|
ABSTRACT
Religious Beliefs, Media Use, and Wishful Thinking
in the 2000 U.S. Presidential Election
A long-documented link exists between vote intention and a prediction of
who will win a political campaign, described in some research as wishful
thinking. A number of factors are thought to enhance the effect, such as
selective exposure to media content and similar others, and moderate the
effect, such as exposure to public opinion polls and knowledge about
politics. Religious factors were found to increase wishful thinking, an
impact moderated to a small degree by exposure to the news media.
Religious Beliefs, Media Use, and Wishful Thinking
in the 2000 U.S. Presidential Election
During a political campaign, it is not uncommon to hear people ask one
another who they think will win. While on its surface a relatively simple
question, one in which the respondent might sample the most recent news
stories and poll results, there exists a powerful tendency to respond with
the preferred candidate. More specifically, wishful thinking occurs when
an individual's personal preference exerts influence on the prediction of
some external outcome.
Almost sixty years ago, scholars noted "a close relationship between vote
intention and expectation of the winner" (Lazarsfeld, Berelson, & Gaudet,
1944, p. 106). A host of studies in various academic disciplines from
political science to psychology have identified and validated aspects of
this bias, teasing out numerous theoretical reasons for why it occurs and
exploring factors that either increase or decrease the effect.
At the core of wishful thinking is the theory of motivated reasoning,
which posits that people are selective in what they choose to believe and
will undertake various strategies that fit new information to their
predispositions (Dolan & Holbrook, 2001; Kunda, 1990). By relying on their
own position or on those of selected others who tend to share their
preference, individuals offer a biased prediction of an election
outcome. Given that people are more likely to be exposed to crosscutting
political viewpoints from the mass media rather than from interpersonal
discussion (Mutz & Martin, 2001), individuals with homogenous social
networks should be most likely to engage in wishful thinking. Here, I
examine whether individuals with strong religious beliefs or who are
identified as religious conservatives are more likely to engage in such
wishful thinking when asked to predict the winner of the U.S. presidential
election.
Wishful Thinking
At its simplest, wishful thinking is the link between a preference (wish)
and a prediction (expectation). Respondents are often asked, in a
political study, what candidate or party they prefer and to predict which
candidate or party will win or capture the majority of votes in an
election. Numerous studies have confirmed these distortions in predicting
the outcome of elections (Frenkel & Doob, 1976; Lemert, 1986; Regan &
Kilduff, 1988; Granberg, & Brent, 1983; Granberg & Brown, 1989; Granberg &
Holmberg, 1986, 1988). Outside the political realm, scholars have found
wishful thinking to occur in such diverse areas as among fans at sporting
events and among those placing bets on the outcome of some contest (Babad,
1983; Babad & Katz, 1991; Babad, Hills & O'Driscoll, 1992; Babad, 1995,
Babad & Yacobos, 1993). Examination of this preference-expectation link
also spans nations, with studies published from data gathered in Israel,
New Zealand, Sweden, and the United States.
Wishful thinking studies typically begin by establishing the explanatory
power of personal preference in predicting some outcome, then they attempt
to tease out factors that either enhance or moderate the effect. Intensity
of attachment or affect toward some outcome, such as a political candidate
or party, often leads to greater wishful thinking (Granberg & Holmberg,
1988; Babad, Hills, & O'Driscoll, 1992). Also, if an event is perceived as
within the control of the respondent, wishful thinking is greater (Rehm &
Gadenne, 1990). Theoretically, knowledge should moderate wishful thinking,
but the results are mixed (Babad, Hills & O'Driscoll, 1992; Babad, 1995;
Lemert; 1986; Dolan & Holbrook, 2001). Indeed, access to poll data
appeared ineffective in reducing wishful thinking and, instead, individuals
tended to misremember poll results in such a way as to support their
personal preference (Babad, 1997). Finally, there is some support for a
"neighborhood effect" in wishful thinking, where people who live in a
geographic area strongly supportive of one candidate will tend to see that
candidate as the likely winner.
While a number of processes have been offered to explain such a tendency,
many scholars point to information-retrieval biases in which people tend to
first access information that is readily available, positive in nature, is
more desirable, or is congruent with pre-existing beliefs and preferences
(Sanitioso, Kunda & Fong, 1990; Matlin & Stang, 1978; Yaacov, Gervey, &
Liberman, 1977; Dolan & Holbrook, 2001). Simply put, people tend to
believe what they want to believe in predicting some outcome.
The Role of the Communication
Exposure to newspapers and, to a lesser degree television news, has been
found in a large number of studies to be associated with greater knowledge
about public affairs. Despite the theoretical role such accurate
information should play in reducing wishful thinking, studies discussed
above make it quite clear that not only to respondents ignore such
information, they will alter it to fit their predispositions and preferences.
The few studies that directly address the role of the news media or public
opinion polls in wishful thinking find little to support the notion that
such content will sufficiently expose individuals to different opinions,
giving them a broader view of the world. Dolan and Holbrook (xxxx)
collapsed television news exposure and newspaper exposure into a single
variable ranging from 0 days to 14 days in their analysis of four U.S.
presidential elections. Given the varying effects of newspaper and
television news viewing on a wide range of political measures, including
political knowledge, this offers a number of methodological and conceptual
problems. Not surprisingly, no relationship was found between media
exposure and the ability to accurately predict an election outcome.
A separate analysis of U.S. presidential elections from 1952 to 1984
included no examination of possible media exposure effects, although it did
examine in the 1980 election a question asking about exposure to public
opinion polls (Granberg & Homberg, the polsys). The effect of
self-reported poll exposure was small but statistically significant,
suggesting media exposure may provide a moderating effect. Other studies
have found exposure to polls to influence expectations (Irwin & Van
Holsteyn, 2002; Johnston, et al., 1992).
Even fewer studies directly address the role of interpersonal communication
in wishful thinking, although research demonstrates that people tend to
associate most often with those who hold views similar to their own
(Calhoun, 1988; Huckfeld & Sprague, 1995). Interpersonal communication
networks, are thought to overwhelm any mass media influence in offering
public opinion polls or a sense of where general opinion stands on an issue.
Religion
While there is scant evidence that media content can moderate wishful
thinking, there is virtually no research that examines the potential role
interpersonal effects in general or religious beliefs in particular might
have on this bias. However, given the growing tendency toward avoidance of
crosscutting information in social networks in general, it requires little
effort to extend this to religious networks. McFarland and Warren (1992),
for example, demonstrated that Christian fundamentalists preferred to read
stories supporting their positions. Conservative Protestants tend to
selectively avoid media content and selectively seek out that which fits
their beliefs (Buddenbaum, 1982; Hamilton & Rubin, 1992; Roberts, 1983; see
Stout & Buddenbaum, 1996, for a discussion).
Members learn church norms through a series of influences: social
interaction, formal and informal training, ministerial guidance, and
restricted communication (Wald, 2003). If wishful thinking indeed is a
demonstration of selective exposure to similar others and biased processing
of available information, there is little reason to believe that
church-based social networks and beliefs would not act in a way found in
other, more geographic-based networks such as the "neighborhood effect"
discussed above.
What is not understood, however, is whether all religious factors operate
in the same way if indeed they do influence wishful thinking.
Implications and Research Questions
If religious networks operate in much the same fashion as other
interpersonal networks, then the influence of selective exposure (or lack
of exposure to alternate viewpoints) should lead to greater instances of
wishful thinking. Based on earlier discussion, I generated the following
research questions:
RQ1: Do religious factors promote wishful thinking?
RQ2: Does exposure to the news media moderate wishful thinking?
RQ3: Does news media exposure interact with religious factors to moderate
wishful thinking?
The first research question is fairly straightforward and asks whether
certain religious variables are associated with wishful thinking. These
factors, discussed in methodological detail later, include prayer, church
attendance, religious identification, self-identification as a born-again
Christian, the importance of religion in daily life, and perceptions of the
Bible. The second research question attempts to solidify knowledge about
exposure to the news media (newspapers, national television news, and talk
radio). The third research question combines these two factors to explore
whether, for some, exposure to mass media can reduce the likelihood of
wishful thinking.
Method
Archival data to study these research questions were drawn from the 2000
National Election Study, a survey of 1,807 respondents conducted before and
after the U.S. presidential election. This data set includes a large
number of questions on religious beliefs and behaviors, mass media use, and
items tapping not only presidential preference but also a prediction of the
election outcome.
Measuring Wishful Thinking
The national system under study can influence how scholars measure,
manipulate, and analyze the data. In nations with a parliamentary form of
government, wishful thinking is often the prediction of the margin of
victory one predicts for a political party. Studies of U.S. presidential
campaigns, however, focus instead on predictions of which candidate will
win the presidency. One recent study of four U.S. elections created a
dependent variable measuring the accuracy of a prediction (Dolan &
Holbrook, xxxx). That is, predictions of a likely outcome were juxtaposed
with the actual winner to create a measure of accuracy. The 2000 U.S.
election raises an interesting methodological problem: Who won? After
weeks of ballot counting and court battles, the Electoral College elected
Republican nominee George W. Bush as president. However, Democratic
nominee Al Gore captured the greater number of popular votes. Using
"accuracy" in this situation is ripe with problems so instead respondents
were coded as a "1" if they predicted their own candidate would win the
election (wishful thinking) and a "0" if they predicted another candidate
would win.
Religion and Media Use
I examined three media use questions: exposure to national television
news, newspapers, and talk radio programs. If Mutz and Martin (2001) are
correct, differences should be seen between the two mainstream news sources
and talk radio, which often focuses on more ideological content supported
by its listeners. The first two media exposure items are standard 0-to-7
scales on the number of days that medium is used. Talk radio exposure was
measured on a 5-point scale ranging from no self-reported exposure to
listening to such programs "every day."
Importance of religion is derived from two questions, the first asking
whether religion matters in a respondent's life and the second asking to
what degree it matters, resulting in a 1 (low) to 4 (high)
scale. Interpretation of the Bible is measured by asking respondents which
of the three responses best fit their personal beliefs, that the Bible is
the literal word of God, whether it is the inspired word of God written by
men and therefore not literal, or whether the Bible comes from men, not
God. Attending service is measured by a simple yes-no question on
attendance. Degree of prayer is measured by an item asking how often a
respondent prays and ranges from 1 (never) to 5 (several times a
day). Respondents who identified themselves as Christian were also asked
whether they consider themselves born-again Christians. Responses on this
were coded as "1" for born-again and "0" for all other respondents.
Finally, as a check on how strongly people feel toward a candidate,
affective response was measured with the traditional "feeling thermometer"
of 0-to-100 degrees for each respondent's favored candidate.
Results
Establishing Wishful Thinking
Did wishful thinking occur in the 2000 U.S. presidential
election? Absolutely. As Table 1 shows, the majority of Bush or Gore
supporters predicted their preferred candidate would be victorious. Even
respondents favoring "minor" candidates tended to engage in modest wishful
thinking, although the political leanings of respondents also led some to
predict that the winner would represent the political party they generally
supported.
Clearly preference is associated with prediction. To simplify subsequent
analyses and as discussed above, I combined these two questions into a
simple dichotomous score in which a "1" represented those who predicted
their preferred candidate would win the election (wishful thinking) and in
which "0" represented persons who predicted someone
Table 1: U.S. Presidential Election Preference by Predicted Winner
Who Will Win Election?
Gore
Bush
Buchanan
Nader
Other
Candidate
Preference
Gore
636 ( 11.9)
135 (-12.0)
0 (-1.6)
0 (-1.2)
14 (-1.0)
Bush
102 (-12.8)
548 ( 13.6)
4 ( 1.3)
0 (-1.1)
11 (-1.1)
Buchanan
0 (- 1.7)
6 ( 1.9)
0 (-0.1)
0 (-0.1)
0 (-0.4)
Nader
25 ( 0.7)
14 (- 1.3)
0 (-0.4)
3 ( 6.8)
1 ( 0.6)
Other
32 (- 0.4)
25 (- 1.2)
1 ( 1.7)
0 (-0.4)
10 ( 6.8)
X2=807.6, df=16, p<.001
Note: Entries are raw numbers, standardized residuals in parentheses.
other than their preferred candidate would win. This resulted in 1,507
respondents available for further analysis (some did not answer the
prediction or preference question). Of this group, 1,197 (or 76.4 percent)
engaged in wishful thinking.
Education moderated wishful thinking, an effect common in other studies,
while age was positively associated with wishful thinking. While education
is often used as a surrogate for cognitive ability, in this instance is
more likely represents a greater appreciation of other views or an ability
to see beyond personal preferences and accept other information (Granberg &
Holmberg, 1988). No other demographic factor was associated with the
effect. Respondents with strong partisan or ideological ties were more
likely to predict their own candidate as the likely winner.
How strongly respondents felt about their preferred candidate was strongly
associated with wishful thinking, a basic finding in previous research. In
addition, most religious variables were positively associated with
candidate affect. Affect was positively correlated with greater reading of
the Bible (r = .14, p<.01) and praying (r = .18, p<.01). In addition,
those who believe in a literal interpretation of the Bible also
demonstrated stronger affect toward their preferred candidate (M = 76.2)
than did those who either believe the Bible inspired but written by men (M
= 71.4) or that the Bible was not inspired (M = 64.1, all means different
at p<.001 level by t-test). Those who attend worship service had stronger
affect toward their preferred candidate (M = 74.0) than did those who did
not attend services (M = 66.7, t(1642) = -6.7, p<.001). Born-again
Christians reported higher affect (M = 75.0) than others (M = 70.0, t(1638)
= -4.7, p<.001).
Analyses of Research Questions
Table 2 reports the analysis of wishful thinking with selected religious
factors in which a chi-square is the appropriate test. Self-identification
as a born-again Christian was not significantly associated with wishful
thinking at this level of analysis, nor was attending worship
service. Evaluation of the Bible's origin and nature, however, was
associated with wishful thinking, suggesting that persons who believe the
Bible to be the literal word of God were more likely to engage in wishful
thinking than persons who either believe the Bible to be inspired but not
literal or those who believe the Bible to be only a work of man.
Analyses of variables using t-tests to explore differences in mean scores
found two results of interest. Respondents who described religion as being
important in their daily life were more likely to name their preferred
candidate as the winner of the election (see Table 3). How often a
respondent prays achieved a difference in the same direction near the
traditional level of statistical significance. In addition, none of the
media variables were associated with wishful thinking at the bivariate
level. Evidence above suggests that various aspects of religiosity can
influence wishful thinking, providing a positive answer to Research
Question 1. However, media exposure failed to moderate wishful
thinking. For the third research question, I examined whether media
exposure in tandem with religious factors could, for some individuals,
moderate the effect. The results were mixed. Table 4 provides examples of
the results, often with only one medium test when several media provided
similar results.
Among those who reported religion as being important to their lives, all
three media exposure items worked to moderate wishful thinking. High
exposure to
Table 2: Bivariate Analyses of Wishful Thinking and Selected Religious Factors
Predict Other
Will Win
Predict Own
Will Win
(wishful thinking)
X2
Not a Born-Again Christian
258 (0.8)
779 (-0.4)
Born-Again Christian
111 (-1.1)
408 ( 0.6)
2.3
Bible Is Literal
102 (-1.9)
419 ( 1.1)
Bible Inspired, not Literal
183 ( 0.2)
578 (-0.1)
Bible Written by Man, Not God
70 ( 2.5)
150 (-1.4)
13.0b
Do Not Attend Service
115 ( 0.5)
350 (-0.3)
Attend Service
253 (-0.3)
843 ( 0.2)
0.5
a p<.05 b p<.01
Note: Cell entries are raw numbers with standardized residuals in
parentheses. The residual signals how far off the actual number is as
compared to by chance. A large positive number, therefore, signifies more
respondents in a given cell than one would expect to find by chance.
Table 3: Bivariate Relationships Between Wishful Thinking and Selected
Variables
Predict Other
Candidate Win
Predict Own
Candidate Win
(wishful thinking)
t
Affect toward Candidate
59.4 (24.8)
75.9 (16.6)
-14.6c
How often Pray
3.4 ( 1.4)
3.5 ( 1.3)
- 1.6 +
Read Bible
2.0 ( 1.1)
2.1 ( 1.1)
- 0.8
Religion Important
3.9 ( 1.8)
4.1 ( 1.7)
- 2.8b
Newspaper Exposure
3.6 ( 2.9)
3.4 ( 2.9)
1.4
TV News Exposure
3.2 ( 2.7)
3.4 ( 2.8)
- 1.0
Talk Radio Exposure
1.8 ( 1.2)
1.7 ( 1.2)
0.5
+ p<.10 a p<.05 b p<.01
Note. Cell entries are mean scores with standard deviations in
parentheses. High scores mean respondents were more likely to pray, to
read the Bible, and think religion is important. For media variables, high
scores represent higher exposure.
newspapers (X2 = 5.4, df=1, p<.05), national television news (X2 = 6.1,
df=1, p<.05), and talk radio (X2 = 5.4, df=1, p<.05) led to less wishful
thinking among respondents categorized as being high in religious importance.
Among born-again Christians, greater wishful thinking was found among
those who did not read newspapers (X2 = 2.8, df = 1, p<.10), a result near
the traditional level of statistical significance. High newspaper exposure
resulted in no association between born-again Christian identification and
wishful thinking, moderating the effect. No other media variable
interacted with born-again status.
Respondents who believe the Bible is the literal word of God were
significantly more likely to engage in wishful thinking regardless of the
level of newspaper exposure (X2 = 7.1, df = 1, p<.05 for low-exposure
respondents and X2 = 6.7, df = 1, p<.05 for high exposure
respondents). The same basic result was also found for national television
Table 4: Summary of Selected Religion and Media Variables with Wishful Thinking
No Wishful Thinking
Wishful Thinking
X2
Newspaper Exposure Low
Religion Not Important
120 (-0.6)
445 ( 0.3)
Religion Important
47 (-1.1)
130 ( 0.6)
2.2
Newspaper Exposure High
Religion Not Important
144 (-0.9)
477 ( 0.5)
Religion Important
59 ( 1.8)
131 (-1.0)
6.0a
Talk Radio Exposure Low
Bible Literal
54 (-1.2)
223 ( 0.7)
Bible Inspired
99 ( 0.3)
318 (-0.2)
Bible by Man
35 ( 1.4)
85 (-0.8)
4.6+
Talk Radio Exposure High
Bible Literal
32 (-1.5)
132 ( 0.9)
Bible Inspired
65 ( 0.2)
182 (-0.1)
Bible by Man
27 ( 1.8)
48 (-1.1)
7.5 a
+ p<.10 a p<.05 b p<.01
Note: Cell entries are raw numbers with standardized residuals in
parentheses. The residual signals how far off the actual number is as
compared to by chance. A large positive number, therefore, signifies more
respondents in a given cell than one would expect to find by chance.
news exposure. The results for talk radio differed slightly. While a
finding near the traditional level of statistical significance was found
for those who listen less often to talk radio programs, among those who
listen more often, such programming was associated
with more wishful thinking for those who believe in a literal
interpretation of the Bible (X2 = 7.5, df = 1, p<.05). Taken together, the
mainstream media had less effect on moderating wishful thinking than did
talk radio, although the differences are small. No interaction was found
between the media variables and attendance of worship services, reading the
bible, or prayer.
Finally, how often do people discuss politics with those they share
religious beliefs compared to other likely social networks such as
neighbors, co-workers, and relatives? Here I provide early results of a
separate study that explores which networks are the most important for most
people. Respondents were asked if there was an individual with whom they
speak often about politics and then to identify this individual through a
series of questions. The results suggest that fellow church members are
dwarfed by other social networks such as co-workers (50.2 percent),
relatives (46.1 percent), and neighbors (28.7 percent). Only 12.2 percent
of respondents said the person they discuss politics most often with
attends the same church. However, it is likely many relatives in question,
particularly spouses, are also attend the same church. Further analysis
will explore these questions.
Implications
Exposure to diverse political views is the hallmark of a healthy
democracy, and yet evidence exists that for many persons, the conversation
is often among those who share similar beliefs and opinions. Wishful
thinking is one aspect of that process, the impact of personal preference
on the prediction of who will win a political campaign. While innocent
enough on the surface, the reasons for this robust psychological effect
raise interesting and troubling questions about how people selectively
expose themselves to information and to others who agree with their point
of view.
Simply put, our interpersonal social networks tend to encourage the belief
that all people think the same way we do. When asked to predict the winner
of an election, for example, we conduct a biased search of available
information and provide an answer that represents a limited sample: our own
opinion and that of similar others. Even accurate knowledge about a
campaign does little to attenuate this effect. Exposure to the mass media,
which provide public opinion poll information and providing alternate
viewpoints space and time to be heard, should moderate wishful
thinking. Previous studies have found little support for this hypothesis
and, indeed, little direct influence of media effects was uncovered
here. As such, the power of interpersonal communication to enhance wishful
thinking and the failure of mass media factors to moderate the effect does
suggest some threat to the notion of a pluralistic society envisioned by
John Stuart Mill. Wisdom, according to Mill, comes from "hearing what can
be said about it by persons of every variety of opinion, and by studying
all modes in which it can be looked at by every character of mind (Mill,
1859, p. 28).
Here, I examined wishful thinking from the perspective of religious
beliefs, an area little studied. Certainly for many, religious
affiliations are tantamount to representing an important—if not the most
important—social network. If indeed such interpersonal networks have an
unusually powerful influence on predictions through the selective exposure
to similar others, then one would expect those who feel most strongly about
their religious attachments would demonstrate the greatest wishful
thinking. Indeed this was true in certain cases, enough to warrant further
study. Media exposure should moderate wishful thinking by providing
information about the "other side" or the distribution of opinion from
polls, but no main effects for media use were found. However, for some
religious factors, media use could moderate the tendency to see one's own
preferred candidate as the likely winner.
What does this mean? Media exposure may for those with strong religious
beliefs, and one would assume strong social networks of a religious nature,
provide information outside the norm of interpersonal discussion or even
selective exposure to preferred channels of communication such as religious
television and radio programming, the Internet, or specialty newsletters
from churches and religious organizations.
References
Babad, E. (1987). Wishful Thinking and Objectivity Among Sports
Fans. Social Behaviour, 2, 231-240
Babad, E. & Katz, Y. (1991). Wishful Thinking—Against All Odds. Journal
of Applied Social Psychology, 21, 1921-1938.
Babad, E., Hills, M. & O'Driscoll, M. (1992). Factors Influencing Wishful
Thinking and Predictions of Election Outcomes. Basic and Applied Social
Psychology, 13, 461-476.
Babad, E. (1995). Can Accurate Knowledge Reduce Wishful Thinking in
Voters' Predictions of Election Outcomes? The Journal of Psychology, 129,
285-300.
Babad, E. & Yacobos, E. (1992). Wish and Reality in Voters' Predictions of
Election Outcomes," Political Psychology, 14, 37-54.
Babad, E. (1997). Wishful thinking among voters: Motivational and
cognitive influences. International Journal of Public Opinion Research,
9, 105-125.
Buddenbaum, J. M. (1982). News about religion: A readership
study. Newspaper Research Journal, 3, 7-17.
Calhoun, C. (1988). Populist Politics, Communication Media and Large
Scale Societal Integration. Sociological Theory, 6, 219-41;
Dolan, K. A., & Holbrook, T. M. (2001). Knowing Versus Caring: The Role
of Affect and Cognition in Political Perceptions. Political Psychology,
22, 27-44;
Kunda, Z. (1990). The Case for Motivated Reasoning. Psychological
Bulletin, 108, 480-498.
Frenkel, O. J., & Doob, A. N. (1976). Post-decision Dissonance at the
Polling Booth. Canadian Journal of Behavioural Science, 8, 347-350
Granberg, D. (1983). Preference, Expectations, and Placement Judgments:
Some Evidence from Sweden. Social Psychology Quarterly, 46, 363-368.
Granberg, D., & Brent, E. (1983). When Prophecy Bends: The
Preference-Expectation Link in U.S. Presidential Elections. Journal of
Personality and Social Psychology, 45, 477-491.
Granberg, D., & Brown, T. A. (1989). On Affect and Cognition in
Politics. Social Psychology Quarterly, 3, 171-182.
Granberg, D., & Holmberg, S. (1986). Preference, Expectations, and Voting
Behavior in Sweden's Referendum on Nuclear Power. Social Science
Quarterly, 66, 379-392.
Granberg, D., & Holmberg, S. (1988). The Political System Matters: Social
Psychology and Voting Behavior in Sweden and the United States (Cambridge:
Cambridge University Press).
Hamilton, N. F., & Rubin, A. M. (1992). The influence of religiosity on
television viewing. Journalism Quarterly, 69, 667-678.
Huckfeldt, R., & Sprague, J. (1995). Citizens, Politics, and Social
Communication: Information and Influence in an Election Campaign. New York:
Cambridge University Press.
Irwin, G. A., & Van Holsteyn, J. J. M. (2002). The influence of opinion
polls on expectations. Public Opinion Quarterly, 66, 92-104.
Johnston, R., Brady, H. E., Blais, A., & Crete, J. (1992). Letting the
People Decide: Dynamics of a Canadian Election. Stanford, CA: Stanford
University Press.
Lazarsfeld, P., Berelson, B., & Gaudet, H. (1944). The People's Choice:
How the Voter Makes up his Mind in a Presidential Campaign. New York:
Columbia University Press.
Lemert, P. (1986). Picking the Winners: Politician vs. Voter Predictions
of Two Controversial Ballot Measures. Public Opinion Quarterly, 50, 208-221.
McFarland, S. G., & Warren, J.C. (1992). Religious orientations and
selective exposure among fundamentalist Christians. Journal for the
Scientific Study of Religion, 31, 163-174.
Matlin, M. W., & Stang, D. J. (1978). The Pollyanna Principle: Selectivity
in Language, memory and Thought. Cambridge, MA: Schenkman.
Mill, J. S. (1859). On Liberty. London: Oxford University Press.
Mutz, D. C., & Martin, P. S. (2001). Facilitating Communication Across
Lines of Political Difference: The Role of Mass Media. American Political
Science Review, 95, 97-114.
Regan, D. T., & Kilduff, M. (1988). Optimism about elections: Dissonance
reduction at the ballot box. Political Psychology, 9, 101-107.
Roberts, C. L. (1983). Attitudes and mass media use of the Moral
Majority. Journal of Broadcasting, 27, 403-410.
Sanitioso, R., Kunda, Z., & Fong, G. T. (1990). Motivated Recruitment of
Autobiographical Memories. Journal of Personality and Social Psychology,
59, 229-241
Stout, D. A., & Buddenbaum, J. M. (1996). Religion and Mass Media:
Audience and Adaptations. Thousand Oaks, CA: Sage Publications.
Trope, Y., Gervey, B., & Liberman, N. (1997). Wishful Thinking from a
Pragmatic Hypothesis-Testing Perspective. In The Mythomanias: The Nature
of Deception and Self-Deception. Editor: Myslobodsky, M.S. Mahwah,
NJ:Lawrence Erlbaum Associates.
Wald, K. D. (2003). Religion and Politics in the United States (4th
Edition). New York: Roman & Littlefield Publishers Inc.
|
