Persuasive Essay
lovlyl
where_is_my_issue.pdf
Journal of Broadcasting & Electronic Media/December 2008
Where Is My Issue? The Influence of News Coverage and Personal Issue
Importance on Subsequent Information Selection on the Web
Young Mie Kim
Researchers believe that the Web functions to supplement traditional news
media. Little is known, however, about how traditional news media consump-
tion influences Web use patterns. This study investigates how prior TV news
exposure influences individuals’ subsequent Web use by testing 3 theories that
may explain individuals’ information selection patterns—accessibility, instru-
mental utility, and personal issue importance. The results of this study reveal
the strong effects of personal issue importance when selecting information on
the Web, regardless of news coverage in traditional media. The findings also
indicate higher levels of information selection when there is no prior exposure
to news coverage.
In the past few decades, the Web has grown exponentially and become an
important source of political information. Recent statistics show that in 1 month
alone (March 2007), approximately 210 million people in the United States used
the Internet at least once (Nielsen/NetRatings, 2007). During the 2004 election,
75 million Americans (37% of the adult population) used the Internet for political
information acquisition; 18% of these users said the Internet was their primary
source of political information, according to a study by Pew Internet & American
Life (2005).
Coinciding with this expanded use of the Web, network television viewership and
newspaper readership have declined. Indeed, many believe that network television
viewing and newspaper circulation will continue to diminish. Coffey and Stipp
(1997), for instance, suggest that traditional media uses will decline over time
because younger generations will grow up using computers more than their parents’
generation and because the computer now replaces other free-time activities. Televi-
sion news viewing in particular appears to be most vulnerable to this decline. Rogers
(1985) found that these early adopters of computers tend to spend less time viewing
Young Mie Kim (Ph.D., University of Illinois at Urbana–Champaign) is an Assistant Professor in the School of Communication at the Ohio State University. Her research interests include new communication technologies and political communication.
© 2008 Broadcast Education Association Journal of Broadcasting & Electronic Media 52(4), 2008, pp. 600–621 DOI: 10.1080/08838150802437438 ISSN: 0883-8151 print/1550-6878 online
600
Kim/WHERE IS MY ISSUE? 601
television. Accordingly, the Pew survey indicated that about 27% of respondents
replaced television viewing with the Web (Pew Research Center, 1997).
However, some researchers argue that the shift from traditional news to reliance
on the Web is, at best, exaggerated. While the Web has replaced some outlets as
a source of news consumption, it is less dramatic than some would claim. Ahlers
(2006), for instance, illustrated that only 12% of news consumers migrated directly
from traditional news media to electronic news media. Furthermore, another 22%
of U.S. adults reported that they use the Web as a complement to traditional news
media rather than a substitute (Ahlers, 2006; Dutta-Bergman, 2004). Thus, it appears
more likely that the Web functions as a supplement, rather than a substitute, to
traditional news media.
Little is known, however, about how traditional news media influence the pat-
terns of Web use (cf. Dutta-Bergman, 2004). What theories offer an insight into
the relationship between traditional news and Web uses? If the Web serves as a
supplement or a complement to traditional news media, how does this take place?
This study investigates the way news coverage in the traditional media influences
individuals’ subsequent Web use patterns in terms of selectivity. By testing three
theoretical frameworks that explain individuals’ information selection patterns—the
accessibility effect from priming, instrumental utility of information, and personal
issue importance—this study explores how traditional news media influence what
information individuals choose to view on the Web.
What Drives Selectivity on the Web? Three Theories
For nearly 60 years, selectivity has remained an enduring concept in communi-
cation research. The concept of selectivity is especially relevant to the new media
environment, particularly the Web. Whereas traditional news media focus on larger
markets with little interest in tailoring the content to specific markets, information
obtained through the Web is almost inherently specialized by topic (e.g., Green-
berg, 1999; Rash, 1997; Sunstein, 2001).1 An enormous amount of information
is far more accessible on the Web than was previously available through tradi-
tional media. In addition, because the Web promotes a high level of interactiv-
ity, users can be selectively attentive—and selectively exposed—to information.
These two major characteristics of the Web—specialization and interactivity—offer
a high potential for increased selectivity (Tewksbury, 2003). While researchers
accept that the Web offers a higher level of selectivity and therefore offers greater
benefits compared to traditional news media, regrettably little is known about
what factors influence individuals’ information selection on the Web. This is es-
pecially true when examining traditional news and its relationship with Web use
patterns: How does prior exposure to news influence what people choose to view
on the Web? Three theoretical frameworks can be applied to explain the poten-
tial influence of prior exposure to news coverage on information selection on
the Web.
602 Journal of Broadcasting & Electronic Media/December 2008
Accessibility From Priming. One explanation for how news coverage influences
subsequent online information selection would be the accessibility effect from prim-
ing (Higgins, 1996). Tulving (1993) describes this broadly as the facilitative effect
of performing one task on the subsequent performance of the same or a similar
task. Similarly, a media effect of priming broadly refers to the effect media content
has on individuals’ later behavior (Roskos-Ewoldsen, Roskos-Ewoldsen, & Dillman-
Carpentier, 2002).
The priming event activates a knowledge construct, which temporarily increases
the probability of using that construct when performing subsequent tasks. A con-
struct’s temporal accessibility2 is due to the recency and frequency of activation. In
general, a recent construct is likely to be employed in a subsequent task, especially
when the time lag between priming and the subsequent task is short; but as the
time delay increases, the most frequently primed construct takes precedent over a
recently primed construct (Higgins, Bargh, & Lombardi, 1985). Prototypical priming
studies in social psychology thus expose participants to a particular construct and
then have them immediately perform a seemingly unrelated task (e.g., person per-
ception). Participants in Higgins, Rholes, and Jones’s (1977) original priming study,
for instance, were exposed to one or another set of trait-related constructs and
later asked to characterize the ambiguous behaviors of a target person in a reading
comprehension task. Participants were more likely to use the trait-related recent
constructs primed by the initial task to categorize the target person’s behaviors rather
than equally applicable alternative constructs. This accessibility effect of priming
has been replicated in many studies with different methods (for details, see Higgins,
1996).
Some scholars have theorized that the structure or organization of constructs
influences the effects of priming. Anderson (1983) and others (Althaus & Kim,
2006; Collins & Loftus, 1975; Kim, 2005; Price & Tewksbury, 1997), for example,
illustrated that concepts are associated with one another as nodes and nested within
a network structure through linkages. When priming activates a particular construct
in a network, the activation radiates from this particular node to others with which
it is associated, thereby increasing the probability that similar concepts unspecified
in the original stimulus come to mind and influence subsequent tasks. Within this
network structure, the activation is a function of the strength of the associations
between constructs. In support of this, Valentino and colleagues (Valentino, 1999;
Valentino, Traugott, & Hutchings, 2002) found that cues in media (news and politi-
cal advertising, respectively) boosted constraint of opinions regarding related issues.
However, when the issue was less relevant, exposure to media coverage did not
appear to increase constraint of opinions.
Although the effect of prior exposure to news coverage on subsequent online
information selection has not been directly tested, the theory of accessibility effects
of priming suggests that the presence or absence of certain issues in news coverage
should be sufficient to produce accessibility effects and influence immediately
individuals’ subsequent online information selection behavior. The groundbreaking
study by Iyengar and Kinder (1987) addressed the accessibility effect of priming
Kim/WHERE IS MY ISSUE? 603
with a series of experiments that either included or excluded certain issues. Their
findings showed that participants in the experimental condition that included a
certain issue became more concerned with the included issue. Similarly, Domke
and colleagues (Domke, 2001; Domke, Shah, & Wackman, 1998) found that the
presence of some issues in media triggered the issues primed in the coverage and
formed individuals’ perceptions of candidates. Given all this, if the accessibility
effect from priming applies to selectivity on the Web following traditional news
exposure, it is reasonable to expect that online selectivity on a certain issue should
be higher when individuals are exposed to the issue on the news immediately prior
to their information selection on the Web than when they are not exposed.
H1: (Accessibility effect of priming). The level of online selectivity on a particular
issue (i.e., the weight of particular issue content in information selection)
would be higher when individuals are exposed to the issue on traditional
news outlets prior to their Web use than when they have not been exposed
to the issue.
Instrumental Utility Theory. While Festinger’s (1957) theory of cognitive dis-
sonance has long dominated the selective exposure literature with its emphasis
on an avoidance strategy in selectivity, communication scholars have conceptu-
alized selectivity as an approaching strategy rather than an avoidance strategy. In
other words, audience members actively seek information that satisfies their needs,
orientations, and motivations. The uses and gratification approach, for example,
emphasizes the utility function of mass media. When information provided by mass
media is useful and fulfills their needs, orientations, and motivations, audience
members use the mass media. For instance, Blumler and McQuail (1969) explain
how the political content of television is linked to audience members’ various mo-
tivations: voting guidance, reinforcement of existing decisions, general surveillance
of the political environment, as well as anticipated utility in future interpersonal
communication. Similarly, Chaffee and McLeod (1973) illustrated how social utility
(i.e., future involvement in interpersonal communication) motivated individuals to
selectively expose themselves to election campaign information.
As an extension of the uses and gratifications approach but with an emphasis on
the information selection process, selectivity has also been explored through the
function of extrinsic information utility. Atkin’s (1973) extrinsic instrumental utility
theory proposes that selective information seeking can be explained as having a util-
itarian purpose, where an individual uses it directly as a means toward solving prac-
tical problems. The key to this instrumental utility model is the reduction of extrinsic
uncertainty. Intrinsic uncertainty or intrinsic motivations, including individuals’ in-
trinsic interest (e.g., personal importance), intrinsic curiosity, and intrinsic pleasure,
are not considered in Atkin’s instrumental information utility theory because in-
trinsic motivations are noninstrumental. Intrinsic motivations derive consummatory
gratifications in the form of intrinsic satisfaction, but this does not directly apply
604 Journal of Broadcasting & Electronic Media/December 2008
to problem-solving. For this reason, intrinsic motivations are noninstrumental and
beyond the scope of Atkin’s instrumental information utility theory (Atkin, 1973).
Excluding individuals’ intrinsic motivations or predispositions, Atkin argues that
the need for information is ‘‘a function of extrinsic uncertainty produced by a
perceived discrepancy between the individual’s current level of certainty about
important environmental objects and a criterion state he seeks to achieve (p. 206).’’
Extrinsic uncertainty is channeled toward particular objects to solve problems that
arise in one’s everyday environment as an adaptation strategy. Because the range
of one’s knowledge is in general severely limited, the ‘‘discrepancy’’ exists between
the current state of certainty and the criterion state of certainty. When individ-
uals perceive that they have insufficient knowledge about environmental objects
(i.e., primitive uncertainty, ‘‘which product is good?’’) or existing knowledge is
inadequate for situations that require orientations, decisions, and performance (i.e.,
complex uncertainty, ‘‘am I supporting the right candidate?’’), extrinsic uncertainty
increases. New information that an individual does not possess already indeed
diminishes the discrepancy and ultimately reduces extrinsic uncertainty. For this
reason, individuals seek information that they do not know already. Seen this way,
selectivity is an adaptation to extrinsic uncertainty. Within the instrumental utility
theory framework, Knobloch, Dillman-Carpentier, and Zillmann (2003) confirmed
that participants selected the articles associated with the news leads where salient
features induced high levels of extrinsic uncertainty.
Unlike the accessibility effect from priming, the instrumental utility theory im-
plies that if issue information is provided prior to information seeking, the level
of information selection on the issue would be reduced because the utility of the
additional information is low. However, if individuals see additional information
that has not been covered previously, the utility of this additional information is
high. As individuals’ extrinsic uncertainty increases when faced with information
not previously exposed to them, they seek the new information to reduce extrinsic
uncertainty. Valentino, Hutchings, and Williams’ (2004) study of the effect of polit-
ical campaign advertisements on subsequent information-seeking behavior is worth
notice. In this study, participants were exposed to both Gore and Bush campaign
advertisements that contained information on their issue positions. Participants were
then asked to visit either or both of the candidates’ campaign Web sites containing
the issues covered in the advertisement as well as other issues. The study found that
compared to the control group not exposed to the campaign advertisements, those
exposed to political campaign advertisements containing issue information sought
a lesser amount of information in the subsequent information search task. Valentino
et al. (2004) explained that the exposure to advertisements reduced further demand
for issue information.
Taking all this into consideration, this study expects that if the instrumental utility
theory is correct, the level of information selection for a particular issue on the
Web is high when there is no prior exposure to the issue because the information
on the Web not presented in prior news coverage appears to be novel and useful
to decrease individuals’ extrinsic uncertainty.
Kim/WHERE IS MY ISSUE? 605
H2: (Instrumental utility). The level of online selectivity on a particular issue (i.e.,
the weight of particular issue content in information selection) would be
higher when the individuals are not exposed to the issue through traditional
news outlets prior to Web use than when they have been previously exposed.
Personal Issue Importance. When Atkin examined information-seeking behavior,
he did not consider intrinsic motivation because it is noninstrumental. Recently,
however, other scholars have explained selectivity as a function of noninstrumental
considerations such as personal issue importance (Berent & Krosnick, 1993, unpub-
lished, as cited in Boninger, Krosnick, Berent, & Fabrigar, 1995; Petty & Krosnick,
1995). Most people have few resources and little motivation to carefully attend to all
of the nations’ issues to reduce extrinsic uncertainty because the information cost for
becoming fully informed is substantial (Berent & Krosnick, 1995). Therefore, indi-
viduals must be selective in their information gathering, processing, and structuring
(Wang, 1977). Indeed, individuals do not have to be cognitively highly sophisticated
to form attitudes regarding the issues they consider personally important. Therefore,
people select information relevant to issues that are personally important to them. As
motivated tacticians, individuals have multiple strategies for information processing
and choose among them based on goals, needs, and motives (Chaiken, Liberman,
& Eagly, 1989; Fiske & Neuberg, 1990; Petty & Cacioppo, 1986). When individuals
think the issue is of high personal relevance, they desire more information regarding
the issue and engage in a greater level of message scrutiny than when the issue is
perceived to be of little personal importance (e.g., Chaiken, 1980; Petty & Cacciopo,
1986).
Personal issue importance is an intrinsic and chronic tendency rather than an
extrinsic and deliberate strategy influencing selectivity (Berent & Krosnick, 1995).
First, even though individuals see an issue as important to the country as a whole,
(e.g., national defense), they may not consider the issue as personally important
if they do not subjectively value the issue. Second, situational relevance does
not define personal issue importance. For example, individuals may consider the
issue of gay marriage important even if the issue does not directly influence their
lives. Research has shown that individuals tend to distinguish ‘‘I’’ and ‘‘others’’
in perceiving issue importance (Glynn, Ostman, & McDonald, 1995), and as a
result, individuals’ personal experience and collective experience are generally
independent (Kinder & Kiewiet, 1979; Mutz, 1998).
Unlike instrumental utility, the notion of personal issue importance underscores
the fact that individuals are naturally drawn to information personally important
to them regardless of its extrinsic and instrumental value. Consumer research, for
example, has documented that those exhibit a high level of personal interest in
a product want to acquire more information about the product class in general
(McQuarrie & Munson, 1992; Richins, Bloch, & McQuarrie, 1992). Berent and
Krosnick’s 1993 study (unpublished; cited in Boninger et al., 1995) clearly illustrated
this tendency in the area of political communication. In their study, participants
were asked to evaluate political candidates and were given six issue statements
606 Journal of Broadcasting & Electronic Media/December 2008
from each candidate. They were instructed, however, to read only three out of the
six statements for each candidate. Participants selected information relevant to the
issues that were personally important to them at the expense of information relevant
to issues they considered personally unimportant. Furthermore, Berent and Krosnick
provided party affiliation information so that participants could easily infer each
candidate’s position regarding the issues. If selectivity is solely based on extrinsic
utility, selective exposure to issue statement information would be reduced when
the party affiliation information is provided because the utility of seeking further
information is small. The results, however, showed that participants personally
concerned with an issue did select issue information just as much as when party
affiliations were unknown. This implies that information selection is not influenced
by a deliberate calculation of the value of information.
In light of all this, this study expects that if the personal importance account is
correct, individuals would select information on the Web based on their personal
issue importance regardless of prior exposure to news coverage, as opposed to
what the accessibility effect of priming or the instrumental utility theory suggests.
Selectivity on a particular issue thus would be higher when individuals have a high
level of personal issue importance compared to when they do not.
H3: (Personal issue importance). Regardless of prior exposure, the level of online
selectivity on a particular issue (i.e., the weight of particular issue content in
information selection) would be higher when individuals consider the issue
personally important than when they do not consider the issue important.
Interplay of Accessibility, Instrumental Utility, and Personal Issue Importance.
Three theoretical frameworks offer different explanations and predictions as to how
exposure to traditional news media would influence subsequent information se-
lection behavior on the Web. These theories are based on different assumptions
concerning the nature of individuals per se, their motivations and behavior, as well
as information itself. For instance, whereas the accessibility effect of priming theory
suggests presence of issue information in the news should increase information
selection on a particular issue, the instrumental utility of information theory proposes
that prior exposure to issues in the news might reduce information selection given
that the utility of the information becomes low when individuals are exposed to and
learn about the issue already. Rather, new issue information on the Web, which is
not presented in the news prior to Web browsing, has a higher level of instrumental
utility because individuals do not know much about the issue. The concepts of
instrumental utility and personal issue importance emphasize different motivations
behind individuals’ information selection in the sense that information utility is
largely driven by extrinsic environment, while personal issue importance is intrinsic
and chronic by nature. Unlike the accessibility of priming or instrumental utility,
personal issue importance explains selectivity on a particular issue regardless of
prior exposure to the issue information.
Kim/WHERE IS MY ISSUE? 607
However, a careful examination of these theories suggests some interplay be-
tween these factors and provides nuanced theoretical qualifications for the effect
of traditional news coverage on subsequent information selection online. A couple
of possibilities exist in terms of the interaction between the accessibility effect of
priming, instrumental utility of information, and personal issue importance.
For example, when individuals consider an issue personally important but do not
find out about the issue prior to Web browsing, what would happen? Psychological
research on vigilance proposes that individuals monitor their environment for sur-
vival purposes; thus, the implication of negative information increases uncertainty
and tends to grab one’s attention (Pratto & Oliver, 1991). Accordingly, Marcus, Neu-
man, and Mackuen (2000) note that information-seeking behavior (i.e., surveillance)
is most likely to occur when external stimuli threaten individuals’ personal interests.
A threat signals when past learning is not enough for an individual to handle a
new situation and when a stimulus is powerful or dangerous to the individual. The
vigilance effect thus suggests that individuals explore information about an issue
on the Web when they are not exposed to the information in the traditional news
media prior to Web use, but the issue is of personal concern. Therefore, this study
expects an interaction effect of instrumental utility and personal issue importance if
the vigilance theory is correct.
H4: (Vigilance effect). The level of online selectivity on a particular issue (i.e.,
the weight of particular issue content in information selection) should be
amplified when individuals are not exposed to the issue prior to their Web
use and when the issue is personally important to the individuals.
On the other hand, it is also possible that even if there is prior exposure to
issue information, individuals still seek further issue information as long as the
issue centers on one’s personal issue importance. In support of this, Hutchings’
(2003) series of studies demonstrated that the perceived importance of an issue
boosted when the issue was covered in the news and when individuals consider
the issue personally important as well. Thus, in the present study, online information
selection concerning an issue would stay about the same or become amplified when
individuals have a high level of personal issue concern regarding the issue and
when they are exposed to the issue through traditional news outlets prior to using
the Web. This study therefore expects an interaction effect of priming and personal
issue importance.
H5: (Escalation effect). The level of online selectivity on a particular issue (i.e.,
the weight of particular issue content in information selection) would be
amplified when individuals are exposed to the issue prior to their Web use
and when the issue is personally important to the individuals.
In summary, three theoretical frameworks suggest different predications regarding
the influence of news coverage on subsequent information selection on the Web.
608 Journal of Broadcasting & Electronic Media/December 2008
Regrettably, little research has yet empirically tested the impact of traditional news
coverage on the way individuals seek information on the Web. The present study
attempts to test different theories and provide a theoretical framework to explain the
process by which the Web supplements (or not) traditional news media by exam-
ining how exposure to news coverage on traditional media influences individuals’
subsequent information selection on the Web.
Method
Overview
The study examines how exposure to traditional news would affect individuals’
subsequent information selection behavior on the Web. In particular, this study tests
competing selectivity hypotheses that would offer explanations for the relationship
between traditional news and online information consumption. This study consisted
of four phases: A pretest, TV news viewing (no news condition skips this stage), Web
surfing, and a posttest.
Design
The study employed 3 (issue coverage condition: no news vs. issue presence vs.
issue absence) � 2 (personal importance: high vs. low) factorial designs to test the
effects of the presence or absence of an issue in the news and the individuals’
personal issue importance, on their subsequent information selection behavior on
the Web.
Sample
A total of 306 subjects (18 years or older) completed all stages of the study.
Subjects were recruited from lecture classes from a large midwestern university.
Extra credit was offered to those who completed the study.
Procedure
Subjects were asked to come to a computer lab and complete a short com-
puterized pretest. The pretest included questions on personal issue importance
concerning four focused issues (abortion, gay rights, the economy, and the war
in Iraq), general political interest questions, and distractor questions (e.g., media
uses patterns).
After completing the pretest, subjects were asked to watch a video clip that
contained a collection of news stories about the four different issues. Under the
Kim/WHERE IS MY ISSUE? 609
‘‘absence’’ condition, one of the issues was excluded so that only three issues were
covered. Under the ‘‘presence’’ condition, all four issues were covered. The TV
viewing stage ran about 20 minutes including introductions. The control condition
(no video condition) skipped this stage.
Next, subjects were asked to find and read information on the four focused
issues on the Web. A portal site designed by the research team was introduced to
subjects for ease of information searching. The portal included 60 actual Web sites
from diverse sources (including news, issue advocacy groups, and research think
tanks) with information on the four focused issues. The portal site also included
instructions for the information search. Subjects were instructed to browse and
examine whatever information they desired for up to 20 minutes. The instructions
stated that subjects would not be able to read everything linked to this site to
induce selectivity. The 20-minute time limit was also given to induce selectivity in
the information search. Each individual’s information selection was recorded in real
time by an individual level, click-by-click, Web behavior recording program.
After the online information search task, subjects were asked to fill out another
short computerized posttest. The posttest included recall questions as well as basic
demographic questions (e.g., gender, age, race, party identification).
TV News Viewing Manipulation
Evening news stories on abortion, gay rights, the economy (tax and budget), and
the war in Iraq from a network television station (ABC) were selected about 5
months prior to the study. Each news story had a package format that included
an anchor’s leading comment, episodes, interviews, and reporters’ comments. The
average length of a news story was 126 seconds (abortion, 123 seconds; gay rights,
115 seconds; the economy, 135 seconds; war, 130 seconds).3 For the presence
condition, a video clip contained four news stories covering each of the focused
issues. The presence condition was broken into four subgroups depending on the
order of the four news stories. Therefore, four different tapes were made for the
presence condition, but because the tapes were randomly assigned to the subjects
under the presence condition, the subgroups were pooled in the analysis. For the
absence condition, a total of four tapes were created; each tape was missing one
of the focus issue stories. In the analysis, each issue absence condition (n D 51
for each issue) was compared to the presence (n D 51) and the no news exposure
conditions (n D 51).
Web Behavior Record
Given the problems of self-reporting exposure measures (see Chaffee & Schleuder,
1986; Price & Zaller, 1993; Tewksbury, 2006; Valentino et al., 2004), this study
developed an innovative method for measuring information search behavior by
adopting a real-time, click-by-click tracking method. For each page that participants
610 Journal of Broadcasting & Electronic Media/December 2008
accessed, the computer program recorded the individuals’ unique identification
number, time stamp (including time spent on the link), the Web page file name,
and the actual content of each Web page. Each individual’s Web viewing data were
automatically transmitted to a university server and saved as a data file with a unique
user ID. Due to the transitory nature of the Web, Web pages may have changed
or disappeared prior to the analysis. To overcome this problem, the program also
downloaded each of the Web pages viewed by the individuals in real time and
transmitted these to the research server.4 The program is based on a client-side pro-
gramming technique, which allows it to capture existing real-world Web behavior
beyond artificially created research Web sites, so participants can visit any Web site
existing in the real world. Using unique user IDs, the program enables researchers to
analyze data at an individual level (as opposed to IP addresses) and combines these
with survey data, which is not possible with log analyses or server-side tracking
programs.
Study Venue
A portal site was constructed as the venue for the study. At the top level, five
categories of Web pages were presented in a menu format: main menu, abortion, gay
rights, the economy, and the war in Iraq. The design and structure of the portal was
the same as any regular Web site in order to provide subjects with easy surfing ability
and to increase the validity of the study. The menu bar in the window was designed
to stay during the study for ease of information selection. In this way, subjects did
not need to go back to the menu page to move around different issue categories.
The list of relevant political Web sites was presented in alphabetical order. A total
of 60 real Web sites were linked to the portal under these categories. Selection of
these sites was based on nonpartisan, voting information consolidator sites, such as
Project Vote Smart (national and local election information) and Politics1 (national
and local election information).
Measures
Personal Issue Importance. Personal issue importance was measured in the pretest
by a 7-point scale, ranging from 1 (not at all important) to 7 (extremely important)
for each of the four focus issues (Mabortion D 5.0, SDabortion D 1.87; Mgay D 3.82,
SDgay D 1.98; Meconomy D 6.11, SDeconomy D 1.09; Mwar D 5.91, SDwar D 1.25.
Selectivity. Selectivity was operationalized in terms of centrality at the issue level,
which was conceptualized by the weight of a particular kind of issue content in
information selection. It was measured by the number of pages on a category out of
the total number of pages (Mabortion D .36, SDabortion D .33; Mgay D .16, SDgay D
.24; Meconomy D .22, SDeconomy D .26; Mwar D .26, SDwar D .28). 5 This measure of
selectivity was based on individual Web behavior records. Individuals’ Web viewing
Kim/WHERE IS MY ISSUE? 611
records were coded by looking into each of the Web pages an individual viewed
(therefore, the coding unit is a page). After the content categorizing into four focus
issues and others, the hit number (the number of clicks) and the time spent on a page
for each category were calculated.6 For the reliability test, 20% of the cases were
randomly selected and coded separately by two trained coders. The two coders’
agreement rate was about 98.5%, yielding Cohen’s kappa .96.
Other Variables. Basic demographic variables including gender (female D 66%,
male D 34%); age (18–24 D 40%, 25–29 D 54%, 30 or older D 6%); race (White D
80.4%); party identification (Democrat D 46.4%, Republican D 35.1%, Indepen-
dent D 14.2%); and political ideology (M D 3.68, SD D 1.61) were measured and
used as control variables.
Results
Manipulation Check
To see if TV viewing manipulation was successful, subjects were asked what
issues they viewed on the news during the study. Two individuals who failed to
correctly identify issue coverage were excluded in the final analysis.
The study also examined differences between conditions in terms of basic de-
mographics (i.e., gender, age, race, party identification, and political ideology) to
check whether the random assignment was successful. No statistically significant
difference was found between the conditions. Therefore, experimental manipulation
and random assignment appeared to be successful.
Testing the Accessibility From Priming Hypothesis
First, this study tested whether subsequent online information selection would be
affected by the presence of issues in news coverage. If the accessibility effect from
priming occurs, when all four issues are covered (presence condition), subjects’
online information selection should be higher compared to absence and no news
conditions. Across all four focus issues, however, this study did not find any statisti-
cally significant differences between the three conditions—presence, absence, and
no news conditions—suggesting no main effect of the presence of issues in news
coverage on subsequent online information selection, abortion: F (2, 128) D .273,
p D .762, observed power D .16; gay rights: F (2, 122) D 1.621, p D .20, observed
power D .46; the economy: F (2, 123) D 1.51, p D .23, observed power D .44;
the war in Iraq: F (2, 123) D .549, p D .58, observed power D .23. Therefore, no
accessibility effect from priming on subsequent online information selection was
found, rejecting H1.
612 Journal of Broadcasting & Electronic Media/December 2008
Testing the Instrumental Utility of Information Hypothesis
It appeared that there was no difference between ‘‘all four issues covered’’ (pres-
ence condition), ‘‘one issue missing’’ (absence condition), and ‘‘no news.’’ The data
were then pooled and divided into two groups per issue: no prior issue coverage
(issue absence and control condition) and prior issue coverage (issue presence
conditions).
The study then tested H2, whether prior exposure to issues or no prior exposure
to issues influenced individuals’ subsequent online information selection patterns.
In particular, the study examined whether individuals selected more information
when they were not previously exposed to the issue than when they were, driven
by the instrumental utility of news information. Supporting H2 in general, the results
suggest the existence of instrumental utility effects. Individuals tended to find more
information about an issue on the Web when they were not previously exposed to
the issue, except for the abortion issue, F (1, 250) D .042, p D .41, Power D .11.
This tendency was found to be consistent across the other three issues, gay rights,
F (1, 250) D 4.45, p D .015, Partial �2 D .04 (no prior issue coverage M D .18,
SD D .26; prior issue coverage M D .11, SD D .21), the economy F (1, 250) D 1.9,
p D .09, Partial �2 D .02 (no prior issue coverage M D .27, SD D .31; prior issue
coverage M D .20, SD D .23), and war F (1, 250) D 2.65, p D .05, Partial �2 D
.03 (no prior issue coverage M D .22, SD D .29; prior issue coverage M D .13,
SD D .26).
The findings again suggest (at the very least) the accessibility effect of priming
by prior exposure to issues did not explain individuals’ issue information selection
patterns on the Web. Rather, the instrumental utility of new issue information, which
was excluded in news but included in the Web, appeared to better explain the
subsequent issue information selectivity online, supporting H2.
Testing the Personal Issue Importance Hypothesis
Based on the notion of personal issue importance, the present study expected
that personal issue importance as individuals’ predisposition would influence in-
dividuals’ online information selection patterns regardless of prior exposure to the
issue (H3). As expected, personal issue importance made a significant difference
between conditions across all four focus issues, the abortion issue, F (1, 250) D
38.55, p D .000, Partial �2 D .14 (high personal importance M D .44, SD D .34;
low personal importance M D .19, SD D .23), gay rights, F (1, 250) D 14.69, p D
.000, Partial �2 D .06 (high personal importance M D .21, SD D .27; low personal
importance M D .10, SD D .18), the economy, F (1, 249) D 2.58, p D .05, Partial
� 2
D .01 (high personal importance M D .25, SD D .26; low personal importance
M D .19, SD D .26), and war, F (1, 249) D 17.19, p D .000, Partial �2 D .07
(high personal importance M D .30, SD D .29; low personal importance M D .14,
SD D .20).7 When individuals showed higher levels of personal issue importance,
Kim/WHERE IS MY ISSUE? 613
the selectivity on the issue of personal importance increased to a greater extent. The
findings suggest strong effects of personal issue importance on online information
selection behavior, confirming H3.
Testing Interaction Effects
Next, the study tested the interaction effects of TV news coverage and personal
issue importance on subsequent information selection patterns on the Web. In par-
ticular, the study explored the possibility of vigilance effects (H4) and/or escalation
effect (H5). When individuals care about an issue but do not find out about the
issue prior to Web browsing, their subsequent online information selectivity on that
particular issue would be amplified (vigilance effect). On the other hand, it may also
be possible that even if there is prior information regarding an issue, an individual
would still seek information further as long as the issue centers on one’s interest.
In this case, information selection on the Web would stay about the same or be
amplified when individuals have a high level of personal issue concern and when
they are exposed to news coverage on the issue (escalation effect).
As shown in Figure 1, the gay rights issue yielded a vigilance effect, supporting
H4. Those who were concerned with the issue of gay rights exhibited higher levels
of selectivity when the issue was not found prior to their Web browsing, F (3, 81) D
Figure 1
Vigilance Effect: Interaction of News Coverage and Personal Issue Importance on
Subsequent Information Selection on the Web (Gay Rights)
614 Journal of Broadcasting & Electronic Media/December 2008
Figure 2
Escalation Effect: Interaction of News Coverage and Personal Issue Importance
on Subsequent Information Selection on the Web (Abortion)
2.58, p D .05, Partial �2 D .10 (Figure 1). However, none of the other interac-
tion effects of personal issue importance and instrumental utility was statistically
significant.
Interestingly, in the case of the abortion issue, an escalation effect was found,
F (3, 87) D 4.45, p D .006, Partial �2 D .15 (Figure 2). When individuals had a high
level of personal issue importance on the abortion issue and they were exposed to
the issue prior to Web browsing, their selectivity on abortion was significantly high,
confirming H5.
Discussion
Using a Web behavior recording program that measures naturally occurring Web
viewing, click-by-click, in real time, the study examined how traditional news cov-
erage (in this case TV news) influences individuals’ subsequent information selection
on the Web. Although many media pundits have claimed that traditional news me-
dia have been threatened by the growth of the Web, research has consistently shown
that the Web functions as a supplement to traditional news. However, relatively
little is known about how people use the Web to supplement traditional news. By
employing a unique method that measures online information selection, click-by-
Kim/WHERE IS MY ISSUE? 615
click at an individual level, the study enables one to better understand the process
by which traditional news coverage influences subsequent information selection on
the Web. As Valentino and colleagues (2004) noted, the lack of sound theoretical
explanations and inconsistent empirical findings concerning how media messages
influence subsequent selective exposure may be partly due to the imprecise methods
used to measure information exposure; the problem becomes more obvious in the
case of online information exposure (Tewksbury, 2006). The individual level Web
behavior recording method employed in this study, therefore, should allow one to
better understand individuals’ information selection patterns.
This study first empirically tested three competing theoretical accounts for selec-
tive information searches on the Web in relation to prior exposure to traditional
media in an attempt to offer a better theoretical framework to explain the relation-
ship between traditional news and Web use patterns: the accessibility effect from
priming, instrumental utility of information, and personal issue importance. To some
extent, these theories predict seemingly conflicting outcomes, or at least provide
competing explanations for online selectivity following exposure to traditional news
coverage. Prior research has viewed the relationship between traditional news media
and the Web from the displacement (e.g., Coffey & Stipp, 1997; Stephens, 1998) or
complementary perspective (e.g., Ahlers, 2006; Dutta-Bergman, 2004), but has not
specified the processes by which the displacement or complementary relationship is
formed. Dutta-Bergman’s noteworthy study, for instance, reveals significant overlap
between traditional and online news consumption at the level of domain-specific
topic, but not by medium, clearly representing evidence for the complementary
relationship within topic areas. Although it goes one step further by illustrating
the details of the complementary relationship using a large scale of survey data
(Pew Research data), Dutta-Bergman’s study still does not allow one to identify
the processes underlying the relationship between traditional news and Web use
because of limitations in the survey method.
The findings of this study have several important implications regarding selectivity
on the Web and its relationship with traditional news use. Most importantly, the
findings of this study exhibit the strong effects of personal issue importance on
individuals’ information selection on the Web across all four focus issues. Including
the control group (no news condition), the impact of personal issue importance
was sustained. The main effects of personal issue importance also survived when
news coverage or the prior issue exposure factor was introduced. This suggests that
individuals’ information selection online is perhaps primarily driven by personal
issue importance as an intrinsic motivation. At a more macro level, the results may
imply that the Web may greatly facilitate sharper distinctions or a highly segmented
specialized audience as found in Tewksbury’s (2003) research on online news
consumption. Ultimately, this trend may lead to the development of distinctive
groups along issue lines, namely ‘‘issue publics’’ (Converse, 1964). With increased
specialized information and interactivity, issue publics can more easily focus on
the topics they are concerned with in their information selection. By using the
additional information that is available through more specialized and interactive
616 Journal of Broadcasting & Electronic Media/December 2008
media, individuals would advance their knowledge and consolidate their attitude.
However, this narrow range of information exposure resulting from the new media
consumption has been an enduring concern among communication scholars (e.g.,
Davis, 1999; Katz, 1996). With the rapid diffusion of specialized and interactive new
communication technologies, the intensified issue publics and their narrow range
of information selection may weaken social consensus and polarize the public to
more extreme positions on the issues of their concerns (Havick, 2000; Sunstein,
2002).
Still, the strong impact of personal issue importance on online selectivity should
not completely discount the influence of traditional news on Web use patterns.
Consistent with prior research (Atkin, 1973; Knobloch et al., 2003), the results of the
study generally supported the notion of the instrumental utility of new information
on the Web in relation to issue coverage on traditional news. While prior research
on the instrumental utility of information has centered on how features of messages
(e.g., salience features; Knobloch et al.) cause individuals to view the utility of the
messages differently, this study expands the theory to how prior exposure (i.e., ab-
sence vs. presence of information) influences later judgments about the instrumental
utility of related issue information, especially when involved with the consumption
of two different media: traditional and new media. Individuals tended to select more
information about a particular issue online when they were not exposed to the issue
prior to their Web browsing. Individuals actively seek online information ‘‘missing’’
from traditional news media in an attempt to reduce extrinsic uncertainty. When
issues are already covered by traditional news media, its information utility becomes
relatively low. Because of this, individuals would rather find new information when
they have a chance to browse the Web. More importantly, when personal concern
on an issue is relatively high and the issue is not covered in the news in the prior
exposure setting, their level of subsequent information selection becomes amplified
(vigilance effect), as shown with the gay rights issue in this study. In some cases,
though, individuals’ level of selectivity stay about the same or become amplified
in the prior presence of the issue in the news setting, as shown in the abortion
issue in this study. At any rate, prior exposure to traditional news still matters in
information selection on the Web. It is not so clear, however, why the study found
mixed patterns of interaction effects. Given that the results varied by issue, one can
speculate that the nature of the issue might have played a role. Future research
should replicate this study using a wider range of issues and consider this factor in
developing a theoretical framework.
At the very least, the findings of this study imply that the simple accessibility effect
of priming may need some qualifications when explaining the influences of prior
exposure to traditional news on information selection on the Web. Individuals do
not merely follow what traditional news media suggest they should pursue on the
Web. It appears that information selection on the Web is primarily determined by
individuals’ habitual and intrinsic inclination (i.e., personal issue importance) and
motivational calculation about what would be the most useful in consideration of
extrinsic utility (i.e., instrumental utility). As Higgins (1996) noted, individuals at
Kim/WHERE IS MY ISSUE? 617
times do not apply primed constructs to subsequent judgments. Chronically acces-
sible constructs (Higgins & Brendl, 1995) as well as motivational factors (Martin,
1986; Martin & Achee, 1992; Martin, Seta, & Crelia, 1990) also play substantive
roles in priming effects. In this sense, the effects of personal issue importance and
instrumental utility and the interplay of the two found in this study may all need to
be incorporated to expand the theoretical framework of priming as well.
The findings of this study may have noteworthy implications for the news media
industry in particular. Most of all, the news media industry (including online news)
should more carefully consider that their audience may have various reasons for
seeking information, which may alter the scope or type of information provided.
Especially, because online news outlets find it difficult to set agendas (Althaus &
Tewksbury, 2002; Tewksbury, 2003), the need to consider individuals’ intrinsic and
extrinsic motivations would be imperative. Faced with the significant threat of the
Web, the traditional news media has already actively adopted the Web as an addi-
tional news outlet. However, some studies (e.g., Tewksbury & Althaus, 2000) have
shown that the content of traditional and online news almost mirror each other. The
results of this study suggest that the news media industry should consider significant
differences in information consumption between traditional and online news outlets.
This study reinforces the idea that the two media outlets should be supplementary
but different in terms of the scope and type of information that each provides.
The present study, of course, has some limitations. Although the study assumes
that the processes of selectivity and media use patterns are common psychological
processes shared by all human beings, the use of a convenient sample of college
students might still have limitations in the generalizability of the results of the study.
The power turned out to be relatively low for some of the statistically insignificant
findings, suggesting some caution in interpreting those results. To increase the ex-
ternal validity, the study employed real Web sites as a venue for online information
selection, but the findings of the study might have been influenced by uncontrolled
real Web sites (although the between-subject design should have randomized the
potential error). Perhaps the biggest limitation is the limited number of issues tested
in this particular study. Since the study found mixed results in the interaction effects
by issue, it is imperative to explore the influence of the nature of issues on selectivity
on the Web.
Despite all, the study shed light upon one’s understanding of how individuals
use traditional news and the Web and how news coverage influences subsequent
information selection on the Web in terms of selectivity. The findings of the study
should better inform one of the ways to untangle the relationship between traditional
news and Web uses in the rapidly changing media environment.
Notes
1One might argue that high levels of specialization already have been observed in other news media such as cable television. However, specialization of online information is still
618 Journal of Broadcasting & Electronic Media/December 2008
distinguishable from cable television. The ownership structure of cable television constrains the ability to provide greater specialization and diversity; for example, increased competition and lower profitability in cable television, where information providers maintain an oligopoly system (Gomery, 1993), have reduced message diversity, and programming largely depends on market size and other market factors.
2In some cases, constructs are chronically, as opposed to temporally, accessible. Chronic accessibility may come from different sources (Price & Tewksbury, 1997): Frequent activation and self-concept (i.e., predispositions).
3According to the Annual Report on American Journalism (2004), edited packages in the evening news of three networks were 84% of the time on these programs, 6.8 packages on each evening. An average length of an edited package was 138 seconds.
4Visual image files were not captured to avoid delayed information in loading individuals’ Web activities.
5Centrality was measured by the time spent on a content category out of total time spent as well. The patterns were similar to hit measures (Mabortion D .32, SDabortion D 1.37; Mgay D .12, SDgay D .19; Meconomy D .16, SDeconomy D .22; Mwar D .18, SDwar D .22. The results using hit measures were reported in this paper.
6If the duration of reading a page exceeded 10 minutes, this was regarded as an interruption of the study and the duration was calculated as just 10 minutes.
7Instead of dichotomizing personal issue importance scores, the present study regressed individuals’ selectivity scores on their 7-point personal issue importance scales, TV viewing conditions, the interaction of the two, and other control variables including gender, race, age, political ideology, and party identification. The patterns of these results are about the same. The effects of personal issue importance were statistically significant and survived across all four focus issues.
References
Ahlers, D. (2006). News consumption and the news electronic media. Harvard International Journal of Press/Politics, 11(1), 29–52.
Althaus, S. K., & Kim, Y. M. (2006). Priming effects in complex information environments: Reassessing the impact of news discourse on presidential approval. Journal of Politics, 68(4), 960–976.
Althaus, S. L., & Tewksbury, D. (2002). Agenda-setting and the ‘‘new’’ news: Patterns of issue importance among readers of the paper and online versions of the New York Times. Communication Research, 29, 180–207.
Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press.
Annual Report on American Journalism (2004). The state of news media 2004. Retrieved September 15, 2007, from http://www.stateofthenewsmedia.org/narrative_networktv_ contentanalysis.asp?catD2&mediaD4
Atkin, C. K. (1973). Instrumental utilities and information seeking. In P. Clarke (Ed.), New models for communication research (pp. 205–242). Beverly Hills, CA: Sage.
Berent, M. K., & Krosnick, J. A. (1995). The relation between political attitudes importance and knowledge structure. In M. Lodge & K. McGraw (Eds.), Political judgment: Structure and process (pp. 91–110). Ann Arbor: University of Michigan Press.
Blumler, J. G., & McQuail, D. (1969). Television in politics. Chicago: University of Chicago Press.
Boninger, D. S., Krosnick, J. A., Berent, M. K., & Fabrigar, L. R. (1995). The causes and consequences of attitude importance. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 159–190). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Kim/WHERE IS MY ISSUE? 619
Chaffee, S. H., & McLeod, J. (1973). Individual vs. social predictors of information seeking. Journalism Quarterly, 50, 237–245.
Chaffee, S. H., & Schleuder, J. (1986). Measurement and effects of attention to media news. Human Communication Research, 13, 76–107.
Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39, 752– 766.
Chaiken, S., Liberman, A., & Eagly, A. H. (1989). Heuristic and systematic information pro- cessing within and beyond the persuasion context. In J. S. Uleman & J. A. Bargh (Eds.), Unintended thought (pp. 212–252). New York: Guilford Press.
Coffey, S., & Stipp, H. (1997). The interactions between computer and television usage. Journal of Advertising Research, 37, 61–67.
Collins A. M., & Loftus, E. F. (1975). A spreading-activation theory of semantic processing. Psychological Review 82, 407–428.
Converse, P. E. (1964). The nature of belief systems in mass publics. In D. A. Apter (Ed.), Ideology and discontent (pp. 206–261). New York: Free Press.
Davis, R. (1999). The web of politics. New York: Oxford. Domke, D. (2001). Racial cues and political ideology: An examination of associative priming.
Communication Research, 28, 772–801. Domke, D., Shah, D. V., & Wackman, D. B. (1998). Media priming effects: Accessibility,
association, and activation. International Journal of Public Opinion Research, 10, 51–74. Dutta-Bergman, M. J. (2004). Complementarity in consumption of news types across traditional
new media. Journal of Broadcasting & Electronic Media, 48, 41–60. Festinger, L. (1957). A theory of cognitive dissonance. Palo Alto, CA: Stanford University Press. Fiske, S. T., & Neuberg, S. L. (1990). A continuum of impression formation, from category-
based to individuating process: Influences of information and motivation on attention and interpretation. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 23, pp. 1–74). New York: Academic.
Glynn, C. J., Ostman, R. E., & McDonald, D. G. (1995). Opinions, perception, and social reality. In T. L. Glasser & C. T. Salmon (Eds.), Public opinion and the communication of consent (pp. 249–277). New York: Guilford Press.
Gomery, D. (1993). Who owns the media? In A. Alexander, J. Owers & R. Carveth (Eds.), Media economics: Theory and practice (pp. 47–70). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Greenberg, A. (1999). Reply to Pippa Norris’s ‘‘Who Surfs?’’ In E. C. Kamarck & J. S. Nye, Jr. (Eds.), Democracy.com? Governance in a networked world (pp. 95–98). Hollis, NH: Hollis Publishing.
Havick, J. (2000). The impact of the Internet on a television-based society. Technology in Society, 22, 273–287.
Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability, and salience. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 133–168). New York: Guilford.
Higgins, E. T., Bargh, J. A., & Lombardi, W. (1985). Nature of priming effects on categorization. Journal of Experimental Psychology, 11, 59–69.
Higgins, E. T., & Brendl, C. M. (1995). Accessibility and applicability: Some ‘‘activation rules’’ in influencing judgment. Journal of Experimental Social Psychology, 31, 218–243.
Higgins, E. T., Rholes, W. S., & Jones, C. R. (1977). Category accessibility and impression formation. Journal of Experimental Social Psychology, 13, 141–154.
Hutchings. V. L. (2003). Public opinion and democratic accountability: How citizens learn about politcs. Princeton, NJ: Princeton University Press.
Iyengar, S., & Kinder, D. R. (1987). News that matters: Television and American opinion. Chicago: University of Chicago Press.
Katz, E. (1996). And deliver us from segmentation. Annals of American Academy of Political and Social Science, 546, 22–33.
620 Journal of Broadcasting & Electronic Media/December 2008
Kim, Y. M. (2005). Use and disuse of contextual primes in dynamic news environments, Journal of Communication, 55, 737–755.
Kinder, D. R., & Kiewiet, D. R. (1979). Economic discontent and political behavior: The role of personal grievances and collective economic judgments in Congressional voting. American Journal of Political Science, 23, 495–527.
Knobloch, S., Dillman-Carpentier, F., & Zillmann, D. (2003). Effects of salience dimensions of informational utility on selective exposure to online news. Journalism & Mass Communi- cation Quarterly, 80, 91–108.
Marcus, G. E., Neuman, W. R., & Mackuen, M. (2000). Affective intelligence and political judgment. Chicago: The University of Chicago Press.
Martin, L. L. (1986). Set/reset: Use and disuse of concepts in impression formation. Journal of Personality and Social Psychology, 51, 493–504.
Martin, L. L., & Achee, J. W. (1992). Beyond accessibility: The role of processing objectives in judgment. In L. L. Martin & A. Tesser (Eds.), The construction of social judgments (pp. 195– 216). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Martin, L. L., Seta, J. J., & Crelia, R.A. (1990). Assimilation and contrast as a function of people’s willingness and ability to expend effort in forming an impression. Journal of Personality and Social Psychology, 59, 27–37.
McQuarrie, E. F., & Munson, J. M. (1992). A revised product involvement inventory: Improved usability and validity. Advances in Consumer Research, 19, 108–115.
Mutz, D. C. (1998). Impersonal influence: How perspectives of mass collectives affect political attitudes. New York: Cambridge University Press.
Nielsen/NetRatings (2007). Average Web usage. Retrieved March 15, 2007, from http://pm.net ratings.com/nnpm/owa/NRpublicreports.usagemonthly
Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and peripheral route to attitude change. New York: Springer-Verlag.
Petty, R. E., & Krosnick, J. A. (1995). Attitude strength: Antecedents and consequences. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Pew Internet and American Life (2005, October 17). The Internet and democratic debate. Retrieved May 15, 2007, from http://www.pewinternet.org
Pew Research Center for the People and the Press (1997, October 17). TV news viewership declines. Retrieved March 15, 2007, from http://www.people-press.org/mediaque.htm
Pratto, F., & Oliver, P. J. (1991). Automatic vigilance: The attention-grabbing power of negative social information. Journal of Personality & Social Psychology, 61, 380–391.
Price, V., & Tewksbury, D. (1997). News values and public opinion: A theoretical account of media priming and framing. In G. Barnett & F. Boster (Eds.), Progress in communication science (Vol. 13, pp. 173–212). London: Ablex Publishing.
Price, V., & Zaller, J. (1993). Who gets the news? Alternative measures of news reception and their implications for research. Public Opinion Quarterly, 57, 133–164.
Rash, W., Jr. (1997). Politics on the Nets: Wiring the political process. New York: W. H. Freeman.
Richins, M. L., Bloch, P. H., & McQuarrie, E. F. (1992). How enduring and situational involvement combine to create involvement responses. Journal of Consumer Psychology, 1, 143–153.
Rogers, E. M. (1985). The diffusion of home computers among households in Silicon Valley. Marriage and Family Review, 8, 89–100.
Roskos-Ewoldsen, D. R., Roskos-Ewoldsen, B., & Dillman-Carpentier, F. R. (2002). Media priming: A synthesis. In J. Bryant & D. Zillmann (Eds.), Media Effects: Advances in theory and research (2nd ed., pp. 97–120). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Stephens, M. (1998). Which communications revolution is it, anyway? Journalism & Mass Communication Quarterly, 75, 9–13.
Sunstein, C. R. (2001). Republic.com. Princeton, NJ: Princeton University Press. Sunstein, C. R. (2002). The law of group polarization. Journal of Political Philosophy, 10, 175–
195.
Kim/WHERE IS MY ISSUE? 621
Tewksbury, D. (2003). What do Americans really want to know? Tracking the behavior of news readers on the Internet. Journal of Communication, 53, 694–710.
Tewksbury, D. (2006). Exposure to the newer media in a presidential primary campaign. Political Communication, 23, 313–332.
Tewksbury, D., & Althaus, S. L. (2000). Differences in knowledge acquisition among readers of the paper and online versions of a national newspaper. Journalism & Mass Communication Quarterly, 77, 457–479.
Tulving, E. (1993). Elements of episodic memory. New York: Oxford University Press. Valentino, N. A. (1999). Crime news and the priming of racial attitudes during evaluations of
the president. Public Opinion Quarterly, 63, 293–320. Valentino, N. A., Hutchings, V. L., & Williams, D. (2004). The impact of political adver-
tising on knowledge, Internet information seeking, and candidate preference. Journal of Communication, 54, 337–354.
Valentino, N. V., Traugott, M. W., & Hutchings, V. L. (2002). Group cues and ideological constraint: A replication of political advertising effects studies in the lab and the field. Political Communication, 19, 29–48.
Wang, G. (1977). Information utility as a predictor of newspaper readership. Journalism Quar- terly, 54, 791–794.
