ArticlePDF Available

Expanding and Validating Applications of the Willingness to Self-Censor Scale: Self-Censorship and Media Advisers' Comfort Level with Controversial Topics

Article

Expanding and Validating Applications of the Willingness to Self-Censor Scale: Self-Censorship and Media Advisers' Comfort Level with Controversial Topics

Abstract

Research on the Willingness to Self-Censor (WTSC) scale posits that the desire to withhold one's opinion is an intrinsic, as opposed to situational, trait. This study of high school media advisers (N=563) revealed that advisers who rated high on WTSC were more likely to state lower levels of comfort with the coverage of five controversial topic areas in their student media. These findings held even when accounting for key demographics, fear of reprisal for running the stories, job enjoyment, and the advisers' perception of their principal's comfort level with the topic.
EXPAND~VGAND
VDA~G
APPLlCAlTONS
OF
7H.E
W..LZNGN€SS
TO
SELF-CENSOR
SCALE.'
SELF-CENSORSHIP
AND
MEDIA
ADVISERS'
COMFORT
LEVEL
WITH
CO~OVERSLAL
Toms
3y Vincent
F.
Filak, Scott Reinardy, and Adam Maksl
Research
on
the Willingness to Self-censor
(WTSC)
scale posits that
the desire to withhold one's opinion is an intrinsic, as opposed to situa-
tional, trait. This study of high school media advisers
(N=563)
revealed
that advisers who rated high on
WTSC
were more likely to state lower
levels of comfort with the coverage of jive controversial topic areas in
their student media. These findings held even when accounting for
key demographics, fear of reprisal for running the stories,
job
enjoyment,
and the advisers' perception of their principal's comfort level with the
topic.
The ability to express one's self in an unfettered manner is at the
core of the Bill of Rights to the
U.S.
Constitution and is often viewed as
a "natural right," thus not endowed by government but rather inherent
to the human condition. However, students who learn about these rights
in
a classroom often find them less applicable
in
practical settings, such
as student media. At the high school level alone, the Student Press Law
Center reports an almost daily increase in incidents of censorship of
high school media, to say nothing of the cases that
go
unexamined.'
While much research has examined the legal and social issues sur-
rounding overt censorship,2 very little research has studied self-censor-
ship and its role in limiting expression in high school media. Dickson
argued that while the student press is often stifled by heavy-handed
administrators,
more
surreptitious acts of self-censorship by advisers
often
go
unnoticed. It is at this level, he argued, that the greatest amount
of damage is done.3
Recent research reveals that self-censorship
is
a measurable con-
struct that differs among individuals? Built as an extension and revision
of
earlier spiral of silence models: the Willingness to Self-censor scale
was used to study opinion suppression during conversations of contro-
versial topics6 and discussions in polarized political
climate^.^
In
each
case, individuals who rated higher on the WTSC scale were less likely to
Vincent
F.
Filak is an assistant professor in the Department of Journalism at the
University
of
Wisconsin-Oshkosh; Scott Reinardy is an assistant professor in the
William Allen White School of Journalism and Mass Communications at the University
of Kansas; and Adam Maksl is an instructor at Ball State University.
J6MCQuarterly
Vol.
86,
No.
2
Smmm
2009
368-382
02009
AEIMC
368
IOURNAUSM
b
MASS
COMMUNICATION
QUARTERLY
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
speak out when they felt the climate was polarized or when the conver-
sation clearly favored a position they did not support.
While these studies are valuable in establishing a scale to measure
the concept and affirming that individuals do differ as to the degree and
nature of their self-expression, several key elements of
this
theory remain
unexplored. First, while the topics in previous studies are controversial,
the hypothetical risks are not as easily perceivable as those posited in this
study. While social isolationism is detrimental to one’s sense of self,
expressing an unpopular opinion regarding a political figure or as part
of a hypothetical scenario does not share the same level of concern as
espousing an unpopular opinion when one’s career is potentially at
stake. Second, while the studies have attempted
to
control a number of
variables that might otherwise explain the WTSC, such as dispositional
shyness, none of the examinations controlled for specific outcomes, such
as fear of job loss. Though the “fears” in
this
study necessarily remain
“hypothetical,” we believe they are easier for our participants to concep-
tualize and react to, as opposed to some of the other fears examined in
previous work. Thus,
this
study seeks
to
examine the following question:
Is the individual desire to express or withhold one’s opinion something
that can be explained away by practical concerns or does it exist beyond
specific situational fears?
We are attempting to replicate and extend the budding self-censor-
ship research through an application of the WTSC to a group of individ-
uals who are pressured to control controversial expression-high school
media advisers. By applying the WTSC scale and simultaneously con-
trolling for specific job-oriented variables,
this
study may advance the
research concerning self-censorship as an inherent and measurable trait.
Censorship
and
High School
Media.
For the past twenty years, the
Literahrre
U.S.
Supreme Court case that has controlled high school media has been
Review
Hazelwood
v.
Kuhlmeier,
in which the majority ruled that principals could
censor student media if they could demonstrate a legitimate educational
reason to do
so?
In the years that followed, research demonstrated that
the vague wording of
this
decision, along with a clear power imbalance
between administrators and advisers, led not only to a great deal of cen-
sorship; but also to a tense environment for advisers and editors.
However, even prior to
Hazelwood,
what high school media outlets
were allowed to report was not usually a matter of law, but instead dic-
tated by “the whim of those in charge.”1° The seminal
Captive Voices
report found that censorship was rarely enacted due to fear of legal
issues.” Instead, principals censored material to limit dissemination of
information they viewed to be too controversial for the students.
When student media run controversial material, editors and advis-
ers often face serious consequences.
In
March
2008,
Linda Kane, the
adviser of a Naperville, Ill., high school newspaper, was told she would
not be allowed to teach journalism or advise the paper the following year
after students ran an article that administrators felt “glorified drug use.”
In Kane’s official letter of reprimand, the principal had specifically noted
EXPANDING
AND
VALlDATING APPLICATIONS
369
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
Kane’s decision to speak to the local press about the controversy as one
of the reasons for her removal. (Kane told the local newspaper that
administrators at her school “don’t know squat” about First
Amendment law.)12
Kay Powers, a high school teacher and media adviser in the state
of Washington, was fired
in
November
2007
after she helped students
produce an “underground newspaper. It was the second time in
two
years the school had been embroiled in a censorship issue. In
2005,
stu-
dent editors of the newspaper sued the district after a new principal
enacted a prior review policy. The suit was settled when the court ruled
that the principal could review material, but he could not prevent it
from being p~b1ished.l~
Amy Sorrell,
an
adviser of an Indiana high school newspaper, was
suspended in March
2007
when she allowed a student to publish a col-
umn
that advocated tolerance of homosexuals.14 In order to be reinstat-
ed, she had to agree to teach only English courses, transfer to another
school in the district, and not advise a student publication for three
years. Sorrell noted that she did not have the financial standing to fight
the
school in courts. Her attorney said if Sorrell fought the school and
lost, it would
be
“a death sentence” for her career. In the end, Sorrell
resigned and took a job at a private high school.
These examples illustrate the potential harm advisers face as a
result of free expression. Other examples include
two
instructors in
Ohio who declined to return to advising after school officials balked at
articles in their papers regarding sex and oral sex,15 and an adviser in
Virginia who was removed after the student newspaper ran a series of
stories on homosexuality and another article on an anonymous Web site
called “Post Secret.”16 In each case, it was more than a fear of ostracism
that advisers faced, and in none of these cases was there a legal concern
with the material the students published.
The
Willingness
to
Self-censor.
In
the years since the advent of
Noelle-Neumann’s spiral of silence theory, researchers have examined
the model as a whole and in various stages.17 The theory outlines a
causal chain in which individuals gain a “quasi-statistical sense” regard-
ing public opinion on an issue. They then assess to what degree
their opinion is congruent with the majority opinion and, based on
the level of public support they feel for their position, they either decide
to engage in public debate or retreat from it. Spiral of silence theo-
rists argue that a fear of social isolation drives
this
desire to self-
censor.
Research has found qualified support for this model in studies
about public opinion about gays in the military,18 abortion,19 children
with AIDS attending public schoo1s,2° English as an “official lan-
guagelrrZ1 and affirmative action.” In a rare examination of the model in
a journalism environment, Filak and Price found that freelance news
photographers were likely to fit the pattern of the spiral if they felt iso-
lated from others in their field. When they felt part of a larger group,
they were more likely to express themselves, even in the face of a hos-
tile crowd.= This supported the work of some critics of the spiral of
3
70
lOURNALISM
b
MASS
cOh4MUNICATlON OUARTERLY
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
silence model, who argue that fear of social isolationism is not a driving
force when it comes to self-censorsh~p.~~
More recent research on self-censorship has also yielded findings
counter to the original model. A meta-analysis by Glynn, Hayes, and
Shanahan revealed weak support for the supposition that the perception
of public opinion correlated with a willingness to express one’s own
opinion. To that end, the authors hypothesized that individual differ-
ences are a greater influence in deciding whether to speak out.=
In
other
words, not everyone is equally susceptible to social conditions that can
lead to self-censorship.
The authors developed and validated a scale that attempted to
measure an individual’s willingness to self-censor.z6 In each study, partic-
ipants who rated higher on the WTSC scale were less likely to express
their opinions in a climate that appeared
to
be unfavorable to their opin-
ion. The authors found that self-censors were more anxious about com-
municating with others, more likely to worry about what other people
thought about them, and lower in self-esteem. Thus, when faced with sit-
uations in which self-censors might be called upon to defend an unpop-
ular opinion, they are likely to acquiesce to the demands of superiors.
In
validating the scale, Hayes and his c011eagues~~ found that when
a conversation was manipulated to be either congruent or incongruent
with the participants’ opinion on a controversial subject, individuals in
the congruent condition were more likely to speak out. That said, the par-
ticipants who scored higher on the WTSC scale were more likely to be
limited by the opinion climate. This study reinforced previous evidence
that, though it manifests itself in specific situations, self-censorship is an
intrinsic trait. Additional work in this area has also reinforced this
In this limited amount of research, little has been done to examine
areas in which the risk is germane to a practical concern. In the early
work by Hayes and c011eagues,2~ participants were told to imagine they
were at a party in which a group of people began discussing a controver-
sial topic. The condition (friendly vs. hostile toward the participant’s
opinion) and the topic (the death penalty, affirmative action, or the van-
dalizing of
SUVs
as an environmental statement) were manipulated. The
participants were then asked how likely they were to express their true
opinion. While the model held, none of those conditions pose adverse
risk based on outcome, such as a fear of job loss, for expressing those
opinions.
Similarly, in Hayes, Uldall, and Glynn’s follow-up to the valida-
tion study, the authors used an experimental approach to codify earlier
findings. In this case, student participants were left alone with
two
confederates who were beginning to discuss a screening system at the
university that would limit freedom but improve safety on campus. The
condition was again manipulated to be either hostile or friendly toward
the participant’s opinion. The climate did not affect all participants
equally and the WTSC scale revealed differences in speaking out
between those who scored high on the WTSC and those who did not.30
While
this
study moved closer to actual risk and personal attachment
EXPANDING
AND
VALIDATING APPLICATIONS
3
71
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
(the screening would occur on the participants’ campus and had the
potential to affect them), no clear-cut punitive measure was attached to
these outcomes.
In the case of self-censorship as it pertains to high school newspa-
per advisers, a palpable fear of being fired or reprimanded directly
attaches itself to each controversial story the paper decides to cover.
In
some cases, advisers and students fear reprisal from administrators to
the point that they avoid covering important and yet controversial top-
ics in the paper. In the
Captive
Voices
report, Nelson revealed that stu-
dents often operated under self-imposed censorship, in which they
shied away from reporting on topics because they feared getting into
trouble. “Such a chilling effect [students not reporting on topics for fear
of punishment] discourages most students and results in the most per-
vasive form of censorshipthat imposed by the students on them-
sel~es.’’~’ In a follow-up study to
Captive
Voices,
Bowen reported advis-
ers and editors continued to shy away from controversial topics because
they feared censorship or worse from the school’s
With
this
in mind, we designed the following study
to
advance
research regarding the willingness to self-censor with three specific
ideas in mind. First, we wanted
to
extend the research into another pop-
ulation (educators and media advisers) and into another realm (work-
place environment). The scale has been validated with students and
within the realm of politics, but little other research has been conducted
outside of these areas. Second, we wanted to move the research into an
area in which the fear associated with potentially speaking out is more
germane to the participants. By examining controversy as it pertains to
a job-related outcome in an area in which controversy often leads to
deleterious outcomes, we believe we can more clearly define whether
self-censorship is situational or intrinsic. Finally, we wanted
to
account
for specific variables that have the potential to influence one’s willing-
ness to self-censor. If advisers report high levels of job enjoyment and
yet high levels of fear that they might lose that job, those variables might
co-vary with the self-censorship variable
to
the point of overtaking the
variance associated with the
WTSC.
Therefore, we hypothesized the following:
H1:
Adviser ratings on the self-censorship scale will
independently predict the advisers’ own comfort level
regarding coverage of sexual issues in the advisers’ media
outlet, even when controlling for their perception of their
principal’s comfort level regarding the coverage of the issue,
the level to which they fear reprisals, and the degree to
which they report enjoying their job.
H2:
Adviser ratings on the self-censorship scale will
independently predict the advisers’ own comfort level
regarding coverage of substance use
/
abuse in the advisers’
media outlet, even when controlling for their perception of
their principal’s comfort level regarding the coverage of the
3
72
JOURNALISM
&
MASS
COMMUNICATION QUARTERLY
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
issue, the level to which they fear reprisals, and the degree to
which they report enjoying their job.
H3:
Adviser ratings on the self-censorship scale will
independently predict the advisers’ own comfort level
regarding coverage of student misdeeds in the advisers’
media outlet, even when controlling for their perception of
their principal’s comfort level regarding the coverage of the
issue, the level to which they fear reprisals, and the degree to
which they report enjoying their job.
H4
Adviser ratings on
the
self-censorship scale will
independently predict the advisers’ own comfort level
regarding coverage of the school’s curriculum in the advis-
ers’
media outlet, even when controlling for their perception
of their principal’s comfort level regarding the coverage of
the issue, the level to which they fear reprisals, and the
degree to which they report enjoying their job.
H5:
Adviser ratings on the self-censorship scale will
independently predict the advisers’ own comfort level
regarding coverage of administration issues in the advisers’
media outlet, even when controlling for their perception of
their principal’s comfort level regarding the coverage of the
issue, the level to which they fear reprisals, and the degree to
which they report enjoying their job.
A
sample of high school media advisers was selected from the
membership list of the Journalism Education Association, which is ”the
only independent national scholastic journalism organization for teach-
ers and advisers.”= We selected the names and contact information of the
2,108
members who noted they were media advisers and found that only
1,577
listed an e-mail address. We sent an e-mail to those members,
which contained a link to an online survey, as well as a
PDF
copy
of
the
instrument that the adviser could print and return via mail. We followed
up with a reminder e-mail about one week later.
About one-fifth of the sent messages
(293)
were returned as unde-
liverable, inactive,
or
sent to those who no longer advise a student pub-
lication, leaving
1,284
potential participants. Of those individuals
remaining in the pool,
602
responded, giving
us
a response rate of about
47%.
The return rate
for
this survey is about three times that of similar
studies involving student media advisers.%
We eliminated
39
responses because they did not complete at least
half of the form, leaving us with
563
cases for analysis.
For
the cases in
the data set, we used mean substitution to replace missing data, which
allows for a more robust analysis of the data and internally consistent
results ~ets.3~
To
prevent risk of substantially changing the data set, no
more than
5%
of any one case
or
variable was replaced in this manner.
Method
EXPANDING
AND VALIDATING
APPLICATIONS
373
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
Advisers were asked to react to several controversial topics in
terms of how comfortable they would feel in seeing articles on these top-
ics in the media outlet they advise. The topics fit into five areas, three of
which were multiple-item variables and
two
others were single-item
variables. The multiple-item variables pertained to sex (oral sex, sexual
intercourse, birth control, and homosexuality), substance
use
/
abuse
(drug use, alcohol use, and smoking) and misdeeds (hazing, vandalism,
and crime). Two single-item variables pertained to the coverage of cur-
riculum at the school and criticism of the administration.
The topics were drawn from previous work% which drew the top-
ics from the Student Press Law Center’s Web site. The SPLC tracks cases
in which advisers
or
student journalists are censored
or
otherwise pun-
ished for engaging in a variety of activities, including publishing contro-
versial topics. Thus, we were assured that not only were these topics rel-
evant to the student papers, but also that there would be a likelihood of
advisers seeing them as controversial.
Prior to presenting the topics to the advisers, we explained that
they were to assume that any story on the topic would be newsworthy
and free of error. In other words, they should be simply reacting to the
topic.
A
7-point Likert-like scale was used for each item, ranging from
not at all comfortable
(1)
to very comfortable (7).
Advisers then were asked to rate the same topics a second time,
but this time they should estimate to what degree they felt their princi-
pal would feel comfortable if a story on that topic were to run in the
adviser’s media outlet. While previous work in this area used a
matched-pairs approach examining both principals and advisers from
specific schools, we eschewed that approach this time and instead relied
on what Noelle-Neumann called the “quasi-statistical sense” approach
to estimating the principals’ moods.
We also had the participants complete the Willingness to Self-
Censor scale. In order to maintain congruency among
our
measures, we
used a 7-point scaled, as opposed to the 5-point version used in the
Hayes, Glynn, and Shanahan studies. The structure of the measure,
however, remained the same with higher scores indicating a greater will-
ingness to self-censor. The scale contains eight items and measures an
individual’s willingness to withhold one’s own opinion during interper-
sonal contact when faced with what could be a hostile
or
disagreeable
audience. The items met an acceptable reliability level for the creation of
a single variable (alpha=
.83).
In addition, we measured
two
variables that attached themselves
directly to the risks associated with speaking out if the adviser were to
do
so
in a real situation.
A
three-item variable was used to measure to
what degree the participants enjoyed their current job (e.g., “All in all,
I
am satisfied with my job;” alpha
=
32).
A second three-item variable
measured the degree to which they feared reprisals if their media outlet
ran a controversial story (e.g., “I feel my job is at risk when
(THE
MEDIA
OUTLET) publishes certain types of material; alpha=
.88).
The survey asked for demographic information, including in what
state the adviser teaches, age, gender, number of years teaching, number
3
74 JOURNALISM
&
MASS
COMMUNICATION
QUARTERLY
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
of years advising student media, number of years working in current
school, number of years in current position, and whether or not the
adviser is tenured. Respondents were also asked if they or their schools
are members of state or national student journalism organizations, if
their state has a student free expression law, if their school or district
has a free expression policy,
and
if
the publication is considered by
school policy or practice to be a public forum. Advisers were asked
what publications they advise and if they teach at public or private
institutions.
Participant Demographics.
Our sample contained advisers from
forty-five states and the District of Columbia. More advisers in this
sample came from public institutions than private ones
(90%
and
lo%,
respectively) and more respondents were women than men
(76%
and
24%,
respectively). The average adviser was
41.7
years old and had
13.9
years of teaching experience and
10.9
years of advising experience.
Professional experience was prominent, with
62.3%
of the respon-
dents stating they had worked in a newsroom of some kind prior to
advising.
Data Analyses.
Prior to engaging in our hypotheses testing, we
examined a correlation matrix to assess whether any of the demo-
graphic variables we collected would impact our outcome variables.
At the bivariate level, we found that age and years spent advising both
correlated significantly with the outcome variables and thus were kept
for future analyses. In addition, we noted that the variables pertaining
to whether the participants thought their state had a law protecting
student p~blications~~ and whether the school district had a policy that
declared student publications open public forums also correlated at the
bivariate level.
We conducted five (one per topic), four-step regressions in which
we entered the demographic variables into the first block, the
two
job-
oriented variables into the second block, the perception of the princi-
pal's reaction to the topic in the third block, and the Willingness to Self
Censor variable in the fourth block. This allowed
us
to better examine
changes along the various points of the regression. Each regression
used the individual's own comfort ratings on the topic as the depend-
ent variable.
H1
stated that adviser ratings on the self-censorship scale will
independently predict the advisers'
own
comfort level regarding cov-
erage of sexual issues in the advisers' media outlet, even when control-
ling for their perception of their principal's comfort level regarding the
coverage of the issue, the level to which they fear reprisals, and the
degree to which they report enjoying their job. As Table
1
shows, the
regression was predictive and strong, with the demographic variables
losing power in each subsequent step and the final step showing a sta-
tistically significant R-square change.
The final
two
steps of the regression revealed that the highest
beta weights rested with
the
perception of the principal's attitude
EXPANDING
AND VAUDAJINC
APPLICATIONS
3
7.5
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
TABLE
1
Hierarchical Regression Analysis of Predictors of Advisers' Comfort Level in Publishing Stories
on
Sexual Topics (Betas,
N=
563)
Predictors Step
1
Step
2
Step
3
Step
4
Age
Years Advising
Percept. State Law
Percept. Dist. Policy
Job
Enjoyment
Fear of Reprisal
Percept.
of
Principal's Comfort
WTSC
R-square
Adj. R-square
R-square Change
Sig. of Change
.02
.01
.06
.14**'
.03
.02
.03
.003
.01
-.04
.02
.12**
.03
-.30***
.12
.ll
.09
.ooo
.02
-.05
.oo
.lo**
.03
-.06
.52***
.33
.32
.21
.ooo
.oo
.oo
-.03
.lo**
.oo
-.02
.50***
-.19***
.36
.35
.03
.Ooo
(beta=
.50)
and the willingness to self-censor variable (beta=
-.19)
while
a
perception of a supportive school district policy also remained signif-
icant (beta=
.lo).
The
direction of
the
beta weight for the
WTSC
indi-
cates that the higher an individual scored on that variable, the less com-
fortable he/she felt regarding the publication of stories on sexual topics.
Thus, while the sense of how the principal might react was heavily pre-
dictive
of
the adviser's
own
comfort level, higher ratings on the WTSC
scale independently predicted the adviser's own rating as well.
H1
was
supported.
H2
was tested with a similar model, but with the adviser's own
comfort level regarding coverage of substance use/abuse in the advis-
er's media outlet as the dependent variable. The regression again was
predictive with both the principal's perceptual comfort level (beta=
54)
and the WTSC scale (beta=
-.14)
proving to be significant predictors in
the full regression.
H2
was supported (see Table
2).
H3
used the adviser's comfort level regarding coverage of student
misdeeds in the adviser's media outlet
as
the
dependent variable.
As
Table
3
shows, the regression was once again predictive, following the
pattern of regression in
H1,
where perception of a supportive school dis-
trict policy (beta=
.09),
principal's perceptual comfort level (beta=
.47),
and WTSC (beta=
-.15)
remained significant predictors in the full model.
H3
was supported.
H4
used adviser comfort level with coverage of the
school's
cur-
riculum as the dependent variable. The full model,
as
outlined in Table
4,
was predictive
and,
once again, perception of a supportive school dis-
trict policy (beta=
.08),
principal's perceptual comfort level (beta=
.a),
and
WTSC
(beta=
-.14)
remained significant predictors in the full model.
H4
was supported.
376
JOURNALISM
b
MASS
COMMUNICATION
QUARTERLY
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
TABLE
2
Hierarchical Regression Analysis of Predictors of Advisers' Comfort Level
in
Publishing Stories
on
Substance UselAbuse
Topics
(Betas,
N=
563)
Predictors Step
1
Step
2
Step
3
Step
4
Age
Years Advising
Percept. State Law
Percept. Dist. Policy
Job
Enjoyment
Fear of Reprisal
Percept. of Principal's Comfort
WTSC
.11*
.01
.01
.lo*
.lo*
-.02
-.02
.08
.04
-.03
-.03
.05
.03
-.02
-.03
.05
.03
-.21***
.01
-.Ol
-.02
-.02
.55***
.54***
-.14**"
R-square
Adj. R-square
R-square change
Sig. of change
*
p
<
.05;
**p
<
.01;
*p
<
.001
.03
.02
.03
.006
.07
.06
.05
.Ooo
.33
.32
.25
.OOO
.34
.33
.02
.om
Finally,
H5
used the adviser's
own
comfort level regarding cover-
age of administration issues
in
the adviser's media outlet as a dependent
variable. This regression was the strongest of the group (adj. R-square=
.38)
and held several additional predictors into
the
fourth stage of the
model. Years spent advising (beta=
.09)
and fear of reprisal (beta=
-.12)
TABLE
3
Hierarchical Regression Analysis of Predictors of Advisers' Comfort Level
in
Publishing Stories
on
Student Misdeeds Topics (Betas,
N=
563)
Predictors Step
1
Step
2
Step
3
Step
4
Age
Years Advising
Percept. State Law
Percept. Dist. Policy
Job
Enjoyment
Fear of Reprisal
Percept. of Principal's
Comfort
WTSC
.08
-.01
.06
.12**
.07
-.04
.03
.11*
.03
-.03
.oo
.09*
.02
-.02
.oo
.09*
.08
-.21***
.04
-.04
.01
-.01
.48*** .47***
-.15***
R-square
Adj. R-square
R-square change
Sig. of change
.03
.02
.03
.003
.09
.08
.06
.ooo
.27
.27
.19
.ooo
.29
.28
.02
.ooo
~~~
EXPANDING
AND
VAUDATING APPLICATIONS
377
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
TABLE
4
Hierarchical Regression Analysis
of
Predictors
of
Advisers' Comfort
Level
in Publishing Stories
on
Curriculum Topics (Betas,
N=
563)
Predictors Step
1
Step
2
Step
3
Step
4
Age
Years Advising
Percept. State Law
Percept. Dist. Policy
Job
Enjoyment
Fear of Reprisal
Percept.
of
Principal's Comfort
WTSC
~
-.01
.05
.01
.13**
R
-
s
q u
a
r
e
Adj. R-square
R-square change
Sig. of change
.02
.01
.02
.021
-.03
.03
-.01
.12**
-.01
.oo
-.03
.08*
-.02
.01
-.03
.08*
.lo*
-.16'**
.06
.05
.04
.ooo
.04
-.01
.45***
.23
.22
.17
.OOO
.02
.02
.44***
-.14***
.25
.24
.02
.Ooo
were significant in the full model as were the perception of a supportive
district policy (beta=
.lo),
principal's perceptual comfort level (beta=
.45),
and WTSC (beta=
-.21).
H5
was supported (see Table
5).
Discussion
and
Conclusion
The purpose of this work was to advance and support the theoret-
ical suppositions presented by Hayes and his colleagues
in
their work
on
the concept of willingness to self-censor. In addressing five contro-
versial topical areas, advisers who rated higher on the WTSC scale
reported lower levels of comfort
in
seeing these topics covered in their
media outlets. The outcomes
in
each of these analyses held even when
accounting for the opinion climate issues noted in previous research and
job-specific variables of concern. Taken as a whole, these results suggest
that self-censorship is clearly a measurable, intrinsic trait that can vary
from individual to individual.
The outcomes here also serve a practical purpose and unveil a
disturbing concern. Budding journalists often rely on their media advis-
ers
to
be advocates and sounding boards for ideas of journalistic im-
portance. To that end, advisers who rate highest in self-censorship are
likely to be not only self-censors, but surreptitious news censors as well.
The inherent desire
to
avoid conflict will likely have these advisers
encouraging students to avoid a story that might be important, but
would also lead to conflict with administrators, parents, or community
members.
In
addition, previous research has suggested that self-censorship
is not likely to diminish over time or through changes in circumstances.
Those rating lower on the self-censorship scale might not
be
eager to
publish controversial material if they have little experience with
stu-
3
78
JOURNALISM
&
MASS
COMMUNICATION
QUARTERLY
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
TABLE
5
Hierarchical Regression Analysis
of
Predictors
of
Advisers' Comfort Level in Publishing Stories
on
Administrative Topics (Betas,
N=
563)
Predictors Step
1
Step
2
Step
3
Step
4
Age
Years Advising
Percept. State Law
Percept. Dist. Policy
Job
Enjoyment
Fear
of
Reprisal
Percept.
of
Principal's Comfort
WTSC
R-square
Adj.
R-square
R-square change
Sig.
of
change
-.04
.14*'
.08*
.17***
.M
.05
.06
.ooo
-.04
.09
.04
.14**
-.01
-.37***
.19
.18
.13
.Ooo
-.04
.07
.04
.14**
-.01
-.17***
.47***
.35
.34
.17
.OOO
-.M
.09'
.04
.lo**
-.09*
-.12***
.45***
-.21***
.39
.38
.04
.OOo
dent media or if they are under the thumb of a dictatorial administrator.
That said, when those constraints are removed, they would be more like-
ly to flourish than would an adviser who rated high on the WTSC scale.
It is difficult to know what, if anything, can be done to ameliorate self-
censorship. Hayes, Uldall, and Shanahan have argued that studies on
self-censorship have not provided much in
the
way of examining the
cognitive processes that lead some individuals to sense a hostile opinion
climate and withdraw.%
A
deeper look into
this
could help ascertain how
self-censors operate cognitively and thus allow ways to inhibit these out-
comes. Future research should seek to examine these issues.
Future research should also look into potential consequences of
self-censoring behavior that
go
beyond the individual who self-censors.
A
study of high school media editors with high and low self-censoring
advisers could be important to determine if self-censorship leads to
media censorship. Furthermore, additional field research and experi-
mental work should be done on this topic to test the WTSC in a con-
trolled, experimental setting. This will further allow
us
to codify the
results outlined here. Finally, psychology research has demonstrated that
social processes can be transmitted from educators to students.39 To that
end, it would be interesting to see whether self-censorship of superiors
"infects" subordinates.
This research is not without its limitations. First, the variables that
pertained to the principal's attitudes were based on the perception of the
adviser as opposed to the actual opinions held by the principal. While
much of the spiral of silence and willingness to self-censor research has
relied on perception,
this
measure carries with it the inherent risk of
being biased and thus undermining various aspects of the research.
Second, we did not measure some variables, most notably dispositional
EXPANDING
AND VALIDATING
APPLICATIONS
3
79
-
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
shyness, that have been examined in previous work in validating the
WTSC scale. Previous work has demonstrated that the shyness measure
has not impacted the outcomes of WTSC, but given the relative newness
of
this
area of research, additional confirmation would have been worth-
while. Finally, additional and unforeseen variables could have affected
the outcome of the regression. As we noted, practical issues should be
accounted for in
this
theoretical examination.
To
that end, we included
both the job enjoyment and the fear of reprisal variables to help further
study
this
topic. That said, other variables, such as a particularly vocal
parent-teacher organization, could just as easily cause a teacher to fear
running controversial stories in the student paper. Nonetheless,
our
analyses have demonstrated that an overriding sense of what the prin-
cipal does and does not like along with the intrinsic desire to self-censor
were the strongest predictors
in
each regression. As such, other extrane-
ous
variables are unlikely to alter the results of this work.
NOTES
1.
splc.org/newsflash.asp
2.
J.
Willam Click and Lillian Lodge Kopenhaver, ”High School
Newspapers Still Censored Thirty Years After Tinker,”
Journalism
&
Mass Communication Quarterly
78
(summer
2001): 321-39.
3.
Thomas Dickson, “Self-Censorship and Freedom of the Public
High School Press,”
Journalism Educator
49
(autumn
1994): 56-63.
4.
Andrew Hayes, “Exploring the Forms of Self-censorship:
On
the
Spiral of Silence and the Use of Opinion Expression Avoidance
Strategies,“
Journal
of
Communication
57
(December
2007): 785-802;
Andrew Hayes, Carroll Glynn, and James Shanahan, “Willingness to
Self-censor: A Construct and Measurement for Public Opinion
Research,”
International Journal
of
Public Opinion
17
(autumn
2005): 298-
323;
Andrew Hayes, Carroll Glynn, and James Shanahan, “Validating
the Willingness to Self-censor Scale: Individual Differences in the Effect
of the Climate of Opinion on Opinion Expression,”
International Journal
of
Public
Opinion
17
(winter
2005): 443-55.
5.
Elisabeth Noelle-Neumann, ”The Spiral of Silence: A Theory of
Public Opinion,”
Journal
of
Communication
24
(spring
1974): 43-51;
Elisabeth Noelle-Neumann, ”Advances
in
Spiral
of
Silence Research,”
KEIO Communication Review
10
(spring
1989): 3-29;
Elisabeth Noelle-
Neumann,
The Spiral
of
Silence: Public Opinion-Our Social Skin
(Chicago
IL:
University of Chicago Press,
1989).
6.
Hayes, Glynn,
and
Shanahan, ”Validating the Willingness to Self-
Censor Scale.”
7.
Andrew Hayes, Dietram Scheufele, and Michael Huge, ”Nonpar-
ticipation as Self-censorship: Publicly Observable Political Activity in a
Polarized Opinion Climate,”
Political Behavior
28
(September
2006): 259-
83. 8.
Hazelwood School District
v.
Kuhlmeier,
484
U.S.
260 (1988).
9.
Thomas Dickson, ”Attitudes of High School Principals about
380
\OURNAUSM
9
MASS
COMMUNICATION
QUARTERLY
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
Press Freedom after Hazelwood,”
Journalism Quarterly
66
(spring
1989):
10.
Robert Trager and Donna Dickerson, “Prior Restraint in High
School: Law, Attitudes and Practice,“
Journalism Quarterly
57
(winter
11.
Jack Nelson,
Captive Voices: The Report ofthe Commission oflnquiy
into High School Journalism
(NY Schocken Books,
1974).
12.
Kate Fitzgerald, ”Illinois High School to Remove Paper Adviser
over Drug Story,”
Student Press Law Center,
March
17,2008,
http:/ /www.
splc.org
/
newsflash.asp?id=l713&year=
13.
Moriah Balingit, “Wash. Teacher Who Advised Independent
Publications Fired,“
Student Press Law Center,
November
20, 2007,
http:
/
/
www.splc.org
/
newsflash~archives.asp?id=165O&year=2007.
14.
Erica Huddock, ”Suspended Adviser Settles with District,”
Stu-
dent Press Law Center,
April
27, 2007,
http:/
/www.splc.org/newsflash-
archive~.asp?id=1512&year=2007
15.
Erica Huddock, “Two Ohio High School Newspapers now Subject
of Advisory Board Criticism,“
Student Press Law Center,
April
17, 2007,
http:
/
/
www.splc.org
/
newsflash~archives.asp?id=1506&year=2007
16.
Judy Wang, “Virginia High School
to
Revise Policy after
Controversial Articles Published,”
Student Press Law Center,
June
28,2007,
http:
/
/
www.splc.org
/
newsflash~archives.asp?id=1564&year=2007
169-73.
1980): 135-38.
17.
Noelle-Neumann,
The Spiral
of
Silence.
18.
William Gonzenbach, Cynthia King, and Patrick Jablonski,
”Homosexuals and the Military:
An
Analysis of the Spiral of Silence,”
The
Howard Journal
of
Communications
10
(October-December
1999): 281-96.
19.
Charles Salmon and Kurt Neuwirth, “Perceptions of Opinion
’Climates’ and Willingness to Discuss the Issue of Abortion,”
Journalism
Quarterly
67
(autumn
1990):
567-77.
20.
William Gonzenbach and Robert Stevenson, ”Children with AIDS
Attending Public School: An Analysis of the Spiral of Silence,”
Political
Communication and Persuasion
11
(spring
1994): 3-18.
21.
Michael Salwen, Carolyn Lin, and Frances Matera, ”Willingness to
Discuss ’Official English‘: A Test of Three Communities,”
Journalism
Quarterly
17
(summer
1994): 282-90.
22.
Patricia Moy, David Domke, and Keith Stamm, “The Spiral of
Silence and Public Opinion on Affirmative Action,”
Journalism
t3
Mass
Communication Quarterly
78
(spring
2001): 7-25.
23.
Vincent Filak and Thomas Price, ”The Value of Group Identity in
Preventing the Spiral of Silence,”
Visual Communication Quarterly
12
(spring/summer
2005): 46-57.
24.
Vincent Price, ”Social Identification and Public Opinion: Effects of
Communicating Group Conflict,”
Public Opinion Quarterly
53
(summer
1989): 197-224;
Vincent Price
and
Scott Allen, “Opinion Spirals, Silent and
Otherwise,”
Communication Research
17
(June
1990): 369-92.
25.
Carroll Glynn, Andrew Hayes,
and
James Shanahan, “Perceived
Support for One’s Opinion and a Willingness to Speak Out: A Meta-
Analysis of Survey Studies on the Spiral of Silence,”
Public Opinion
Quarterly
61
(fall
1997): 452-63.
EXPANDING
AND VALIDATING
APPLICATIONS
381
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
26.
Hayes, Glynn, and Shanahan, “Willingness to Self-Censor.”
27.
Hayes, Glynn, and Shanahan, ”Validating the Willingness to Self-
Censor.”
28.
Hayes, Scheufele, and Huge, “Nonparticipation as Self-Cen-
sorship
.”
29.
Hayes, Glynn, and Shanahan, “Validating the Willingness to Self-
Censor.”
30.
Andrew Hayes, Brian Uldall, and Carroll Glynn, ”Validating the
Willingness to Self-censor Scale 11: Opinion Expression in a Real
Conversational Setting,” unpublished manuscript,
2007.
31.
Nelson,
Captive Voices.
32.
John Bowen, “Captive Voices: What Progress, Change has
Occurred in
10
years?”
Quill
6.
Scroll
59
(February
-
March
1995): 14-16.
33.
www.jea.org
34.
Bowen, “Captive Voices”; Marcia Kovas, “The Impact
of
Hazel-
wood in the State of Indiana,“
Quill
8
Scrol64
(February-March
1991) 4-
8. 35.
Thomas Hill and Pawel Lewicki,
Statistics: Methods and Appli-
cations
(Tulsa,
OK
StarSoft,
2005).
36.
Vincent Filak and Adam Maksl, “An Examination of the Comfort
Levels
of
High School Principals and Newspaper Advisers in Regard to
Controversial Topics” (paper presented to the annual meeting of
AEJMC, Washington, DC,
2007).
37.
In almost one-third of the cases, individuals had incorrectly stat-
ed whether a state law protected them. For example, California has a
strong press law that is among the best in the nation in protecting high
school journalists. That said, eleven of our forty-four respondents from
California stated that no such law existed. Conversely, Indiana law
offers high school journalists no protection from censorship, but five of
the twenty-one respondents from the state noted that it
does.
We recod-
ed the data for all
the
participants, simply using state legal statutes and
the
participants’ declared location to better study
this.
In the end, we
found no significant correlation between an actual presence of the law
and any of the outcome variables. Unfortunately, we were unable to do
a similar examination for the publications policy variable, as we did not
ask participants what school district they belonged to. However, we
were concerned that our findings in regard to these variables might be
misinterpreted. To that end, we’ve renamed each of the protection vari-
ables ”perception of state protection” and “perception of district protec-
tion” to more accurately reflect the true nature of each variable.
38.
Hayes, Uldall, and Shanahan, ”Validating the Willingness to Self-
Censor Scale 11.”
39.
Luc Pelletier, Chantal %quin-L&esque, and Louise Legault,
“Pressure from above and Pressure from below as Determinants of
Teachers’ Motivation and Teaching Behaviors,”
Journal
of
Educational
Psychology
94
(February
2002) 186-96.
382
~OURNALISM
b
MASS
COMMUNICATION
OUARTERLY
at UNIV OF WISCONSIN OSHKOSH on October 7, 2016jmq.sagepub.comDownloaded from
... n tidak diuji sebagai pembolehubah yang boleh diukur (Lee et al., 2004). Oleh yang demikian, konsep lain yang berkaitan seperti penyaringan kendiri (self-censorship) telah dibangunkan sebagai pembolehubah untuk mengukur tahap seseorang individu menahan diri daripada mengutarakan pendapat kepada publik. Beberapa kajian (contoh : Hayes et. al., 2013;Filak et. al., 2009;Matthes et. al., 2012) menguji konsep penyaringan kendiri untuk secara tidak langsung melihat peranan ketakutan terhadap pengasingan dalam proses ekpresi pendapat awam. ...
... Terdapat kajian lepas yang menemui perhubungan yang signifikan antara topik tertentu (atau situasi tertentu) dengan tahap kelantangan individu dalam menyuarakan pendapat (cth: Kim, 1999). Namun, terdapat juga kajian yang menemui tiada kesan signifikan (cth: Lasorsa, 1991) dan antara justifikasi yang diberi adalah disebabkan pemilihan isu yang tidak sama dalam setiap kajian, dan oleh yang demikian mempunyai risiko pengasingan yang berbeza (Filak, Reinardy, & Maksl, 2009). ...
... Research on self-censorship, or self-limitation as we have called it here, has shown that it has both social and psychological causes (Filak et al. 2009;Hayes et al. 2005). This means that holding back one's opinions may be related to both personality traits, along with certain background characteristics such as gender, age or education, and to the context of conversation itself. ...
Article
Full-text available
The topic of this paper is the social and normative conditions for exercising the freedom of speech in Norway. Based on a representative survey of the population, the authors ask which types of utterances are seen as acceptable on different arenas in Norway, and whether certain groups are more likely to withhold their own opinion, out of fear of offending others or of being exposed to ridicule. The analysis shows that the Norwegian population has a balanced view of the freedom of speech as weighed against other social values, such as the value of protecting vulnerable groups and not offending or harming particular groups or religions. There is a high degree of consensus on the balance of values: a majority chooses the middle standpoints. Furthermore, the analysis shows that many would limit their expressions when presented with the risk of offending or hurting others, a process that we term “self-limitation”. Depending on the perspective on freedom of speech, such self-limitation will be interpreted differently. One important point however, is that self-limitation may lead to spirals of silence, where certain voices are less heard.
Article
This research advances communities of practice (CoPs) theory by documenting a role distinct to CoPs within student newsrooms. Using a survey ( N = 127) of college student media advisers, the data support a liminal role in which advisers are simultaneously inside and outside the CoP. This study also develops and verifies scales for each of the four pillars of a CoP in student newsrooms, that is, practice, community, identity, and meaning. Furthermore, these findings explore the impact of newsroom structure on the formation of CoPs and the differences among advisers regarding time spent in mentorship and assessment.
Article
Prior research into student journalism has indicated that a willingness to self-censor was a stronger predictor than other intrinsic concerns or external threats when it came to covering controversial topics. A recent examination of high-school students, however, found that gender served as both a predictive factor in a reticence to publish controversial content and a source of perceived overt censorship actions. Seeking to provide parsimony to this issue, this study examined online survey results from 142 high-school journalists (n = 142) in relation to controversial topics, self-censorship, and overt censorship. Findings indicate that topics pertaining to sexuality remain the most controversial and that gender directly correlated with reports of overt censorship. In addition, congruity between the gender of the student and adults in a supervisory role showed diminished levels of censorship and reports of a willingness to self-censor.
Article
This research analyzes news-making decisions and self-censorship among student journalists through the lens of gatekeeping theory. Using Q methodology, the study examines responses of 45 student journalists who were asked about variables that would influence their decisions not to pursue a controversial story. Four factors emerged based on analysis of three principal variables: the source of perceived threats, the nature of those threats, and the parties most likely to suffer if the threats come to fruition.
Article
Full-text available
Herkesçe kabul edildiği gibi İnternet, iletişimin geçmişe göre daha kolay gerçekleşmesine yardımcı olmaktadır. Dünyanın küçüldüğü algısını da beraberinde getiren bu görüş, bu küçük dünya içindeki etkileşimlerin daha hızlı ve etkin biçimde gerçekleşmesine yaramaktadır. Bu açıdan bakıldığında, sosyal medya kavramı da bireylerin etkileşim hızına katkı sağlayan bir olgu olarak dikkat çekmektedir. Ancak iletişimin başatlığına paralel olarak ilerleyen denetim kavramı da, İnternet ve sosyal medyanın gelişimine ayak uydurmayı başarmış; farklı yollarla İnternet'in sunduğu özgürlük alanının sınırlanması yolunda kimi adımlar atmıştır. Sansür kavramının karşılayabileceği sınırlamalar, bariz denebilecek biçimlerde İnternet ve sosyal medyanın yarattığı hareketlenmeler dâhilinde de görünür olmuşlardır. Sansürün doğal bir sonucu olarak görülebilecek otosansür de bireylerin sosyal medyadaki etkileşimleri üzerinde söz sahibi olma potansiyeli nedeniyle günümüz İnternet dünyası içinde önemli bir yer teşkil etmeye başlamıştır. Bu çalışma da Facebook örneği üzerinden sosyal medyadaki otosansür eğilimlerini ortaya koymayı amaçlamaktadır. Bu çalışma, gözetim ve suskunluk sarmalı bağlamında, çevrimiçi anket ve yarı yapılandırılmış görüşme aracılığıyla Facebook kullanıcılarının davranışlarını inceleyerek otosansür davranışlarını ortaya koymayı amaçlamaktadır.
Article
This study investigates the effect of news sites’ explicit online comment moderation policies on spiral of silence (SOS) outcome variables. It advances the methodological understanding of SOS research by comparing the traditional willingness-to-share variable to more direct measures of attitudes. The results of two studies show that a one-sided editorial and comments generally silenced the opposition, particularly when participants were asked to provide their own comment. This finding suggests that open-ended comments may best measure SOS effects. With the inclusion of explicit comment moderation, however, participants were less likely to agree with the editorial as evidenced by their closed-ended attitudes.
Article
Full-text available
The typical collective memories of societies involved in intractable conflicts play a major role in the eruption and continuation of the conflicts, while the positive transformation of these memories to being less self-serving promotes peacemaking. A major factor that inhibits such transformation is self-censorship. Self-censorship, practiced by members of a society’s formal institutions, inhibits the dissemination of alternative, more accurate narratives of the conflict that may change dominating biased conflict-supporting memories. Despite the importance of formal self-censorship in maintaining collective memories of conflicts, little empirical and theoretical research has examined this phenomenon. The present study addresses this omission. It examines the self-censorship practiced from 1949 to 2004 in three formal Israeli institutions (the National Information Center, the IDF/army, and the Ministry of Education) regarding the main historical event of the Israeli-Palestinian conflict: the causes of the 1948 Palestinian exodus. This is done by analyzing all these institutions’ publications produced throughout the 56-year research period and interviewing their key position holders. The results show that the institution gatekeepers practiced self-censorship for five reasons: garnering international support, mobilizing citizens, the impact of Zionist ideology, institutional norms, and fear of sanctions. The empirical findings are used to elicit theoretical insights, such as a definition for formal self-censorship, the difference between self-censorship practiced by gatekeepers (from formal and informal institutions) and that practiced by ordinary individuals, the five reasons for such self-censorship (distinguishing between two categories—intrinsic and extrinsic reasons), and the reasons that led the gatekeepers to admit that they had self-censored.
Article
Full-text available
This study surveyed respondents in three disparate communities regarding their willingness to express their opinions about “Official English.” The findings raised questions as to whether perceived media opinion influenced their willingness to “speak out.” People were more likely to express their opinions when they perceived that national opinion (public and media) corresponded with their own opinions.
Article
This study examines Elisabeth Noelle-Neumann's spiral of silence theory in terms of the controversial issue of whether homosexuals should be allowed to serve in the U.S. military. The study tests the theory's hypotheses in light of a content analysis of the media's treatment of the issue. The findings offer tentative and qualified support for Noelle-Neumann's theoretical propositions about the relationship between individuals' perceptions of the issue and the media's tenor of the issue. The study departs from previous research by examining the influence of the media at different stages in the spiral of silence model. Results suggest that the mass media may have separate but similar effects on perceptions of majority opinion and personal opinions. Results further reveal that, in certain situations, individuals who feel that their position is beginning to lose public support may feel compelled to voice their opinions.
Article
This study sought to more fully explicate the key variables involved in Noelle-Neumann's spiral of silence theory, which states that fear of isolation keeps individuals from expressing opinions perceived to be in the minority. We tested the theory in the context of public discussion about affirmative action policies, a domain seemingly ideal due to its moral and value-laden characteristics. Data from 217 randomly selected adults in October 1998 indicate that fear of isolation indeed prevents one from publicly voicing perceived minority opinions. Willingness to speak out on a controversial ballot initiative was predicted also by demographics, media use, and importance of the issue. However, it was perceived consonance of one's opinion with family and friends—rather than society at large—that predicted willingness to speak out.
Article
This study applies Elisabeth Noelle‐Neumann's spiral of silence theory to the controversial issue of whether children with AIDS should be allowed to attend public school. The study tests the theory's hypotheses in light of two content analyses of the media's treatment of the issue. The findings offer tentative and qualified support for Noelle‐Neumann's theoretical propositions about the relationship between individuals' perceptions of the issue and the media's tenor of the issue; particularly that of television. The findings suggest the need to address the role of other agents of influence, including reference groups, and to use time‐based data to unravel the causal order of the relationships.
Article
Public opinion formation is a social and communicative process, and individuals' opinions thus depend in many ways upon the social context surrounding public issues. Consequently this research investigates the interaction between people's social identities and mass media reports of public issues emphasizing conflicts of opinion between social groups. Current theory and research on social identification and intergroup behavior are used to develop a three-stage model of the cognitive, perceptual, and behavioral processes which may be triggered by media reports of group conflict. According to the model, a news report emphasizing group conflict over an issue (1) cues its recipients to think about the issue through their particular group perspective, which (2) leads to polarized or exaggerated perceptions of group opinions, and finally (3) leads to expressions of personal opinion consistent with these exaggerated perceptions of group norms. Factors contributing to and inhibiting this kind of response are also incorporated into the model. Analyses of experimental data from college undergraduates who read and responded to experimentally prepared newspaper articles covering a campus issue are reviewed. The results are found generally to support the social identification model and serve to illustrate how media reports emphasizing group conflicts may play an important role in the formation of public opinion.
Article
Finds that editors and advisors of high school newspapers agreed that only limited self-censorship was being practiced. Reports on the extent to which the advisors' journalism training and advising experience related to the amount of self-censorship as well as censorship reported. (SR)
Article
Reports on a survey of 474 Illinois, Indiana, and Wisconsin public high schools, designed to determine practices and attitudes regarding student publications. Notes that the results indicate that there is no consistent approach to high school journalists and no consistent attitude toward them. (GT)
Article
This 1986 survey of 432 Madison, Wisconsin, residents describes an elaborated version of Elisabeth Noelle-Neumann's model of the “spiral of silence.” Consistent with her hypotheses, persons whose opinions were congruent with those of the national majority were more willing to speak to a stranger than were those whose opinions were shared only by a minority; however, the same result was not obtained for either a different mode of expression or system level. Factors such as involvement and knowledge were found to directly influence opinion expression, whereas education and gender were found to be indirectly related to expression.