Predictors of channel exposure...
Predictors of Channel Exposure and
of Topic-Specific Attention
to Messages about Risk
Nandita Dhume and Sharon Dunwoody
School of Journalism and Mass Communication
University of Wisconsin-Madison
821 University Avenue,
Madison, WI 53706
[log in to unmask]
Marty Kanarek
Department of Preventive Medicine and
Institute for Environmental Studies
University of Wisconsin-Madison
502 N. Walnut St.
Madison, WI 53706
and
Kenneth Bro
Wisconsin Division of Health
144 E. Washington Ave.,
Madison, WI 53706
Presented at a SCIgroup session, Association for Education in Journalism
and Mass Communication, Washington, DC, August 1995.
This work was funded by the University of Wisconsin Sea Grant Institute
under grants from the National Sea Grant College Program, National
Oceanic and Atmospheric Administration, U.S. Department of Commerce and
from the State of Wisconsin.
NA90AA-D-SG469, R/PS-42
Predictors of Channel Exposure and of Topic-Specific Attention to
Messages about Risk
Nandita Dhume, Sharon Dunwoody, Marty Kanarek and Kenneth Bro
University of Wisconsin-Madison and Wisconsin Division of Health
Madison, WI
Abstract
Risk communicators need to make more sophisticated choices of channels
in which to provide messages. In order to do so, it is imperative to
understand the factors that lead to channel selection by individuals
at
risk. Researchers argue that judgments of channel cost and message
relevance both should influence channel preference.
This was tested by surveying Wisconsin anglers (N=333) regarding their
use of information channels particularly in the context of risks
posed
by eating sport-caught fish. This study tests channel cost, channel
utility, as well as two dimensions of risk judgment and issue salience
as predictors of channel exposure and attention to risk messages
within
those channels.
As hypothesized, channel utility and one dimension of issue salience
(level of worry) predicted to attention. Other findings suggest that
type of channel, type of risk, and the relationship between dimensions
of issue salience play an important role in predicting channel use.
Predictors of Channel Exposure and of
Topic-specific Attention to Messages about Risk
Risk communication scholars, among others, often examine information
channel as an independent variable in an attempt to understand
the
effects, if any, that channel selection may have on receivers. This
study looks at a prior stage in the communication process,
examining
possible predictors of channel use. Unlike the vast majority of
risk
communication studies, here channel use will be treated as a
dependent
variable.
The purpose of the former type of study (where channel use is an
independent variable) typically is to determine the degree to which
channel use is a predictor of behavior change, particularly
risk-taking
behavior. Information campaigners, risk communicators, and
other
scientists interested in modifying or predicting behavior are often
interested in the effects of factors such as the amount of
exposure to
channels of information and the amount of attention paid to
information
within such channels.
Such research is sender-oriented and focuses on the communication
process after an individual has encountered a message. For example,
channel use has been studied not only as a predictor of
risk-taking
behavior (Gantz et al., 1990) and of beliefs about science and
technology (Elliott & Rosenberg, 1987) but of political behavior (Feld
man & Kawakami, 1991) and of levels of knowledge (Culbertson &
Stempel,
1986).
These studies operate from the perspective of the communicator. Their
assumption is that channel use will have some sort of impact on
the
behavior of individuals. Risk scholars employing this approach are
attempting to assess the extent to which particular channels of
information will change or moderate risk-taking behavior. Little or no
effort is spent on discovering the channels that individuals
are
predisposed to selecting.
The process of communication, however, is not unidirectional, from
sender to receiver. Rather, individuals may choose from a variety
of
media channels to suit different situations. Uses and
gratifications
research (Katz et al., 1974) for example, conceives of
individuals as
actively using channels of communication to satisfy the needs of
their
current social situations. Factors that are associated with
channel
selection and use need to be explored in order to better explain
and
predict communication processes.
Some risk communication scholars (for example, Krimsky & Plough, 1988)
suggest that, rather than embedding messages in predetermined
channels,
it would be more profitable to first determine what channels
best suit
the information needs of individuals. These researchers
subscribe to the
belief that individuals are rational beings who will use
channels that
meet their needs. Factors that predict to those channel
selections,
thus, need to be determined to fully explore the communication
process.
This paper examines variables that could potentially impact channel
use, our dependent variable. The study first makes a distinction
between
exposure to a channel and attention to specific content within
a channel
as two distinct measures of channel use. Then, patterns in
channel
exposure and the amount of attention paid to particular types of
stories
by Wisconsin anglers are studied within the context of
information about
the health risks associated with eating sport-caught fish.
Channel use: Exposure vs. Attention
Communicators commonly select exposure to a channel as the primary
measure of channel use (for example, Gerbner et al., 1984).
However,
other researchers have found that attention to specific types of
information in a channel rather than mere channel exposure is a much
stronger predictor of channel effects (for example, McLeod &
McDonald,
1985). Chaffee & Schleuder (1986), for example, found that
attention to
media news in specific channels was a significant predictor of
knowledge
about public affairs and politics even after controlling for
channel
exposure.
Such work suggests that exposure over time to any single channel is not
a sufficient measure of channel use if one is interested in
information
effects with respect to a particular issue or content area.
Rather,
attention to particular content within channels should be a
stronger
measure of channel use. This study will explore factors that are
associated with both channel exposure and levels of attention to
specific content within specific channels.
Predictors of Channel Use
Individuals select channels based on a combination of factors. Some of
these factors may be receiver-oriented, others may be related
to the
issue or topic of interest, and still others may be related to
perceived
channel characteristics.
Receiver attributes: Many researchers have sought audience attributes
that influence channel selection. Typical of such work is a
study of
senior citizens by Goodman (1992) which found that education,
age,
income, and living arrangements were associated with exposure to
chan
nels. Other audience attributes uncovered by researchers include
social
and psychological factors such as cosmopolitan lifestyles and
need for
activation (Donohew et al., 1987).
We will employ a variety of individual characteristics in this study
but will use them primarily as control variables in our
regression
equations. Our focus will be on variables discussed in the next two
categories.
Perceived issue/topic attributes: Message characteristics such as issue
salience are likely to have an impact on channel use. Issue
salience in
the context of risk can be characterized as the extent to which
an
individual finds a risk personally relevant in the sense that it
poses a
hazard to him or herself. That is, the greater the individual's
assessment of a risk, the more attention he or she should pay to
risk-related messages. Here, it is important to note the distinction
between individuals' personal (or perceived) judgment of risk and
the
calculated statistical probability of that risk taking place.
For the purposes of this study the term "personal risk judgment"
denotes an individuals assessment of his or her level of risk.
Whether
or not this assessment is accurate (in keeping with scientific
risk
estimates) is not discussed in this paper as accuracy does not
have a
bearing on our research questions that is, we do not hypothesize
that
accurate risk judgments will impact channel use.
Further, the concept of risk judgment includes not only individuals'
evaluations of their personal likelihood of coming to harm but
the
additional dimension of worry generated by the hazard. Dunwoody and
Neuwirth (1991) have characterized risk judgment in terms of these
two
conceptually distinct dimensions -- cognition and affect. The
cognitive
dimension is the individual's perception of his or her
likelihood of
facing adverse consequences of a risk action (for example, the
likelihood of getting sick from eating contaminated fish). The
affective
dimension deals with the level of worry generated by the risk.
It is likely that these dimensions tap into the concept of issue
salience; that is, the higher one rates one's own chances of getting
sick and/or the more worried one is about getting sick
personally, the
more one should consider the issue of fish contamination (in
this case)
to be important. The impact of issue salience on channel use
will be
examined in this study via measures of cognitive and affective
risk
judgments.
Perceived channel attributes: Rubin (1993) argues that apart from
social and psychological factors, attitudes about a medium or
channel
and its content, as well as media orientations, play a role in
channel
use. Individuals' attitudes about channels can be studied in a
number of
ways. Johnson & Meischke (1992), for example, asked a sample of
women to
evaluate channels from which they had received cancer-related
information in terms of three channel dimensions: editorial tone
(credibility), communication potential (presentation and style), and
utility. Similarly, Marin and Marin (1990) analyzed the perceived
credibility of channels and of sources of AIDS information within a
sample of Hispanics.
Sitkin et al. (1992) have offered a model that is based on how
individuals in organizations select channels. They find associations
between perceptions of what they call data-carrying capacity (the
degree
to which a channel is able to efficiently convey task-relevant
data),
symbol carrying capacity (the degree to which a channel is able
to
transfer symbolic meaning), and media use.
Chaffee (1986) has pointed to two concepts that he thinks are important
predictors of channel exposure. It is these two concepts that
will be
central to our analysis. According to Chaffee, the "sources one
consults
for information ... are determined mainly by (a) their
accessibility and
(b) the likelihood that they will contain the information one
might be
seeking" (p. 64).
Chaffee characterizes channel accessibility in terms of both the
frequency with which a source attempts to communicate with a
receiver
and the physical and psychological ease with which the receiver
can use
the channel. Since we are studying receivers here, not sources,
accessibility for the purposes of selective exposure will be measured
in
terms of the latter characteristics in this study. This means
that
perceived channel accessibility or channel access cost is
determined by
perceived cost: time, and energy -- in other words, the effort
an
individual feels that he or she must expend in order to use that
particular channel.
The second factor that Chaffee articulates is perceived channel
relevance or utility, which refers to the perceived likelihood that a
channel will contain information that is useful. This study
defines
channel utility not only by the amount of information available
through
a channel but the perceived accuracy of that information.
Although Chaffee proposed these two predictors of channel use more than
a decade ago, little research has attempted to explore their
actual
relationship to channel choices. In a recent convention paper,
Neuwirth
and Dunwoody (1994) tested the argument by asking a sample of
young
adults about their use of information channels to learn about the
risk
of AIDS. They found that cost of channel access was a
significant
predictor of channel exposure, while respondents' estimates of the
utility of AIDS information in particular channels served as a
predictor
of topic-specific exposure. Thus, although both concepts did
indeed
predict to channel use, they each predicted to a different
dimension of
such use.
Hypotheses and Research Questions
This paper examines the extent to which channel accessibility (cost)
and channel utility (relevance) will predict to channel exposure
and to
attention to a specific type of message within those channels.
Neuwirth
and Dunwoody's (1994) work suggests that measures of utility
and
relevance may predict to different dimensions of channel use,
specifically that cost may be more closely related to exposure but that
judgments of relevance, in contrast, may be more closely
related to
attention measures. We test these notions in the following two
hypotheses:
Hypothesis 1: Higher levels of perceived channel cost will be
associated with lower levels of channel exposure.
Hypothesis 2: Higher levels of perceived channel utility will be
associated with higher levels of attention.
Issue salience, on the other hand, should be entirely a predictor of
attention rather than exposure. We, thus, hypothesize:
Hypothesis 3: Higher levels of personal risk likelihood will be
associated with higher levels of topic-specific attention.
Hypothesis 4: Higher levels of personal worry will be associated with
higher levels of topic-specific attention.
Finally, since cognitive and affective dimensions of risk judgment have
been shown to be conceptually distinct (Dunwoody, Dhume, Bro
and
Kanarek, 1995; Dunwoody & Neuwirth, 1991), we ask the following
research
question:
Which dimension of issue salience (personal risk likelihood or personal
worry) is a better predictor of topic-specific attention?
Hypotheses were examined for four channels of communication --
newspapers, television, radio, and fishing magazines -- using data
collected from a sample of Wisconsin anglers. Independent variables
included not only the four central concepts of channel cost,
channel
utility, personal risk likelihood, and personal worry but also
demographic features such as individuals' age, gender, level of
education, income, and race.
The Risk of Eating Sport-Caught Fish in Wisconsin
Contamination of sport fish in the Great Lakes and other bodies of
water in Wisconsin presents health threats that are sometimes
difficult
to measure. Nonetheless, scientists are convinced that the
threats are
real, and states in the Great Lakes basin (Illinois, Indiana,
Michigan,
Minnesota, New York, Ohio, Pennsylvania and Wisconsin) since
the early
1970s have issued annual consumption advisories that recommend
fish to
avoid as well as ways to prepare fish for eating that minimize
exposure
to chemicals. Wisconsin tests sport fish for polychlorinated
biphenyls
(PCBs), DDT, toxaphene, chlordane, dieldrin, mercury and
dioxin.
Among that cocktail of chemicals, PCBs and mercury have been singled
out for the greatest attention by state agencies. According to
the
state's advisory (Health Guide, April 1994), PCBs have been linked
to
developmental and growth problems in infants born to women who
regularly
eat contaminated fish, and long-term consumption is also
suspected of
causing cancer. The chemical, long used by industry and now
buried in
the sediments of the Great Lakes, works its way up the food
chain and
accumulates in the fat of fish. Bigger fish bring with them
potentially
higher contaminant loads.
Mercury is more of a problem in inland lakes and waterways. Naturally
present in many water bodies and introduced by industrial
emissions to
air and surface water in others, it is stored throughout the
body of a
fish, particularly in muscle tissue. Mercury is toxic and
ingesting
large amounts can harm the central nervous system and may affect
body
movement and senses of touch, taste and sight. Unlike PCBs,
mercury can
be excreted from the system. But fish excrete mercury at a very
slow
rate, so the highest levels of the contaminant are still found in
large,
old fish.
Methods
The sample in this study is comprised of anglers in the state of
Wisconsin. Through the systematic random sampling of fishing
licenses
from seven Wisconsin counties, 786 names and addresses of
anglers were
collected. These anglers were sent a mail questionnaire.
Ultimately, 198
were returned by the post office as undeliverable. Of the
remaining 588
anglers, an initial and two follow up mailings elicited a total
of 333
responses, a response rate of 57%.
Dependent variables: The eight dependent variables in this study were
both exposure and attention to each of the following channels:
(a)
newspapers; (b) television news; (c) radio news; (d) fishing
magazines.
Channel exposure was measured by evaluating responses to the question
"Over the past seven days, how many days did you read a
newspaper
(television/ radio)?" This was immediately followed by the attention
question: "If you were to encounter a story about fish
contamination in
your newspaper, how much attention would you pay to it?"
Responses to
this latter question ranged from a lot of attention ( scored 3)
to no
attention (0).
Exposure to fishing magazines was determined by counting the number of
responses to the question "Do you or someone in your household
subscribe
to or regularly read any fishing publications (e.g. In
Fisherman, Woods
and Waters)? If yes, please list them." Topic-specific
attention to
fishing magazines was measured in the same way as it was for the
other
three channels; that is, respondents were asked to indicate the
degree
of their attention to potential articles regarding fish
contamination.
The mean, median, and standard deviation for these dependent variables
can be found in Table 1.
Independent variables: The primary independent variables for this study
were channel cost, channel utility, cognitive risk judgment,
and
affective risk judgment. These were measured as follows.
1. Channel cost: Respondents were asked to rate the effort (in terms of
cost, time, and energy) of using the four channels listed
above; that
is, they were asked whether a accessing a channel would require
high
effort (effort score=3), some effort( 2), little effort (1) or no
effort
(0). Table 2 provides descriptive statistics for channel cost
for these
four channels and Table 6 provides measures for the same
variable but
for other channels.
2. Channel utility: Respondents were asked to evaluate the usefulness
of fish contamination information that they would find in each
channel.
Channels were rated as very useful (utility score = 2),
somewhat useful
(1) or as not useful (0). Usefulness was defined as a
combination of
amount of detailed information and accuracy of the channel.
Descriptive
statistics for channel utility are presented in Table 3.
3. Cognitive risk judgment: Respondents responded to the following
question on a scale of 0 to 100: "How likely is it that you
personally
will become ill from eating fish caught in Wisconsin waters or
the Great
Lakes? (0 = absolutely no chance of getting sick; 100 = certain
to get
sick).
4. Affective risk judgment: Respondents were again asked to respond on
a scale of 0 to 100: "How worried are you personally about
becoming sick
from eating fish caught in Wisconsin waters or the Great Lakes?
(0 = not
at all worried; 100 = the most worried you could ever be).
Table 4
provides summary statistics for both cognitive and affective
dimensions
of risk.
5. Demographics: Demographic variables in this study were gender, year
of birth (age), level of education, income, and race.
Results
The hypotheses were tested by conducting regression analyses.
Significant regression variables and their beta values are available in
Table 5. The findings for each hypothesis are discussed here,
individually.
Hypothesis 1: Higher levels of perceived channel cost will be
associated with lower levels of channel exposure.
This hypothesis is not supported. Channel cost was not associated with
exposure to any of the four channels of communication. It was,
however,
positively associated with topic-specific attention to
newspapers and
television. In other words, the more effort individuals felt
they needed
to access newspapers and television, the more likely they were
to pay
attention to stories regarding fish contamination within those
media.
Hypothesis 2: Higher levels of perceived channel utility will be
associated with higher levels of channel attention.
This hypothesis is supported. In addition, the data suggest an even
larger pattern. Channel utility, unlike channel cost, was
significantly
associated with exposure as well as with attention for all four
channels. The more useful individuals considered information in a
channel, the more likely they were to use that channel in terms of
both
exposure and attention.
Hypothesis 3: Higher levels of personal risk likelihood will be
associated with higher levels of topic-specific attention.
Contrary to our expectations, personal risk likelihood was
significantly associated with only one of the dependent variables --
exposure to newspapers. The negative beta indicates that greater
use of
newspapers is associated with lower estimates of coming to harm
from
eating sport-caught fish. This hypothesis, thus, is not
supported.
Hypothesis 4: Higher levels of personal worry will be associated with
higher levels of topic-specific attention.
As hypothesized, personal worry was positively associated with
attention for all four channels of communication. The more worried an
individual was about becoming sick from eating sport-caught
fish, the
more likely he or she was to pay attention to stories or
articles about
fish contamination in any of the four channels.
Demographics: Demographic independent variables that were included in
the analyses were gender, age, level of education, and income.
With the
exception of age, there is little in the way of patterns to
analyze.
y Gender was significantly related only to newspaper exposure and
attention with males using the newspaper more than females and
paying
more attention to newspaper stories about contaminated fish than
females.
y Age, not surprisingly, was positively related to three of the four
exposure variables. The older an individual, the more likely he
or she
was to be exposed to newspapers and television; and the more
likely he
or she was to attend to stories about fish contamination in
newspapers.
On the other hand, the younger the angler, the more likely he or she
was to be exposed to fishing magazines.
y Level of education was significantly related to only one of the
dependent variables -- exposure to newspapers. The higher the level
of
education, the more likely the angler was to be exposed to
newspapers.
y Income was positively related to exposure to radio news and public
affairs programming as well as to the number of fishing
publications
that were subscribed to.
y Race was negatively associated with exposure to radio news and public
affairs suggesting that Caucasian anglers were more likely to
use this
channel of information.
Discussion
Three patterns emerge on examination of the results. First, our
exploration of the extent to which channel cost and utility predicted
to
anglers' use of channels for information about contamination in
sport-caught fish supported the value of one dimension -- utility --
but
not the other. In this study, perceived cost of a channel
played no role
whatsoever in anglers' tendency to use a channel for fish
contamination
information. Rather, the major predictor of channel exposure
and
attention was the perception of the relevance of information
available
in a particular channel.
A possible reason for the failure of channel cost to predict to channel
use could be that the channels examined in this study are not
cosnsidered to be costly. Cost measures were collected for a wide array
of channels and these indicate that there is sufficient
variance in
measures of channel cost (see Table 6). However, as exposure and
attention measures were available only for four channels (newspapers,
television, radio and fishing magazines), data were not amenable
to
including the other channels of information. Evidently, these four
channels are not perceived as costly and the lack of variance makes
it
difficult to see the extent to which channel cost would be a
predictor
of channel use. We suspect, however, that channel cost would
predict to
channel exposure if more "costly" sources of information such
as
physicians, government officials, and written materials from
environmental agencies were included in the study.
Another pattern suggests that in the case of information regarding fish
contamination, anglers tend not to discriminate between
exposure and
attention with respect to channel utility. As hypothesized,
channel
utility predicted to topic-specific attention. However, channel
utility
also predicted to channel exposure. Although the latter finding
was not
predicted, it is not implausible. Information about
contamination in
fish while available through many different channels is not
plentiful.
In addition, risks incurred from eating sport-caught fish are
low with
little issue salience for anglers.
It is possible that the impact of channel utility on topic-specific
attention as well as on channel exposure stems from a blurring of
distinctions between the two. In the case of a low-level,
low-salience
risk with limited available information, anglers are not
actively
seeking fish contamination information. Rather, they may access it
when
they encounter it during regular channel use. Thus, there may
be little
difference between exposure and attention. Type of issue
should,
therefore, be a variable rather than a constant when predicting to
channel use.
A third finding was that cognitive risk estimates fail to predict to
channel use. We found support for our hypothesis that worry
estimates
predict to levels of topic-specific attention. However, our
other
measure of issue salience -- likelihood estimate -- did not have an
impact on level of attention.
It may be that likelihood estimates predict to worry estimates, a
theory that is supported in the findings of Griffin et al.(1994).
In a
recent convention paper, these researchers found that personal
risk
likelihood predicts quite strongly to worry about a risk. This
indicates
that cognitive judgments of risk may act indirectly, not
directly, on
channel use.
In summary, it is likely that when cost is not a factor, judgment of
channel utility will be the most important predictor of channel
use.
Channel cost should not be dismissed as a predictor, however. The
study
looked at a wide array of channels and found much variance in
perceived
cost. A comparison of Tables 2 and 6 illustrates this point.
Future
research needs to take both cost and utility into account over a
variety
of information channels.
In addition, risk scholars should consider type of risk as a variable
rather than as a constant. Low-level risks particularly those
with
clearly evident benefits should be compared with high-level risks
that
have no apparent benefits. Furthermore, the relationship
between
different dimensions of risk perception should be evaluated so as to
achieve a better understanding of channel use.
Rather than focusing solely on the effects of channels chosen by
information providers, it is imperative that risk communication
scholars
understand the process by which individuals select channels to
inform
themselves. Risk communicators need to be aware of factors that
lead to
channel use in order to provide information to individuals at
risk as, a
better understanding of the predictors of channel use will
foster
communication that is consistent with the channel exposure and
attention
of those at risk.
References
Chaffee, S.H. (1986). Mass media and interpersonal channels:
competitive, convergent or complementary? In G. Gumpert & R. Cathecart
(eds.), Inter/Media: Interpersonal communication in a media
world (3rd
edition). New York, NY: Oxford University Press.
Chaffee, S.H. and J. Schleuder (1986). Measurement and effects of
attention to news media. Human Communication Research, 13, 76-107.
Covello, V.T. (1987). Informing people about risks from chemicals,
radiation, and other toxic substances: A review of obstacles to
public
understanding and effective risk communication. In W. Leiss
(ed.),
Prospects and problems in risk communication, Waterloo: University
of
Waterloo Press.
Culbertson, H.M. and G.H. Stempel III (1986). How media use and reliance
affect knowledge level. Communication Research, 13(4), 579-602.
Donohew, L., P. Palmgreen, and J.D. Rayburn (1987). Social and
psychological origins of media use: A lifestyle analysis. Journal of
Broadcasting and Elaectronic Media, 31(3), 255-278.
Dunwoody, S. and K. Neuwirth (1991). Coming to terms with the impact of
communication on scientific and technological risk judgments.
In L.
Wilkins and P. Patterson (eds.), Risky Business, New York, NY:
Greenwood
Press.
Elliott, W.R. and W.L. Rosenberg (1987). Media exposure and beliefs
about science and technology. Communication Research, 14(2),
164-188.
Feldman, O. and K. Kawakami (1991). Media use as predictors of political
behavior: The case of Japan. Political Psychology, 12(1),
65-80.
Fischhoff, B., S. Lichtenstein, P. Slovic, S. Derby, and R. Keeney
(1981). Acceptable risk. New York, NY: Cambridge University Press.
Gantz, W., M. Fitzmaurice, and E. Yoo (1990). Seat belt campaigns and
buckling up: Do the media make a difference? Health
Communication, 2(1),
1-12.
Gerbner, G., L. Gross, M. Morgan, and N. Signorielli (1984). Political
correlates of television viewing. Public Opinion Quarterly, 48,
283-300.
Goodman, R.I. (1992). The selection of communication channels by the
elderly to obtain information. Educational Gerontology, 18(7),
701-714.
Griffin, R.J., S. Dunwoody, F. Zabala and M. Kamerick (1994). Public
reliance on risk communication channels in the wake of a
cryptosporidium
outbreak. Paper presented to the Society of Risk Analysis
annual
convention, Baltimore.
Health Guide for People Who Eat Sport Fish from Wisconsin Waters (April
1994). Madison, WI: Wisconsin Department of Natural Resources.
Johnson, J.D. and H. Meischke (1992). Differences in evaluations of
communication channels for cancer-related information. Journal of
Behavioral Medicine, 15(5), 429-445.
Katz, E., J.G. Blumler, and M. Gurevitch (1974). Utilization of mass
communication by the individual. In J.G. Blumler and E. Katz
(eds.), The
uses of mass communications: Current perspectives on
gratifications
research. Beverly Hills, CA: Sage Publications.
Marin, G. and B.V. Marin (1990). Perceived credibility of channels and
sources of AIDS information among Hispanics. AIDS Education and
Prevention, 2(2), 154-161.
McLeod, J. and D. McDonald (1985). Beyond simple exposure: Media
orientations and their impact on political processes. Communication
Research, 12, 3-34.
Neuwirth, K. and Dunwoody, S. (1994). Channel access cost and perceived
utility as predictors of exposure and attention to HIV
information.
Paper presented to the Midwest Association for Public Opinion
Research
annual conference, Chicago.
Rubin, A.M. (1993). Audience activity and media use. Communication
Monographs, 60(1), 98-105.
Sitkin, S.B., K.M. Sutcliffe, and J.R. Barrios-Chaplin (1992). A
dual-capacity model of communication media choice in organizations.
Human Communication Research, 18(4), 563-598.
Slovic, P. (1987). Perception of risk. Science, 36, 280-285.
Table 1
Dependent Variables: Descriptive Statistics
Channel use:
Mean
Median
Standard Deviation
Newspaper Exposure
5.01
6.0
2.36
Newspaper Attention
2.26
2.0
0.79
Television Exposure
4.81
5.0
2.2
Television Attention
2.37
3.0
0.74
Radio Exposure
4.56
5.0
2.59
Radio Attention
2.18
2.0
0.84
Fishing mag Exposure
0.7
0.0
1.03
Fishing mag Attention
2.28
2.0
0.83
Newspaper, television, and radio attention were measured in terms of
number of days. Fishing magazine exposure was measured in terms of
number of fishing publications subscribed to.
Attention to all four channels was measured a lot (3); some (2); a
little (1); no attention (0)
Table 2
Channel Costs
Newspapers
Television
Radio
Fish mags
No effort
24.8%
42.9%
46.0%
19.1%
Little efrt
46.5%
39.0%
35.9%
36.9%
Some effort
22.3%
11.7%
12.0%
37.2%
High effort
6.4%
6.4%
6.1%
6.8%
Mean
1.1
0.82
0.78
1.31
Median
0.05
1.0
1.0
1.0
S. Dev.
1.0
0.88
0.88
0.86
No effort=0; Little effort=1; Some effort=2; High effort=3
Percentages denote the valid percent of the sample that indicated that
effort score for that channel.
Table 3
Channel utility
Newspapers
Television
Radio
Fish mags
Not useful
5.2%
9.0%
13.0%
8.8%
Smwht usefl
49.0%
50.8%
60.1%
61.8%
Very useful
45.8%
40.2%
26.9%
29.4%
Mean
1.41
1.31
1.14
1.2
Median
1.0
1.0
1.0
1.0
St. Dev.
0.59
0.62
0.62
0.58
Not useful=0; Somewhat useful=1; Very useful=2
Percentages denote the valid percent of the sample that indicated that
utility score for that channel.
Table 4
Cognitive and affective risk judgments
Cognitive risk judgment
Affective risk judgment
Mean
14.31
34.81
Median
5.0
25.0
Standard Deviation
19.94
32.77
Responses could range from 0 (no likelihood/not worried) to 100
(absolutely likely/as worried as I've ever been).
Table 5
Prediction of channel use (primary independent variables)
(Significant betas)
Newspaper exposure
Newspaper attention
Television exposure
Television attention
Radio exposure
Radio attention
Fish mag exposure
Fish mag attention
Channel cost
.132*
.124*
Channel util
.223****
.181**
.147*
.178**
.154*
.223**
.222***
.284***
Cogntve risk
-.183*
Affctve risk
.22***
.226***
.225***
.269***
Gender (m=1)
.114*
.119*
Year born
-.343****
-.17**
-.26****
.146*
Education
.141*
Income
.172**
.155*
Race
-.159*
R2
.233
.132
.084
.105
.064
.113
.067
.173
Betas are significant at
* = <.05
** = <.01
*** = <.001
**** = <.0001
Table 6
Costs for channels other than newspaper, television, radio and fishing
magazines
Local govt. offcials
State govt. officials
University researchers
Physician
Govt. publications
No effort
16.8%
18.6%
22.2%
24.9%
17.2%
Little effrt
14.3%
10.2%
6.9%
15.3%
23.5%
Some effort
29.0%
22.0%
21.6%
32.1%
38.9%
High effort
39.9%
49.1%
49.4%
27.7%
20.4%
Mean
1.92
2.02
1.98
1.63
1.62
Median
2.0
2.0
2.0
2.0
2.0
St. Dev.
1.1
1.16
1.21
1.14
1.0
|
