Content-Type: text/html

Agenda Building and the 1992 Presidential Campaign:
Was it a failure to communicate or did the audience set the agenda?
 
by
 
L.M. Walters
Fulbright Fellow
Research Group for Communication Studies,
ELTE and the Hungarian Academy of Sciences
Associate Professor of Jounrlism
Texas A&M University
201 Reed McDonald Hall
College Station, Texas
409-845-4611
 
&
T. Walters
Visiting Professor of American Studies
Budapest University of Economic Sciences
Language Institute
20 Horansky Utca
1129 Budapest
Hungary
36-1-209-27-20
 
As the time draws near for yet another Presidential election, campaign
organizers are again searching for the "golden formula" to communicate
effectively with voters.
Because political campaigns are, above all, exercises in communication, they
require many vehicles to transmit issue and image salience to voters (Kingdom,
1966, p. 109; Walters, 1994). Mindful of this, the communications mix of a
sophisticated campaign employs advertising and public relations techniques as
part of an integrated political strategy.
An element of that mix is the press release. Properly used, releases have the
potential to amplify themes and images stressed by the campaign. Some advocates
believe that they can help stimulate media coverage to further advance those
themes and images (Mitchell, 1992; Ver-meer, 1982, p. 145; Gaby, 1980).
Political public relations practitioners believe, as do others in the field,
that media placements of press release material bring legitimacy and attention
to issues favorable to campaign and candidate. Legitimacy flows from the media,
from the credence and presumption of impartiality the public attaches to Rnews.S
Attention comes from the pressU focus on a particular topic, the addition of the
subject involved to the media agenda. Practitioners believe that this
necessarily means the issue or entity is added to the public agenda which puts
it in a position to shape public opinion (Turk, 1985, p. 12; Walters & Walters,
1992; Walters & Walters, 1994) and official action (Sigal, 1973, p. 135).
Believing in this power of the press, both 1992 presidential campaign staffs
made a concerted effort to focus public attention on desired topics by
distributing issue-oriented press releases to the media.
But, were the two campaigns successful?
This paper examines a census of those releases sent to electronic media outlets
to determine whether they helped establish topic salience for the registered
voting public.
In doing so, Failure to Communicate sheds light on the process of agenda
creation and concludes that researchers should pay more attention to the role an
audience plays in the process. Because of media evolution towards information as
a commodity, this audience plays a larger role than previously thought of or
accounted for.
The audience's selection of key issues in the campaign, in fact, may have been a
pivotal factor in determining the outcome of the presidential election.
 
 
Previous Research
Public relations practitioners are not alone in their belief in the power of the
media to set the public agenda. Since the McCombs and Shaw (1972) study of the
1968 U.S. presidential election, agenda setting research has tested the
proposition that media emphasis on a topic results in public concern with this
topic (McCombs & Shaw, 1972; Shaw and McCombs, 1977; and see also agenda-setting
festschrift in Journalism Quarterly, 69(4).
Many are executed by erecting and comparing rank measures of one set of agendas
with similarly constructed rank measures of another. Such examinations are based
on the assumption that the Rmedia set the public agenda of issues by filtering
and shaping reality rather than by simply reflecting it (Weaver & Elliott, 1985,
p. 87).
Despite promising beginnings, the nearly quarter century of research that has
followed the 1972 study has produced mixed results. One reason is that this
theory usually cannot establish either causality or time order. Thus, while
agenda setting posits that the media agenda influences the public's agenda, it
is equally plausible that the public agenda influences the media agenda (Severin
& Tankard, 1979).
Moreover, the agenda setting process does not always "work." This failure has
been variously attributed to such factors as media usage, contingent conditions
affecting the audience and the actors, and the level of obtrusiveness of the
issues involved (Graber, 1984; Lang & Lang, 1981).
Whatever the case, this ambiguity has led to the search for constructs that
either complement or supplant the theory. One steam of research has led scholars
behind agenda setting to agenda building.
In what McCombs (1992) has called the "fourth phase" of agenda setting studies,
researchers have been attempting to answer the question "Who sets the media's
agenda."
This should have been among the first questions examined by researchers looking
at issue selection by significant publics. The answer should re-veal something
about the relationship between audience interest in, and media attention to,
issues. Preliminary indications are that this relationship appears to be more
complex than that posited by agenda setting studies.
Gandy (1982) attempted to answer that question by examining the way in which the
media interact with other, non-press, actors to create the items that eventually
appear on the media agenda. According to Gandy, those actors include public
information officers, governmental and company spokes-persons and other
professional public relations practitioners. They make up an integral component
of the newsgathering process, since they are the initial source of much of what
appears in the pages of newspapers and on the airwaves (Gandy, 1982; Walters &
Walters, 1992; Walters, Walters, & Starr, 1994).
 
Because these sources have different philosophies, they raise different issues,
focus on different aspects of a story and choose to promote certain perceptions.
This means that source selection helps determine not only who will be given
voice, but also what that voice will be allowed to say.
Certainly, the selection of paper sources is as important to agenda building as
the selection of people sources. And, Walters and Walters (1992), Walters and
Walters, (1994), Walters, Walters, and Starr (1994), Altheide (1985), Hale
(1978), Sachsman (1976), Kaid (1976), Rings (1971), and Glick (1966) all have
provided evidence that suggest written publicity materials, such as press
releases, provide a large part of news content.
Despite the importance of such documents in defining issues and creating images,
few researchers have looked into the press releases of major national election
campaigns. Research on public relations output within the context of a political
contests has been limited, in part, because the press release has not been
treated as important contributor to the process of building the political
agenda. Much of the work that does look at such materials seriously is dated,
tied to concepts such as RThe Selling of the PresidentS of the 1970s, or slanted
to anecdotal considerations of subjects like Rspin doctors.S Quantitative
research, such as it is, is uneven and has dealt mostly with local or state-wide
elections.
Another reason that scholars may have neglected to examine releases is that they
could not secure an adequate sample of the output of competing candidates.
The data set used in this study is different. Examined here are all the press
releases issued by the major party candidates in the critical final month of
the1992 presidential campaign. These materials provide an opportunity to assess
complete articulated political agendas in a visible, vituperative, and public
relations-dependent presidential campaign..
Failure to Communicate also accounts for audience preferences, as it attempts to
explain the relationship of that group to the the agenda building process in an
election campaign. The study does so through an examination of three factors: 1.
the issues raised in the press releases disseminated by offices of the two major
presidential candidates during October, 1992; 2. the issues identified as
important to the voters and 3. the registered voter's candidate preference.
 
 
Research Questions
 
1. What issues do Bush-Quayle and Clinton-Gore campaign press releases identify?
 
2. What are the rank orders of the issues identified in the Bush-Quayle and
Clinton-Gore campaign press releases? How do these ranks orders compare with
each other and with those of surveyed registered voters.
 
3. What is the relationship between the percentages of registered voters who
prefer a candidate and the issues identified in the press releases of the
candidates?
 
 
Method
To answer these questions, three sets of data were examined. The first was
composed of a census of original press releases from the Bush-Quayle and
Clinton-Gore campaign offices. All distributed during the critical month of
October, 1992, these releases were received by the top-ranked all-talk radio
station in a large Texas metropolitan area.
These releases were scanned and formatted for use in Microsoft Word 4.0 using
Omni-Page Professional. They were analyzed in Grammatik 4.0, a grammar-checking
and analysis program designed for the office environment. Systat 3.0 was used to
develop descriptive and inferential statistics.
The second set of data established the issues that were of interest to the
voting public. The issues used for comparison of the agendas of the campaigns
with those of registered voters were drawn from a Gallup Poll taken between
October 8 and 9 1992. This sample included 775 registered voters. They listed
job creation, economic growth, taxes, the federal budget deficit, crime,
education, foreign policy, health care, and the environment as the top 10 issues
of the election.
The third set of data determined the candidate preference of the voting public.
Polling percentage preference data were drawn from two sources: a compendium of
trial heats given in the November /December 1992 issue of The American
Enterprise and The Gallup Poll Monthly for October and November 1992. When the
results of more than one poll were listed for a given day, the polls were
averaged and that average was used as an expression of preference. This
expression of preference provided the Percentage Who Preferred a Candidate on a
Daily Basis.
To measure the difference between the major party candidates, two percentages
were calculated. First, was the Daily Percentage Difference between the
candidates, which was Clinton preference percentage minus the Bush preference
percentage. Second, to eliminate the volatility of daily percentages, a
Three-Day Rolling Average Percentage Difference between the candidates was
calculated. This was the average of three days totals. The first percentage was
the average of the last two days of September and the first day in October. The
rolling average moved forward from this date.
For this study, personal qualities, including character, the draft, Iran, Iraq,
and first and last names of the candidates, were also drawn from questions asked
in the Gallup Poll and added to the basic ten issues. From this enlarged list, a
dictionary of single key words for each issue was constructed. Then, a single
key word dictionary was bounced by computer against the original press release.
The resulting figures were the count of word occurrences for each issue and a
percentage of the total number of words in the original document. All issue
categories were single word with the exception of the first and last names of
the candidates. These were entered as George and Bush and Bill and Clinton. For
ease of analysis, counts and percentages were computed, and collapsed into the
categories Bush Total and Clinton Total.
 
 
Results
Running the dictionary file against the original press releases produced a rank
order of issues for each campaign. This rank order was the same regardless of
whether the numbers so generated were total word mentions or the percentage of
total words in the releases. (See Table 2.) The Bush-Quayle releases ranked The
Other CandidateUs Name first, followed by (2) Taxes, (3) CandidateUs Own Name,
(4) Character, (5) Economy, (6) Jobs, (7) Foreign Affairs, (8) Environment, (9)
Health, (10) Budget, (11) Education, and (12) Crime. The Clinton-Gore releases
ranked The Other CandidateUs Name first, followed by (2) CandidateUs Own Name,
(3) Taxes, (4) Economy, (5) Jobs, (6) Health, (7) Budget, (8) Foreign Affairs,
(9) Character, (10) Crime, (11) Education, and (12) Environment. A SpearmanUs
Rank Order Correlation between the issue agendas of the major party candidates
during October of 1992 Presidential election campaign was .774.
To compare the candidatesU agendas with that of the registered voting public,
the top issues of the day were drawn from a Gallup Poll conducted between
October 8 and 9 in 1992. Be-cause The Other CandidateUs Name and CandidateUs Own
Name did not appear in this poll, the two were deleted, and the re-maining 10
categories were compared between the candidates and the registered voting
public.
Looking at the electionUs top issues, critical differences between the campaigns
and between the campaigns and the registered voting public emerge. (See Table
2.) While registered voters ranked the economy as the number one issue, Clinton
campaign press releases rated it second, and BushUs rated it third. The voters
rated health second, the Clinton camp rated health fourth, and BushUs rated it
seventh. The voters rated jobs the third most important issue, ClintonUs also
rated it third, and BushUs rated it fourth. The voters rated education fourth,
Clinton and Bush both rated it ninth. Finally, registered voters thought the
budget was the fifth most important issue, Clinton rated also it fifth while
Bush rated it eighth.
As might be drawn from this comparison, the issue agendas of the candidates
differed from those of the registered voters. Using the ten issues, the
SpearmanUs Correlation between the campaigns was .600. A SpearmanUs Rank Order
Corre-lation run among Clinton, Bush, and the voters issue agendas shows that
Clinton at .552 more closely matched that of the registered voters than did Bush
at -0.018 when examined for all of October.
When SpearmanUs Rank Order Correlation was computed on a weekly basis,
differences appear when the registered voters agenda (established by the October
8-9 poll) becomes a basis for analysis. For Bush, the correlations to the
voterUs agenda were: Week 1, .291; Week 2, -.562; Week 3, .616; and Week 4,
-.190. So, for two of the four weeks, the Bush campaign agenda was negatively
related to the voterUs issue agenda. Only during the first and third weeks were
SpearmanUs Rank Order Correlation positive.
For Clinton, the relationship was reversed. For Clinton, they were: Week 1,
.247; Week 2, .321; Week 3, .218; and Week 4, .189. Thus, for all of the four
weeks, the SpearmanUs Rank Order Correlation was positive. (See Table 3.)
Bush began to close the percentage gap with Clinton after the third debate
during the third full week of the month. This was a time period during which the
Bush issue agenda more closely matched the VoterUs agenda that did the Clinton
agenda. On October 26, Bush gave a RJames A. Baker Speech,S indicating that
Baker would become the domestic policy czar. For a while Bush gained ground, but
then lost the impetus when his campaign turned to taxes and character as main
issues on October 28, 29, and 30. During this period, BushUs chances slipped
away as his campaign moved from the domestic issues dear to the hearts of
registered voters, and Clinton stayed with those issues, talking mostly about
jobs and the economy. (See Table 4.)
During these last days, the SpearmanUs Rank Order of BushUs agenda fell to .-190
relative to that of Registered Voters while ClintonUs was .189. Bush lost the
impetus of the Baker speech, began falling away, and then lost the election
because his campaign did not pay attention to issues of concern to Registered
Voters.
Besides using the key words in press releases to compare rank orders of issue
agendas among the candidates and with the voters, these words may be correlated
to indicators of preferences for an individual presidential candidate. Such
measures could include the Percentage Who Preferred a Candidate on a Daily Basis
and the Daily Percentage Difference between the candidates. A Three-Day Rolling
Average Percentage Difference between the candidate may be used to smooth out
potential volatility in daily figures.
These figures add to an ex-planation of how the election was won and lost. Based
on their un-derstanding of the process, the Bush reelection camp had hoped to
set an issue agenda for registered voters based on taxes, character flaws, the
economy, jobs, and foreign affairs as its top five issues. Their attack campaign
was wedded to these issues, even though, as the SpearmanUs Rank Order
Correlations have shown, the registered voters agenda differed remarkably from
the Bush camp. (See also Walters & Walters, 1994.)
Because the Bush and registered voterUs agenda differed so greatly, the Pearson
Product Moment Correlations between the Percentage Who Prefer, the Daily Percent
Difference, and the Three -Day Rolling Average Difference and the Issues (as
measured by key word process) were generally low. Relative to the Percentage Who
Prefer, only one positive correlation was more than .40, one was more than .30
and three were greater than .20. These figures were worse for Daily Percent
Difference (only one correlation greater than .20) and Three-Day Rolling Average
(one greater than .50, one greater than .30; and one greater than .10)
Many correlations were negative, indicating that the more the Bush addressed an
issue, the more likely registered voters were to turn away from him. Of the ten
total correlations for Percentage Who Prefer, four were negative; for Daily
Percent Difference, four were negative; and for Three-Day Rolling Average, six
were negative. (See Tables 5 and 6, previous page.)
As outsiders, the Clinton camp campaigned as agents of change who were in touch
with the people and who would not radicalize the government (See Walters &
Walters, 1994.) Their agenda identified more closely with that of the registered
voters. Compared to the Bush camp, the Clinton campaign had more positive and
stronger correlations between campaign identified issues and Percentage Who
Preferred a Candidate on a Daily Basis Daily, the Daily Percentage Difference
between the candidates, and the Three-Daily Rolling Average Percentage
Difference between the candidates. Several moved above the .60 mark, and, across
the three categories, only three issues correlations were negative. This
contrasts with the Bush campaign that 14 negative correlations. (See Tables 5
and 6.)
A SpearmanUs Rank Order Correlation matrix of the ordered correlations of issues
for Percent Who Prefer, Daily Percent Difference, and Three-Day Rolling Average
Difference and the candidateUs agenda demonstrate visually startling differences
between the two campaigns. (See Table 7.) These figures also show the relative
message consistency of the two campaigns over the course of October.
What emerges is the inability of Bush-Quayle to use press releases to redefine
the issue agenda order for the registered voting public. Conversely, the
Clinton-Gore enjoyed success by employing their releases to match its issue
agenda order to that of registered voters. When Clinton-Gore stayed from this
path, they lost potential voters.
When Daily Percent Differ-ence, Rolling Difference, and Those Who Prefer were
looked at as dependent variables that were product of important issues mentions,
the agenda building process of this election becomes clearer. If the top five
issues are looked at with respect to Multiple R-squared, the single instance in
which a candidateUs R2 was larger than that for voter related issues was for
Rolling Difference, and those differences were only marginal. In that instance,
BushUs top five had an R2 of .55 and the VoterUs had an R2 of .54. The single
instance in which a candidateUs p. for top 5 issues was better than (that is,
less than) that of the voterUs agenda was also for Rolling Difference, which was
.024 for BushUs top five issues and .026 for the Voters. (See Table 8.)
Even this small difference is deceptive if not examined in context. While the
CandidateUs R2 for Bush were higher, three of his top five issues had negative
standardized betas. These negative betas were -.612 for jobs and -.064 for both
foreign affairs and taxes. For the voterUs top five issues, these negative betas
were -.666 for jobs and -.504 for education.
When the analysis included the ten campaign issues, the same patterns were still
evident. Both R2 and p. were higher for Clinton across Daily Difference and
Percent Who Prefer, and lower with respect to Rolling Difference. In the Rolling
Differ-ence, Bush had an R2 of .636 and p. of .196, and Clinton had an R2 .514
and p. of .547, but, just as in the case of the top five issues, Bush had
several negative betas. These included: -.751 for Job, -.560 for Environment,
-.435 for Education, -.348 for Taxes, and -.041 for Foreign Affairs. Of these,
Job and Edu-cation were third and fourth res-pectively on the VoterUs Agenda.
Clinton had three negative betas including -.752 for Foreign Affairs, -.544 for
Education, and -.008 for Health. Of these, Health was second and Education was
fifth on the VoterUs Agenda.
Likewise, Bush suffered from differences with respect Daily Difference, Rolling
Difference, and Percentage Who Prefer across the Whole Agenda and the
CandidateUs and VoterUs Top Five Agendas. These figures could be interpreted to
indicate a number of things, including weakness of party regular support for
Bush or enmity felt toward him as a Tory politician and/or support for Clinton
as a outsider, in touch with the people who could concentrate on domestic
issues. A look at standardized betas across all variables and levels of analysis
supports this latter notion, as this was an election in which domestic issues
were judged most important by the registered voters.
 
 
Discussion and Conclusion
This study looked at campaign press releases distributed by the staffs of the
two major presidential candidates during October, 1992, to see 1) what issues
the candidatesU materials identified as important, 2) how the issue agendas
constructed from the releases compared to each other and to that of the
registered voters, and 3) the relationship between issues and the preference
expressed by the voters for a candidate.
The results indicate that the Bush-Quayle and Clinton-Gore teams had differing
views on the importance of issues on which to campaign. They also differed in
the ability to manage that agenda.
Generally, the Bush-Quayle campaign used an issue agenda that was based on an
offensive strategy. It concentrated on attacking ClintonUs character flaws,
including his moral lapses, his failures to reveal truthful information to the
public and especially his attempts to evade the draft. The Republicans also
lashed out at their opponents for their personal lack of expertise in foreign
policy and their partyUs reputation for Rtax and spendS legislation.
The Clinton-Gore campaign had a different strategy. As befits the outsiders
trying to unseat a popular incumbent, the Dem-ocrats steered away from personal
attacks, rarely mentioning concerns based on character. As befits self-professed
Rpolicy wonks,S they focused on the issues. Those they chose had a domestic
orientation. Their releases mentioned topics such as increasing employment,
improving the economy and reforming the health care system. These materials
pictured the challengers as agents of positive change, more in touch with the
needs of the people than the Republican incumbents.
The results of this study indicate that the Clinton-Gore team was correct in its
assessment; the Democrats, indeed, were closer to the public, at least in terms
of topics mentioned in their campaign materials. Every measure used here shows
that the issue agenda of Clinton-Gore was closer to the issue agenda of the
public than was that of the Republican rivals.
This indicates that the Clinton-Gore campaign did not set an agenda for the
voters. Rather, it matched the agenda of the voters. Doing so may have
contributed to the DemocratsU popularity. Through agenda matching, the
Clinton-Gore team maintained a lead in the presidential preference polls,
slipping only during the third week of the month. It was at this time that
Bush-Quayle more closely mirrored the votersU concerns than did their opponents.
When Clinton-Gore moved back into sync with the voters during the fourth week of
October, they again began to pull away from the Republicans. They remained the
leaders for the next few days, thus winning the election and delivering the
Executive Branch of the government into the hands of the Democrats.
This study reveals as much about the agenda creation process as it does about
successful campaign strategies. From the beginning, examinations of this process
have suffered at least two major flaws. The first is the assumption of Rmedia
tropismS in determining time order. The second is that these studies usually do
not account for the rise of Rmarketed media.S
Early studies of agenda creation posited that the process is media tropic, that
is, the audience grows to topics as delineated by the vehicles of mass
communication. This process did not always Rwork;S numerous researchers found
instances in which the publicUs agenda did not mirror that of the media. This
was ascribed to a variety of factors, including media usage, obtrusiveness of
the issues, or other contingent conditions. Usually, little or no regard was
given to the audience as an active, economic force that helps determine media
content. What we may have here may not be a failure to communicate, but a
failure to acknowledge the true communicators, an active, empowered public.
This lack of recognition of the strength of audience control over issues is
wedded to a public utility view of media. Founded on a notion that media leaders
believe in, and practice, a broader notion of public good, such a view envisions
the media benevolently assembling an agenda of issues for the consideration of
the audience.
Even from the beginning of agenda building studies, such an assumption may never
been entirely true. Although examples of crusty editorial types with a sense of
the public good abound, successful media businesses have always depended on
matching content to audience need. Those businesses that succeeded were able to
more closely match the media product with the interests of a critical segment of
the population.
Maximizing cost effectiveness reaching a target audience required that editorial
RfeelS be replaced with full-scale marketing departments. These departments
know what the audience wants and match output to meet those needs. Audience are
tested and probed through surveys, focus groups and audiences. And, as observers
such as James D. Squires, former Chicago Tribune Editor, have noted,
sharkskin-suited executives have taken over (Quoted in Soley, 1993). The new
media moguls, who regard content as software for the media vehicle, are more
likely to have MBAs than reporting experience.
Now more than ever, managers tailor the media product to meet the needs of
targeted groups. Today, this marketing of media is swift and profound. It means
that audience and readers quickly help determine media content and issue
ordering.
Programming formats such as talk and reality-based news and information are
proof of this. And it seems that the process of making the audience more
powerful in, and responsible for, agenda building is continuing un-abated.
Indeed, one Indiana Tele-vision station is even experimenting with letting an
audience vote by telephone on the structure of the dayUs primary news cast.
While such extreme examples indicate that the audience has an more active role
than ever before in this process, they are only the extension of a dynamic trend
that has been developing for years.
Consider that one reason Ronald Reagan was regarded as the great communicator
was that his public relations staff used a notion called Rprecincts of
perception.S ReaganUs team attempted to change what the public felt about Reagan
relative to an issue, instead of merely attempting to establish an issue order.
In doing this, the public relations practitioners recognized, as do all good
marketers, that bands of virtual publics linked not by proximity, but by
interests and attitudes, define the marketplace.
The marketed public perspective not only recognizes that the agenda building
process changes in harmony with the media business, it also changes in concert
with society. Just as the concept of a homogeneous melting pot has given way to
the view of a socially diverse salad bowl, and as the marketplace of ideas has
become the menu of ideas, so too are there changes in the public and its power
Historical forces have brought the United States to a position that emphasizes
diversity over conformity and subgroups over large populations. Clusters of
smaller scale critical mass publics are assuming greater influence in defining
content, and thus issue importance for the public relations practitioner and the
media.
No doubt the power relationship involving publics, public relations
practitioners and the media is a dynamic one. Whether such changes are linear
(and permanent) or cyclical is unclear. But, some, including historian Arthur
Schlesigner, Sr., have viewed these types of events in terms of great cycles.
Perhaps communications researchers would benefit by borrowing not just from
mathematicians and statisticians, but from differing historical and
philosophical perspectives as well.
Studying the problem will not be easy, because applying a cyclical concept to
the agenda building process will require not only longitudinal study, but also a
research perspective that views results in a different manner. This perspective
values exploration, not exclusion. A failure to discover direct agenda building
should not be discarded. Rather, it should be examined with respect to the
influence of the public and the marketing process. Whatever the cause, failures,
like successes, have value, particularly be-cause agenda building involves a
large number of actors.
If there is a failure in this particular study it is that all such actors were
not included. It would have been useful to include media output, such as radio
broadcasts or newscasts based on the press releases. However, the upcoming
presidential election provides the opportunity to test an equation accounting
for public, public relations practitioner and media all in one package.
References
 
 
 
 
Altheide, L. (1985). Media Power
Beverly Hills, California: Sage Publications.
 
Gaby, D.M. (1980). Politics and
Public Relations. Public Relations Journal,36(10):10-12.
 
Gandy, O. H. (1982). Beyond
Agenda Setting: Information Subsidies and Public Policy. Norwood, New Jersey:
Ablex Publishing.
 
Gans, H.J. (1979). Deciding
WhatUs News: A Study of CBS Evening News, NBC Nightly News, Newsweek and Time.
New York: Pantheon.
 
Glick, E.M. (1966). Press
Government Relationships: State and H.E.W. Departments. Journalism Quarterly,
43(1):49-56.
 
Graber, D.A. (1984). Processing
the News: How People Tame the Information Tide, New York: Longman.
 
Hunt, T. ( Ruben, B.D. (1993)
Mass Communication. New York: Harper Collins.
 
Kingdom, J.W. (1966).
Candidates for Office: Beliefs and Strategies. New York:  Random House.
 
Kaid, L.L. (1976). Newspaper
Treatment of a CandidateUs News Releases. Journalism Quarterly, 53(1):135-137.
 
Lang, G.E. & Lang K. Watergate:
An Exploration of the Agenda Building Process. In G.C. Wilhoit &  H. deBok
(eds.) Mass Communications Review Yearbook (2-16.) Newbury Park, California:
Sage.
 
McCombs, M.E. (1972). The
Agenda Setting Function of the Mass Media. Public Opinion Quarterly, 36:176-185.
 
 
 
 
 
 
McCombs, M.E. (1992). Explor-
ers and Surveyors: Expanding
Strategies for Agenda-Setting Research. Journalism Quarterly, 69(4):813-824.
 
Mitchell, G. (1992, May 30).
Media Politics -- How It Began. New York Times,Section 1, Page 19.
 
Rings, W.L. (1971). Public School
Coverage With and Without P.R. Directors. Journalism Quarterly, 48(1):62-67.
 
Sachsman, D.B. (1976). Public
Relations Influence on Coverage of Environment in San Francisco Area. Journalism
Quarterly, 53(1):54-60.
 
Severin, W, and Tankard J
(1979).Communication Theories. New York: Hastings House.
 
Shaw., D., & McCombs, M.E.
 1977). The Emergence of American Political: The Agenda Setting Function of the
Press. St. Paule: West Publishing.
 
Sigal, L.V. (1973). Reporters and
Officials: The Organization and Politics of Newsmaking. Lexington,
Massachusetts: D.C. Heath
 
Soley, L. (1993, September)
Utne Reader, 59: 44,47.
 
Turk, J.V. (1985).Information
Subsidies and Influence. Public Relations Review, 11(3):10-25.
 
Turk, J.V. (1986). Public Rela-
tions' Influence on the News. Newspaper Research Journal, 7(4):15-26.
 
Vermeer, J.P. (1982). For Im-
mediate Release. Westport, Connecticut: Greenwood Press.
 
 
Walters, L.M.& Walters, T.N.
(1992) Environment of Confidence: Daily Newspaper Use of Press Releases. Public
Relations Review, 8(1):31-46.
 
Walters, T.N., & Walters, L.
(1994a). Hobgoblins of Fear: Campaign Press Releases and the Issue Imagery of
the 1992 Presidential Election. College Station, Texas: Texas A&M University.
 
Walters, L.M., & Walters, T.N.
(1994b). The Four Seasons: Patterns of Cyclical Success in Public Relations
Placements. Southwestern Mass Communi-cation Journal, 10(1):86-97.
 
Walters, T.N., Walters, L., & Starr,
(1994). After the Highwayman: Syntax and Successful Placement of Press Releases
in Newspapers. Public Relations Review, 20(4:345-356,
 
Walters, T.N. & Walters, L.
(1995).A Triangulated Needs Assessment Survey of Hungarian Journalism Education,
Preliminary Report. American Journalism Center, Budapest.
 
Walters, L. (1995). The Good, the
Bad, and the Media: The Effects of Mass Communica-tion. Zeszyty Prasoznawcze,
No. 2 Krakow, Poland: The Press Research Center: Jagellonian University.
 
Walters, T. (1995). Media Ethics
and Law. Zeszyty Prasoznawcze, No. 3. Krakow, Poland:The Press Research Center:
Jagellonian University
 
Weaver, D.E. & Elliott, S.N.
(1985). Who Sets the Agenda for the Media? A Study of Local Agenda Building.
Journalism Quarterly, 62(2):87-94.