Social reality effects of the mass media:
The case of the Aum Shinrikyo affair in Japan
Department of Communication
Tokyo Woman's Christian University
Suginami-ku, Tokyo, 167, Japan
[log in to unmask],
[log in to unmask]
Fukushima College for Women
1-1, Chigoike Miyashiro
Fukushima city, 960-01, Japan
[log in to unmask]
social reality effects of the media
Social reality effects of the mass media:
The case of the Aum Shinrikyo affair in Japan
A series of crimes by the Aum Shinrikyo cult, including the sarin gas attack on
the Tokyo subway system on March 20 in 1995, shocked the entire Japanese
society. The Aum Shinrikyo affair and its related issues got massive media
coverage. But it had many problems on various aspects. The whole mass media in
Japan spared no trouble to inform terrified public about the suspects of the
series of affair, the Aum Shinrikyo members and their leader, Shoko Asahara,
even before the police had been done enough investigations. During that period,
most of the media reports relied on the limited sources like leaking by the
police, and they treated Shoko Asahara and the cult members as if they had been
already convicted as guilty.
In addition, the amount of the media reports was so excessive that the
situation was unpararelled in the recent Japanese media history. From late
March to August in 1995 (even after the day Shoko Asahara was arrested on May
16), almost every mass medium focused its concentration overly on the Aum
issues. The only recent example in the U.S. comparable to the Aum coverage in
Japan was probably that of O.J. Simpson's case, although the Aum case was much
more negative toward the culprits.
Arai (1979) introduced the concept of Sou Journalism Jokyo e
Journalism situation), which refers to a situation that the entire mass media in
a country focuses on one particular issue, and reports the issue as much as
possible from the same biased point of view. Under such a circumstance, the
media may give audiences only one viewpoint about the issue, and the audiences
are not informed enough about other important news in the society. The
over-concentrated and uniform media coverage on the Aum Shinrikyo affair was the
very example of Sou Journalism Jokyo.
Many scholars and former journalists severely criticized this unusual
situation, but not much research has been done about the impact of those media
coverage on the audiences. The main purpose of this study is to examine the
impact of the media coverage of the Aum issues upon perceptions of the Japanese
audience. In doing so, the study is based on the perspective of cultivation
theory (Gerbner & Gross, 1976; Gerbner, Gross, Morgan, & Signorielli, 1994),
which is one of the major media effect theories focusing on the contribution of
the media upon audiences' perceptions of social reality.
For more than two decades, research on the influence of the mass media on our
perceptions of social reality has been guided by several well-known theories,
including agenda-setting (McCombs & Show, 1972; Dearing & Rogers, 1996),
cultivation (Gerbner & Gross, 1976; Gerbner, Gross, Morgan, & Signorielli,
1994), spiral of silence (Noelle-Neumann, 1974), and third-person effects
(Davison, 1983). Although there are to some extent common aspects among them,
those theories and models focus on somewhat different aspects of social reality
Cultivation theory attempts to shed light on the unintended influence of TV
upon viewers. Although many studies concerning cultivation theory deal with
perceptions about violence and crime, the theory also has been applied to a wide
variety of topics including sex role stereotypes (Morgan, 1982), political
orientations (Gerbner, Gross, Morgan, & Signorielli, 1982, 1984), or beliefs
about racial integration (Matabane, 1988). The theory postulates that the more
time people spend watching television, the more likely it is that their
conceptions of social reality will reflect what they see on television.
While some studies basically follow Gerbner's assumptions and analytical
procedures (e.g., Matabane, 1988), a number of researchers have questioned or
challenged Gerbner et al.'s assumptions, methodologies, and findings. Many of
the cultivation studies focus on the revision of the original theory as
formulated by Gerbner and his associates. Although an extensive review of such
criticisms is beyond the scope of this article (For more extensive review of the
criticisms, see Potter, 1993), we attempt to extend the scope of cultivation.
This is another objective of this study.
The important point is whether the theory can apply only to the long-term
influence of television. In the over-20 year-tradition of cultivation research,
there have been may arguments about how the theory is defined. Some researchers
used experimental methods to test cultivation hypothesis (e.g., Ogles & Hoffner,
1987), in which the effect they found was inevitably the short-term one. Others
attempted to apply the theory to mass media other than television. Although
Gerbner et al. have emphasized that the theory focuses on "the long-term and
cumulative" impact of television exposure, it is not made clear in the original
formulation that how long is enough to call an effect the long-term one. Here
we propose the idea that the cultivation perspective may be extended to other
media and mid-term (a couple of months or more) media coverage of a particular
issue in some specific situations like the above-mentioned Sou Journalism Jokyo.
Media coverage on the Aum Shinrikyo affair
Before reporting the result of the survey, we briefly look at how the mass
media in Japan reported the Aum Shinrikyo affair and its related issues.
Critics pointed out that there were many problems with the media coverage on the
Aum Shinrikyo affair, which include, but are not limited to, the overemphasis on
the affair over other important issues, over-reliance on the information leaked
by the police, and reports based upon conjectures and from only one point of
view (e.g. Asano 1995; Iimuro 1995; Kamei 1995; Nakagori 1995).
Mizuno et al. (1995) conducted a content analysis of the media coverage
regarding the Aum Shinrikyo issues, including how the TV coverage changed over
time in 1995. According to Mizuno et al., the total amount of time allocated to
the Aum issues by six major TV stations in Tokyo reached more than 10,000
minutes a week during the period between March 27 and May 28. But in late
August, the amount of the coverage dropped to one-fourth of the peak. Their
findings demonstrate that the amount of TV report was most excessive from late
March to early July.
Until when Shoko Asahara, the leader of the Aum, was arrested on May 16, every
commercial TV station had unanimously broadcasted special programs about the
Aum-related issues. According to Broadcasting Report (Media Soken, 1995), five
network stations located in Tokyo broadcasted 34 special programs on the Aum
issues from April 1 to May 15, which included 12 programs by Fuji TV, 11 by
NTV, 5 by TV Asahi, and 3 by TBS and TV Tokyo respectively. It means that
nearly every day at least one special program related to the Aum affair was
broadcasted. Many of those programs recorded high audience ratings (over 20% and
some got even over 30%). Such high audience ratings for the special programs
might be the result of reflection of the viewers' interest for the issues.
However, that type of programs attempted to attract viewers' interest by using
exaggerated program titles even when not much new information was provided.
As for the so-called wide shows, day-time entertainment programs mainly
targeted to housewives, which feature gossips and scandals of celebrities, the
situation was much more extraordinary. According to the data provided in one of
the information programs by TBS's Broadcaster on May 20, 1995, during that
week the total amount of time the whole wide shows (broadcast by five TV
commercial network stations) spared for the Aum issues was about 51 hours. The
second biggest issue at that week (the terrorist bombing targeted at Mayor
Aoshima) was broadcast for only 32 minutes. The situation was basically the
same as this during late March to early July.
Although not as terrible as television, daily newspapers' coverage on the Aum
Shinrikyo affair also deserves criticism. For example, according to
Broadcasting Report, No. 135 (Media Soken, 1995), three major dailies (Asahi
Shimbun, Mainichi Shimbun and Yomiuri Shimbun ) exclusively dealt with the
issues related to the Aum Shinrikyo on their first pages as the top news during
the two months from March 20 to May 20. During these 2-month periods, there
were only twice in the three main dailies without the Aum articles on their top
pages. Similar situation was also seen in other local newspapers. And the
situation was worse for what we call Sport papers, which is equivalent to the
tabloids in the U.S.
The problems of the Aum Shinrikyo coverage were not only its excessive amount
but also its exaggerated content. Every mass medium described the Aum Shinrikyo
cult as the dubious and peculiar group, and after the sarin gas attack, warned
the audience that the Aum members still hid poisonous sarin and were waiting for
the next chance to scatter it. What impact did such a media coverage have on the
audiences? To address this issue, we conducted a survey in Japan.
The sample of this study was those who were over 20 years old, drawn by a quota
sampling method from the Tokyo Metropolitan area (including Tokyo, Kanagawa ,
Chiba and Saitama Prefectures). Questionnaires were administered from July 10
to July 20 in 1995 by 38 trained research assistants, enrolled in a survey
method course, who were instructed to contact approximately equal numbers of
male and female adults in particular age categories: 20-29, 30-39, 40-49, 50-59,
and over 60. Two hundred ninety two usable questionnaires were returned. The
sample included 48.1% males and 51.9 % females. Respondents ranged in age from
20 to 72: 22.7% were 20-29 years old, 18.9% were 30-39 years old, 20.6% were
40-49 years old, 21.3% were 50-59 years old and 16.6 % were 60 or over.
The following types of questions were asked in the survey: (1) the amount of
exposure to the various mass media, such as total TV viewing, TV news viewing,
TV Wide Show viewing, Aum special programs viewing, newspaper reading and
magazine reading, (2) how the audience evaluated the various media reports about
the Aum affair, (3) a variety of opinions, such as the level of social anxiety,
respondents images of new religions, attitude toward the police investigation
into the Aum affair, and so on, (4) demographic variables. The questionnaire
was long and contained items assessing a variety of opinions related to the Aum
Shinrikyo issues. Because of space limitations, we report the results of only
selected items. Most of the items assessing a variety of opinions were measured
on a five-point scale.
As for the respondents' evaluation on the Aum Shinrikyo media coverage, we made
evaluation scales for television and newspaper respectively, by combining three
variables, (1)" the amount of the coverage", (2)"the content of the coverage"
and (3) " the way of reporting the issues". In each question, we coded positive
answers (such as the amount of coverage was not excessive, or the content was
not biased) as - 1, negative answer (such as the amount of the coverage was too
much or the way of reporting was unfair) as +1, and DK answer as 0. Then we
combined the three items into the evaluation scale.
This study was guided by the following four hypotheses in order to examine the
impact of the media coverage of the Aum Shinrikyo affair.
Two hypotheses are related to the rumor that some incidents (like the sarin gas
attack) would happen in Shinjuku (one of the biggest commercial districts in
Tokyo) on April 15, 1995;
(H1-1) The more people were exposed to the media coverage on the Aum Shinrikyo
issues, the more likely they were to believe that something would occur on April
(H1-2) The more people were exposed to the media coverage on the Aum Shinrikyo
issues, the more likely they were to think that others believed that something
would occur on April 15.
Two hypotheses are related to social anxiety;
(H2-1) The more people were exposed to the media coverage on the Aum Shinrikyo
affair, the more likely they were to feel uneasy about the society.
(H2-2) The more people were exposed to the media coverage on the Aum Shinrikyo
affair, the more likely they were to think that others felt uneasy about the
Two hypotheses are related to images of new religions in general;
(H3-1) The more people were exposed to the media coverage on the Aum Shinrikyo
affair, the worse their image of new religions in general becomes.
(H3-2) The more people were exposed to the media coverage on the Aum Shinrikyo
affair, the more likely they were to think that others' images of new religions
(H4-1) The more people were exposed to the media coverage on the Aum Shinrikyo
issues, the more favorable they were toward the excessive police investigation
into the Aum affair.
Regarding to H1 to H3, we asked respondents both about themselves and about
their perceptions of others' opinions because of the following reason. Some
investigators point out that television messages could influence societal-level
but not personal-level beliefs. Doob and MacDonald (1979), for example,
mentioned that "television may well act as a source of information with regard
to questions of fact, whereas it does not change people's view of how afraid
they should be" (p. 179). Tyler and Cook (1984) further examined this impersonal
impact hypothesis of mass media across various areas of risk judgment. Their
experimental studies demonstrated that (a) personal- and societal-level
perceived risks are distinct and (b) the mass media influence judgments
primarily on the societal level, not on the personal level. Like Doob and
MacDonald, Tyler and Cook concluded that "mass media reports often have an
impact upon views about the seriousness of a problem as it affects society in
general but not in altering views about personal risk" (1984, p. 707). In the
review of the cultivation literature, however, Saito (1995) pointed out that
this impersonal impact hypothesis gained only limited support. Thus, in this
study, we also tried to examine this hypothesis.
First, we examined the influence of the mass media report about the rumor that
something would happen in Shinjuku on April 15. As above mentioned, the
respondents were asked two questions related to this: whether they believed the
rumor (personal question) and whether they thought that the other people
believed it (impersonal question). Most of the respondents knew the rumor
through some kinds of media such as TV or magazines (only 6.2% of the
respondents didn't know it). Data shows that 9.3% of the respondents thought
that something would happen, while 19.4% answered that they thought the other
people believed the rumor. Similarly, 11.1% answered that they did not believed
it, but only 3.1% of the respondents said that they did not think others
believed the rumor. This result seems to provide an evidence supporting the
third-person effect hypothesis proposed by Davison (1983). The hypothesis
postulates that people think that the effect of persuasive communication is
stronger to others than to themselves, and people also tend to overestimate the
impact of the mass media on the other people. Our result shows that the
respondents in this study tended to think the impact of the media report to be
stronger to others than themselves.
In order to examine the relationship between the items related to the rumor and
the amount of exposure to various media coverage, we conducted cross-tabular
analysis. First, we found the significant relationship between the personal
question and the amount of exposure to the magazine coverage on the Aum
Shinrikyo affair (c2=24.5, p<.01; gamma=.391): the more the respondents read
the magazine articles about the Aum Shinrikyo issues, the more likely they were
to believe the rumor. But there were no clear relations between the item and
levels of exposure to other media reports.
Table 1 demonstrates the result of the cross-tabular analysis between the item
and the amount of reading magazine articles related to the Aum issues within
various subgroups. Although the pattern is slightly different depending upon
each subgroup, there is a basic tendency that the more the people read the Aum
coverage on the magazine, the more people think something will happen.
Insert Table 1 around here
As for the other item on the rumor (impersonal question), no significant
associations was found between the item and amounts of total television viewing,
television news viewing, special program viewing, or newspaper reading.
However, the significant association was observed between the item and the
amounts of wide show viewing and magazine reading: the more the respondents
were exposed to the media coverage on the Aum issues, the more likely they were
to think others believed the rumor.
Table 1 also shows the result for the impersonal question when the amount of
magazine reading was used as the media exposure measure. This table shows
percentages of the respondents who said that "many people thought something
would happen". The pattern of the association differs somewhat depending on the
subgroup. The relationship between the amount of magazine reading and the
impersonal question was strongest among the male respondents and the highly
educated. For example, the percentages of choosing the answer that many people
thought something would happen were 92.3 % of males who were most heavily
exposed to the magazine coverage, but 77.8% among those who read the magazine
moderately, and 51.9% among those who didn't read the magazine at all (c2 =12.0,
p<.01; gamma=.447 ). Similar tendency was observed in the subgroup of the highly
As seen in Table 1, our data demonstrates that both personal- and
impersonal-questions were significantly correlated with the amount of magazine
reading. Thus the data provides a contradictory result for the impersonal Impact
In the cross-tabular analysis, we controlled only one variable at a time.
Therefore we also conducted the multiple regression analysis controlling for
several variables simultaneously. First of all, since the two items on the
rumor (the personal and impersonal questions) showed a relatively strong
correlation (r=.491, p< .01), we combined these two items into a single scale
called the rumor scale (the higher points the respondents got, the more they
believed the rumor). Table 2 shows the result of the multiple regression
analysis using the rumor scale as the dependent variable.
Insert Table 2 around here
As the table shows, the amount of exposure to the Aum coverage in magazines was
significantly associated with the rumor scale (b=.261, p<.01). The television
evaluation scale was another significant predictor variable (b=.175, p<.01):
the higher they evaluate the Aum coverage on television, the more likely they
were to believe the rumor.
In sum, H 1-1 and H 1-2 were supported when the amount of magazine reading on
the Aum issues was used as the predictor variable.
As for the personal question, there were significant associations between the
item and the amount of TV news viewing, special program viewing, or newspaper
reading: the more people were exposed to these media reports, the more they
felt uneasy about the society.
Table 3 shows the result of the cross-tabular analysis when the amount of
exposure to TV special programs on the Aum Shinrikyo affair was used as the
predictor. The females, the younger respondents, those with lower education, and
the respondents whose evaluation for TV reports is higher, show the association
between the item and the amount of viewing: the more they watched the TV special
programs on the Aum issues, the more they felt social anxiety.
Insert Table 3 around here
Similar results were observed with some other media exposure variables such as
total TV viewing, TV news viewing or newspapers reading. Thus, the results
supported Hypothesis 2-1.
As for the relation between levels of the media exposure and the impersonal
question (the perceived opinion about others' anxiety), the result of the
cross-tabular analysis showed that only amount of TV news viewing was
significantly associated with the item. The percentages that the respondents
thought that others had felt social anxiety were: 42.7% for those who did not
watch TV news so much, 59.7% for those who frequently watched them and 70.8% for
those who watched the TV news very frequently (c2=12.5, p<.01, gamma=.339).
Thus the result indicates that the more people watched TV news on the Aum
Shinrikyo affair, the more likely they thought that the other people had felt
social anxiety (See table 3).
As table 3 shows, although not every association in the subgroups is
significant, there indeed are significant relations between the two variables in
many subgroups. That is, as for the amount of TV news viewing, Hypothesis 2-2
was also supported.
The results also indicate that the impersonal impact hypothesis does not seem
to be supported because there were significant relations even between the
impersonal item and the media exposure measure.
In addition, we also conducted the multiple regression analysis. First, because
the personal and impersonal questions were moderately correlated (r=.498, p<
.01), we combined them into a social anxiety scale (The higher point on the
scale corresponds to higher social anxiety). We used this scale as the dependent
variable in the regression analysis.
As table 4 shows, the level of TV news viewing was significantly related to the
scale, even simultaneously controlling for several third variables (b=.172,
p<.01). The more the respondents watch TV news about the Aum Shinrikyo affair,
the higher their points on the social anxiety scale are. In addition, the table
shows that those with higher education felt less social anxiety (b=-.127,
Insert Table 4 around here
We asked the respondents if their images of new religions in general changed
after being exposed to the massive media coverage of the Aum Shinrikyo issues.
The reason for asking this was that the mass media tended to report or depict,
not just the Aum Shinrikyo cult, but also new religions in general in a negative
Understandably, none of the respondents said their images of new religions
became better. All respondents said that their own images were unchanged (43%)
or became worse (57%). As for perceptions of others' images of new religions,
the result shows that except one respondent who said others' images changed
positively, more than 85% of the respondents thought that other people's images
of new religions became worse. This finding indicates that the respondents
thought that others were influenced by the media coverage and changed their
images of new religions negatively, but their own images were not affected so
much. That is, they overestimated the impact by the media content on others.
The data shown in Figure 1 thus provides further evidence supporting the
third-person effect hypothesis.
Insert Figure 1 around here
Table 5 shows the comparison of the respondents' own images of new religions
between before and after being exposed to the media coverage of the Aum issues.
The result indicates that the more negative images the respondents had prior to
the media coverage, the more likely their images were to become worse. Only 4%
of the respondents whose prior image was neutral changed their image negatively.
However, 28% of whose prior image was somewhat negative and 40.7% of whose prior
image was negative changed their images even more negatively. That is, those
who had more negative images toward new religions in general were more likely to
change their images worse.
Insert Table 5 around here
Next, we examined the relationship between these image items and various media
exposure measures. Again, we conducted the cross-tabular analysis. The results
show that only the amount of TV Wide Show viewing was significantly associated
with both the personal and impersonal items of the images of new religions. The
percent of the respondents who changed their image negatively (personal item)
was 20.5% of the light viewers and 35.2% of the heavy viewers (c2 =15.94,
p<.05; gamma=.169). Thus, the more they watched the Wide Shows, the more
negatively their image of new religions changed (See Table 6).
Insert Table 6 around here
Even after controlling for a third variable at a time, significant associations
were observed among some subgroups such as those with lower education
(c2=13.02, p<.05; gamma=.306) or those who talked about the Aum issues with
friends or family members less frequently (c2=13.92, p<.05; gamma=.356). To sum
up, Hypothesis 3-1 was supported when the amount of Wide Show viewing was used
as the predictor variable.
Table 6 also shows the results of the cross-tabular analysis for the impersonal
item. Similar to the case of the personal item, the tendency was observed that
the more the respondents watched the Wide Shows, the more likely they were to
think that others changed their images of new religions negatively. For
example, in the subgroup of the less educated, 34.8% of the light viewers of the
wide shows, but 60.7% of the heavy viewers chose the answer that other people's
images of new religions changed more negatively (c2=10.87, p<.05; gamma=.319).
Thus, Hypothesis 3-2 was also partly supported when the amount of wide show
exposure was used as the predictor variable.
We further conducted the multiple regression analysis. First, we combined the
personal and impersonal questions about images of new religions into a single
scale called the religion image scale (two items were significantly correlated;
r=.498, p< .01). Higher points on the scale mean more negative images of new
religions. We used this scale as the dependent variable.
As a result of a series of multiple regression analyses, it was shown that
within the highly educated group, none of the variables entered into the
equation was significant. But in the less educated group, as table 7 shows, the
amount of wide shows viewing was marginally significant (b=.191, p<.08). In
addition, in the equation model of this multiple regression analysis, sex was
also a significant variable (b=-.234, p<.05): the female respondents made their
images more negative.
Insert Table 7 around here
Among the problems with the media report of the Aum affair, one of the most
serious issues was that the mass media relied heavily on the information leaked
by the police. Critics argue that the media played a kind of the publicity role
for the police. Regarding this issue, we hypothesized that the more people were
exposed to the mass media report on the Aum Shinrikyo issues, the more positive
their attitudes toward the police investigation became.
Before analyzing our data, we briefly look at the results of surveys by
Yomiuri Shimbun. In May and June in 1995, Yomiuri conducted the public opinion
surveys including the questions about the police investigation into the Aum
Shinrikyo affair. According to the surveys, 63.3% (in March) and 69.2% (in
June) of the respondents said that the police was doing well.
Thus, although some critics pointed out that the police investigation was
excessive like arresting Aum members on other charges or with minor charges,
the general public's attitude was rather tolerant to the authority. Only 3.5%
(in March) and 2.7% (in June) of the respondents answered that the police did
too much, and one fourth of the respondents said that the police should
investigate harder (27.8% in March and 24.4% in June).
As for our data, only 4.1 % of the respondents answered that the police did too
much. 32.8% said that the police did a little too much but it could not be
helped, and 17.2% said the police did their investigation in a proper way. 12.1%
of the respondents answered that the police should do a little harder and as
much as 27.2% of the respondents said that the police should do much harder.
We examined the relation between attitudes toward the police investigation and
the various media exposure measures by cross-tabular analysis, but no
significant association was found. We further tested the following indirect
effect model of the media because there were some relationships between the
amount of media exposure and the level of social anxiety.
[Exposure to the media coverage on the Aum Shinrikyo affair]
[Feel social anxiety] [Positive attitude toward the police investigation]
However, none of partial correlations between the media exposure measures and
attitudes toward the police investigation controlling for the level of social
anxiety reached the significant level at p <.05 (partial correlations ranged
from .01 to .08). Hypothesis 4 therefore was not supported.
Although any significant relationship between the amount of the media exposure
measures and the attitudes toward the police investigation was found, there
were significant associations between levels of evaluation of media and the
attitudes toward the police investigation. As table 8 shows, except for
Newspaper evaluation scale and the evaluation for daily newspapers, we can see
the tendency that the higher the respondents evaluate the media coverage, the
more positive their attitudes toward the police investigation were.
Insert Table 8 around here
Such associations were also found when simultaneously controlling for other
variables such as the amount of media exposure, sex, and age. Table 9 shows, as
an example, the result of the multiple regression analysis when we entered the
evaluation scale for television and the amount of viewing TV programs on the Aum
issues into the regression equation.
Insert Table 9 around here
Critics pointed out that there were many problems with the massive media
coverage on the Aum Shinrikyo affair and its related issues. What impact did
such a media coverage have on the audiences? In this study, we addressed the
issues of whether the audiences felt anxiety about the society, their images of
the new religions became worse, or they formed positive attitudes toward the
excessive police investigation.
The results of the survey indicate the following: the more the respondents were
exposed to the various media reports (such as special TV programs on the Aum
issues, TV news and newspapers) on the Aum Shinrikyo affair, the more they felt
social anxiety; the more the respondents read the magazine articles about the
Aum Shinrikyo, the more likely they were to believe the rumor that something
would happen on April 15; the more the respondents watched the wide shows about
the Aum issues on television, the more their images of new religions in general
changed negative. Moreover, these tendencies were found not only on the items
about themselves (personal questions) but also about the other people
(impersonal questions), although the associations found were somewhat weaker for
the impersonal questions. The results also showed that the higher the
respondents evaluated the television coverage on the Aum issues, the more they
had positive attitudes toward the police investigation into the Aum Shinrikyo
affair. In sum, these results would provide some empirical evidence for
journalism in Japan to reconsider the way of reporting a big social issue.
As we noted earlier, this study examined the impact of the media coverage
during several months. In other words, it focused on the mid-term rather than
long-term influence of the mass media. In addition, we employed various media
exposure measures, not just total television viewing, as the predictor
variables. According to Gerbner et al.'s original definition, therefore, the
analysis in this study might not be a test of cultivation theory. Gerbner and
his associates insisted that message elements which lead to cultivation should
be those which cut across most programs and genres.
As we already pointed out, the Aum Shinrikyo affair and its related issues got
the massive media coverage and the nearly whole mass media provided the uniform
message on the limited issues to the general public. The media reports and
depictions of the Aum issues were so intensive and other important issues
virtually disappeared for a while from the media in Japan. Thus, almost
everyone could not escape from the media coverage. Under this Sou Journalism
situation, we assume that a cultivation perspective can be extended to the
impact of the mid-term (several months) media coverage and to mass media other
than television. It should be noted, however, that although we assumed that the
media reports may have an strong influence on the audience under Sou Journalism
situation, we do not claim that everybody should be equally influenced by the
media content, but as shown by our data, the degree of the media influence
differs depending upon the individual's social situation.
Adoni, H., & Maine, S. (1984) Media and social construction of reality.
Communication Research, 11(3), 323-340.
Arai, N. (1979) Journalism, Toyo keizai shimposha (in Japanese).
Asano, K. (1995) Serious problems of witch-hunting media report on the Aum
issues (Aum majogari hodo no daimondai)", Hosobunka, July, 52-65 (in Japanese).
Berger, A.A. (1995) Essentials of mass communication theory. CA:Sage
Davison, W. P. (1983) The third-person effect in communication. Public Opinion
Quarterly, 47, 1-15.
Dearing, J. W., & Rogers, E. M. (1996) Agenda-Setting. CA: Sage Publication.
Doob, A .W. , & MacDonald, G .E . (1979) . Television viewing and fear
of victimization: Is the relationship causal ? Journal of Personality and
Social Psychology, 37, 170-179.
Gerbner, G., & Gross, L. (1976) Living with television: The violence profile.
Journal of Communication, 26(2), 172-199.
Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1982). Charting the
mainstream: Television's contribution to political orientation, Journal of
Communication, 32(2), 100-127.
Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1994). Growing up with
television: The cultivation perspective. In J. Bryant & D. Zillmann (Eds.),
Media effects: Advances in theory and research (pp. 17- 41). Hillsdale, NJ:
Iimuro, K. (1995). "The quality of Media questioned in the extraordinary
situation ('Ijo na toki' ni towareru media no shitsu)" , Shimbun Kenkyu, 528,
25-28 (in Japanese).
Kamei, J. (1995). The days of poisonous gas and the Aum (Dokugasu to Aum no
hibi)", Hoso Hihyo, July issue ,30-35 (in Japanese).
Matabane, P.W. ( 1988). Television and the Black audience: cultivating
moderate perspectives on racial integration. Journal o f Communication,
38 (4), 21 - 31 .
McCombs, M. E., & Show, D. L. (1972) The agenda-setting function of the mass
media. Public Opinion Quarterly, 36, 176-187.
Media Soken (1995). Broadcasting Report (Hoso Report) , No.135, Banseisha (in
Mizuno,H. et al. (1995). Mass media coverage on the Aum Shinrikyo issues and
the opinion survey in Tokyo (Aum Shinrikyo o meguru Masukomi hodo to tomin no
ishiki chosa)" , Saitama Daigaku Kiyo, 31(2), 63-108 (in Japanese).
Morgan, M. (1982 ). Television and adolescents' sex-role stereotypes:
A longitudinal study. Journal of Personality and Social Psychology, 43, 947
Nakagori, H. (1995). Neglecting the ABC of reporting, a flood of information
relied on conjectures and rumors (Shuzai no iroha mushi, suisoku denbun hodo no
kohzui)" , Sogo Journalism Kenkyu, 153, 33-39 (in Japanese).
Noelle-Newman, E. (1974) "The spiral of silence: A theory of public opinion",
Journal of communication, 24, 43-51.
Ogless, R .M., & Hoffner, C. (1987) Film violence and perceptions of crime: The
cultivation effect. In M. L. McLaughlin (Ed.), Communication yearbook 10,
384-94, Newbury Park, CA: Sage.
Potter, W.J. (1993). Cultivation theory and research: A conceptual critique.
Human Communication Research, 19, 564-601.
Saito, S. (1995) Cultivation theory revisited: Another look at the theory and
implications for future research. Paper presented at the annual meeting of the
Communication theory and methodology division of the Association for Education
in Journalism and Mass Communication, Washington, D.C., August.
Sparks, G.G. , & Ogles, R.M. (1990 ). The difference between fear of
victimization and the probability of being victimized: Implications for
cultivation. Journal of Broadcasting and Electronic Media, 34, 351 -358.
Tyler, T.R. , & Cook, F.L. (1984 ). The mass media and judgments of
risk: Distinguishing impact on personal and societal level judgments.
Journal of Personality and Social Psychology, 47, 693 - 708.