
citizen-based journalism had been doing so longer than the seven-week period
during which newspaper content was analyzed. Persuasion theory suggests that
attitudes, such as trust in media, are enduring and therefore would likely be
swayed more by a long-term commitment to citizen-based journalism as opposed to
picking up any effects of content, based on seven weeks worth of election
stories. It should be noted, however, that intent was calculated based on
coverage of the 1996 election and may or may not reflect any long-term
commitment to public journalism.
CONCLUSIONS
The persuasive effects of involving citizens in the news gathering process, as
reflected by measurement of citizen-based journalism intent, provided at least a
partial explanation as to the missing link, which is effects on citizen's trust
in media by citizen-based journalism content, in studies of public journalism.
Intent to conduct citizen-based journalism was shown to produce long-term
improvement in public trust of media while there was no apparent effect related
to citizen-based journalism content. Overall, however, the lack of effects on
citizen attitudes remains an enigma. Perhaps, as suggested by persuasion
theory, attitudes are enduring and the public will require longer exposure to
citizen-based journalism before it begins to lose the cynicism and mistrust that
have built over time. Content effects may also be overshadowed by other
effects, such as changes in newsroom attitude, and newspaper participation in
community events. Such things won't necessarily be reflected by newspaper
content. Public attitudes may be set at pre-existing levels. Communities with
greater trust and social capital may have inspired the local paper to adopt
public journalism, changing the effect to the cause. Appendix 1 - Frequency
tables for media trust in August (TRMEDA), November (TRMEDN), and the shift from
August to November (TRMEDS=TRMEDN-TRMEDA). Frequency for the categorized shift
variable is also included (TRMEDSC).
Actual Indexed Values
Value TRMEDA TRMEDN
________________________________________________________
0 281 293
1 212 210
2 117 106
3 13 14
________________________________________________________
Shift Values
Value TRMEDS
_______________________________________
-2 23
-1 123
0 350
1 108
2 17
3 2
_______________________________________
Value TRMEDSC
_________________________________________
-1 146
0 350
1 127
_________________________________________
Where -1 represents any negative shift, 0 represents
no shift and 1 represents any positive shift. REFERENCES
[1] John D. Bare, "Toward a Definition of Public Journalism," (Ph.D. diss.,
University of North Carolina at Chapel Hill, 1995); Davis Merritt Jr. and Jay
Rosen, Imagining Public Journalism: An Editor and Scholar Reflect on the Birth
of an Idea (Bloomington, IN: Roy W. Howard Chair, School of Journalism, Indiana
University, 1995): J. Rovner, "Covering Health Care: Are Reporters Missing the
Beat," Advances 3 (1996): 1-4.
[2] Citizen-based journalism is also referred to as public journalism, civic
journalism, community journalism, and communitarian journalism.
[3] Alicia C. Shepard, "The Gospel of Public Journalism," American Journalism
Review 16 (1994): 28-34.
[4] Keith R. Stamm and Lisa Fortini-Campbell, "The Relationship of Community
Ties to
Newspaper Use," Journalism Monographs 84 (1983): 1-27.
[5] Newseum et al, "News Junkies, News Critics," (study by the Newseum, The
Freedom Forum's Media Studies Center in New York, and the Roper Center for
Public Opinion Research at the University of Connecticut, 1997). URL:
http://www.newseum.org/survey/index/html; Robert L. Stevenson, "The
Disappearing Reader," Newspaper Research Journal 15 (1994): 22-31.
[6] Davis Merritt Jr., "Missing the Point," American Journalism Review 18
(1996): 29-31.
[7] Merritt, "Missing the Point," 30.
[8] Tony Case, "Public Journalism Denounced," Editor and Publisher 127 (1994):
14.
[9] Daniel Yankelovich, Coming to Public Judgment: Making Democracy Work in a
Complex World (Syracuse, NY: Syracuse University Press, 1991).
[10] Yankelovich, "Coming to Public Judgment," 3-4; Ed Rubenstein, "Right
Data," National Review 48 (1996): 20.
[11] Everett C. Ladd, "The Data Just Don't Show Erosion of America's 'Social
Capital,'" The Public Perspective 7 (1996): 1, 5-21.
[12] Robert M. Entman, Democracy Without Citizens: Media and the Decay of
American Politics (New York: Oxford University Press, 1989); Yankelovich,
"Coming to Public Judgment," 3-4; Robert D. Putnam, "Bowling Alone: America's
Declining Social Capital," Journal of Democracy 6 (1995): 65-79; Robert N.
Bellah, Richard Madsen, William M. Sullivan, Ann Swidler, and Steven M. Tipton,
"House Divided," preface to the 1996 edition of Habits of the Heart:
Individualism and Commitment in American Life (Berkeley, CA: University of
California Press, 1985, 1996) 7-39.
[13] Putnam's findings are disputed by Ladd who believes Putnam has
misinterpreted the data. Ladd presents data showing increased membership in PTA
and no substantial change in group memberships from 1975 to 1994. See Ladd,
"The Data Don't Show Erosion," 5-21.
[14] Putnam, "Bowling Alone," 65-79.
[15] Putnam, "Bowling Alone," 65-79.
[16] Mike Hoyt, "Are You Now, or Will You Ever be a Civic Journalist?"
Columbia Journalism Review (Sept./Oct. 1995): 27-33.
[17] Davis Merritt Jr., Public Journalism and Public Life: Why Telling the News
is Not Enough (Hillsdale, NJ: Erlbaum, 1995).
[18] Case, "Public Journalism Denounced," 15.
[19] Ed Fouhy and Jan Schaffer, "Civic Journalism--Growing and Evolving,"
Nieman Reports 49 (1995): 16-18.
[20] Philip Meyer, "Public Journalism and the Problem of Objectivity," (paper
presented at a conference of the Investigative Reporters and Editors, Cleveland,
OH, 1995).
[21] The People Project, among other things, produced voter guides, devoted
extensive coverage to issues, encouraged voter participation, provided
information on how people could become involved through volunteering, sought
public input as to what issues were most important to the community, and avoided
focusing on experts and politicians as story references. The People Project was
an expansion of the Voter Project, conducted by the Eagle and KAKE-TV.
[22] Shepard, "The Gospel of Public Journalism," 31.
[23] Philip Meyer and Deborah Potter, "Effects of Citizen-Based Journalism in
the 1996 National Election." (paper presented at the National Press Club
Seminar, Washington, D.C., 1997). Condensed version available at URL:
http://www.unc.edu/~pmeyer.
[24] Unpublished data produced by graduate students in Prof. Meyer's seminar.
Content variables studied included percentage of stories mainly about
candidates' criticisms of each other, and frequency and prominence of citizen
references--neither produced any significant effect on public trust.
[25] Shepard, "The Gospel of Public Journalism," 31.
[26] Shepard, "The Gospel of Public Journalism," 31.
[27] Genelle I. Belmas, Jennifer L. Lambe, and William A. Babcock. "Can News
Councils Help Newspapers Regain Public Trust?" (paper presented at the Southeast
Regional Conference of the AEJMC. New Orleans, LA., 1998).
[28] For a detailed description of the Hawthorne experiments, see Homans,
George C. (1963). "Group Factors in Worker Productivity." In Sociological
Research I: A Case Approach. Matilda White Riley, ed. New York: Harcourt,
Brace, & World, Inc.
[29] Daniel J. O'Keefe, Persuasion Theory and Research (Newbury Park, CA: Sage
Publications, Inc, 1994) 17.
[30] Meyer and Potter, "Effects of Citizen-Based Journalism," 1-5.
[31] Some markets had previously expressed, publicly, either disdain or support
for public journalism practices. Other markets were known to have a previous
commitment to public journalism.
[32] Garth S. Jowett and Victoria O'Donnell, Propaganda and Persuasion (Newbury
Park, CA: Sage Publications, Inc., 1986); O'Keefe, "Persuasion Theory and
Research."
[33] Meyer and Potter, "Effects of Citizen-Based Journalism," 8-9.
[34] Bare, "Toward a Definition of Public Journalism."
[35] Robert D. Putnam, Making Democracy Work: Civic Traditions in Modern Italy
(Princeton, NJ: Princeton University Press, 1993).
[36] Stamm and Fortini-Campbell, "Community Ties," 1-27; Jay Rosen, Community
Connectedness: Passwords for Public Journalism (St. Petersburg, FL: The Poynter
Institute for Media Studies, 1993).
|
