This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line,
send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Feb 2006)
Thank you.
Elliott Parker
====================================================================
An empirical investigation of the relationships
among fear and efficacy of breast cancer, media use, and knowledge
about breast cancer prevention
in Caucasian and African American women
Kenneth Fleming
Cynthia Frisby
University of Missouri-Columbia
Correspondence to the first author:
Kenneth Fleming
School of Journalism
University of Missouri-Columbia
135 Neff Annex
Columbia, MO 65211
Tel: (573) 882-3396
Fax: (573) 882-2890
Email: [log in to unmask]
Research paper submitted to the topic of "The Operation, Function and
Impact of Ethnic and Minority Media on Society" of the Mass
Communication & Society Division of the Association for Education in
Journalism & Mass Communication Convention
in San Antonio, Texas in Aug. 10-13, 2005
An empirical investigation of the relationships among fear and
efficacy of breast cancer, media use, and knowledge about breast
cancer prevention
in Caucasian and African American women
Abstract
This study examines the relationships among attitudes toward breast
cancer, knowledge about the disease, religious beliefs, and use of
various news media channels in Caucasian (n=240) and African American
(n=206) women randomly selected in eleven metropolitan areas in the
U.S. Results show that magazines were negatively related to fear of
breast cancer, and radio was positively related to efficacy of the
disease for African American women. Use of the Internet was a
predictor of efficacy for Caucasian women. Radio significantly
contributed to knowledge about breast cancer for both groups of
women. Religious beliefs and knowledge about breast cancer were also
predictors of fear and efficacy of the disease, but with opposing
effects and to different women.
An empirical investigation of the relationships among fear and
efficacy of breast cancer, media use, and knowledge about breast
cancer prevention
in Caucasian and African American women
African American women are more likely than Caucasian women to be
diagnosed at later stages of breast cancer and are more likely to die
from the disease (e.g., NCI, 2004; Hoffman-Goetz, 1999). Perhaps one
of the reasons lies with the differential learning patterns of the
two groups of women. Previous research has shown that individuals
with high education levels tend to learn information from news media
sources at a faster rate than those with low education levels (e.g.,
Rosenstock et al, 1988; Frisby, 2002; Gantz et al., 2004). The
present study, then, tests whether knowledge about breast cancer,
along with use of various news media sources and religious beliefs,
would influence two important but opposing attitudes toward the
disease -- fear and efficacy of breast cancer -- in both Caucasian
and African American women.
Knowledge about the disease and its risks was poor among African
American women who did not want or desire mammography, and those who
expressed desire for a mammogram perceived themselves as more
susceptible to breast cancer and considered the disease severer than
those who did not want a mammogram (Price et al., 1992). Other p
revious research shows that African Americans generally display a
greater tendency to delay or avoid contact with health providers
(AMA, 2004; Ford & Cooper, 1995; Howard, Penchansky, &Brown, 1998).
Meanwhile, they rely on cultural and religious beliefs in coping with
breast cancer and other diseases (e.g., Royak-Schaler et al., 1995;
Bobo, 1995; Cardwell & Collier, 1981). Thus, steps to increase
participation in breast cancer screening programs may require
increasing public efforts that not only effectively disseminate
information but also help increase women's knowledge about the
disease with adequate consideration of the factors that women feel
important in their breast cancer prevention.
The research reported here is part of a comprehensive research in
health communication investigating the relationships between
Caucasian and African American women's perceptions of breast cancer,
their knowledge about the disease, and cognitive involvement with
messages about breast cancer prevention from various sources of news
media. Using hierarchical regression analyses, the present study
seeks to build on the relationship between breast cancer and mass
communication by examining and analyzing the differences in knowledge
and attitudes toward the disease as a function of use of news media
between two important groups of women. Although it has been suggested
that media messages can be effective for behavioral changes such as
early breast cancer detection specifically in African American women
(e.g., Ansell et al., 1988), we felt it necessary to examine the
effects of news media on women's attitudes toward the disease since
attitudes are closely related to knowledge and religious beliefs that
more or less determine one's behavior. In addition, we anticipated
that that the present study would help identify the media sources
that could be used to effectively reach women with information about
risks related to breast cancer and its prevention.
Literature Review
The role of mass media in public health communication
Most American adults closely follow health news in the media (e.g.,
Gantz & Wang, 2004). According to a Kaiser Health Poll in July/August
2004, 42% of the American public closely followed stories about the
release of new cholesterol guidelines at the time when stories about
military action and peacekeeping in Iraq (87%) and about the 2004
U.S. presidential election campaigns (77%) were dominant in the media
(Kaiser Family Foundation, 2004). In another Kaiser Health Poll in
September/October 2003, knowledge questions regarding public health
were answered correctly on average by 56% of those who reported
following the news story closely versus 40% of those who reported not
following it closely (Brodie et al., 2003).
In addition to delivering important health news, news media broadcast
important public service announcements (PSA), alarming reports about
toxic substances, incidence of diseases, and report critical medical
breakthroughs, availability and quality of professional healthcare
organizations, and etc. Further, mass media disseminate health
information through health campaigns designed specifically to educate
and informant. In raising public awareness of an illness, news media
play an important role because it is through various media outlets
that issues health and illness are placed at the forefront of public
consciousness (Brashers et al., 2002). The media are also thought to
be influential in shaping public attitudes toward health-related
issues such as risk factors for cancer (Hoffman-Goetz, 1999).
Magazines, for example, were ranked as the third leading source of
cancer information for women in the U.S. (Meissner et al., 1992), and
more women rated magazines as more important sources of cancer
information than did men (Murray & McMillan, 1993).
Health information seeking has a number of important implications for
health outcomes (e.g., Gantz & Wang, 2004). For example, insufficient
information can potentially lead to poor decisions about alternative
treatment regimens (Lerman et al., 1993). Further, previous research
has shown that patients who are adequately informed about their
illnesses are better able to maintain a sense of control (Felton &
Revenson, 1984; Lerman et al., 1993; Viney & Westbrook, 1984) and to
cope with the uncertainty of the illnesses and their treatments
(Burish & Lyles, 1983; Evans & Clarke, 1983; Molleman et al., 1986).
Meanwhile, well-informed patients tend to follow their therapeutic
plans of care more closely (Eraker, Kirscht, & Becker, 1984). African
Americans traditionally have been less active seekers of information
regarding their illnesses than other ethnic groups (Freimuth, Stein,
& Kean, 1989).
Although it is often asserted that systematic communication can
improve health behavior, less profound impact of the media has not
been found (Piotrow et al., 1997). In examining the psychosocial
mechanisms through which symbolic communication influences human
thought, affect, and action, Bandura (2001) argued that "[m]ost
external influences affect behavior through cognitive processes
rather directly" (p. 267). Human beings are capable of
self-regulation and proactive in response to environmental events and
inner forces. On the one hand, cognitive factors help determine what
environmental events will be observed and how the information
conveyed through the events will be processed for future use. On the
other hand, one's knowledge, direct experience, and perceived
barriers and benefits will come into play and become a critical part
of the cognitive process that eventually determines whether the
environmental events will have any lasting effects in one's behavior.
In health reporting the news media are not free of bias and mistakes
(e.g., Hoffman- Goetz, 1999; Jones, 2004). For example, although
medical and health-related studies have repeatedly shown that women
over 40 are more likely to have breast cancer than those younger than
40, 44 percent of the breast cancer-related advertisements in ten
top-selling women's magazines and three weekend newspapers in an
Australian study featured women younger than 40 (Jones, 2004). In
both print media stories or advertisements, mammographic screening of
women over 50 was rarely mentioned (Jones, 2004). In a content
analysis of full-length cancer stories in magazines targeting African
Americans such as Jet, Ebony, and Essence from 1987 to 1995,
Hoffman-Goetz (1999) found that more than two-thirds of the stories
used strong words of shock, anger, panic, and fear to describe
negative emotions or reactions to an initial diagnosis of cancer.
Also frequently expressed in the descriptions were pain and side
effects related to cancer or cancer treatments. On the other hand,
73% of the documents described positive fatalism related to delivery
from death and the idea of being a survivor.
Although Bandura's Social Cognitive Theory (1997, 2001) suggests an
indirect model of media effects, it is not known whether frequent
consumption of news media is a causal influence on American women's
knowledge and attitudes toward breast cancer, as far as breast cancer
prevention is concerned. Nor are we sure what impact, either direct
or indirect, news media have on perceptions of breast cancer
messages. Do the media determine the amount of attention that is
given to a particular message about breast cancer at the individual
level? And if attention to the media messages is important, how does
it affect women's attitudes and perceptions of breast cancer? We feel
exploratory research is still needed to address the issues of how the
amount of media used to obtain health information, along with other
barriers to breast cancer prevention such as knowledge and religious
beliefs, would affect American women's attitudes toward breast
cancer prevention. Therefore, we wanted to test our first hypothesis:
H1. The information about breast cancer from various sources of news
media will be positively related to fear and efficacy of the disease
in Caucasian and African American women.
The role of religious beliefs in African American women's cancer prevention
Psychological responses such as fear, fatalism, and efficacy have
been identified as personal factors that inhibit African Americans
from attending to messages as well as participating in cancer health
promotion behaviors (Cardwell & Collier, 1981; Long, 1993; Barber et
al., 1998; Matthews et al., 2002; Lewis & Green, 2000). It has been
suggested that cultural-based beliefs about the causes of cancer are
what keep some African American women from seeking medical treatment
and tests for breast cancer. Although socioeconomic factors such as
income and social status have been found to account for half of the
cancer-related deaths in Caucasian women, previous research has shown
that attitudes and beliefs about causes of cancer are commonplace
among some African-American women (e.g., Johnson, 1998).
Researchers have begun to speculate that African American women are
more likely to have been brought up with misinformation about breast
cancer and its treatment, and their misinformation is directly passed
down from their mothers and grandmothers. This rationale suggests
that it is not only important for doctors to be aware of the cultural
and religious beliefs of their patients, but also that health
communicators need to be aware of these beliefs in order to create
advertising messages, campaigns, and appeals that more effectively
speak to African American women.
In a recent study on the perceptions African American women have
concerning breast cancer, data show that with respect to cancer, some
women believe that exposure to "air" causes cancer to grow (MSNBC,
1999). Moreover, the MSNBC study showed that more African American
women, compared to their counterparts, believe that "worry" or
anxieties worsen (or cause) their cancer, that God and prayers will
heal them of the cancer without a reliance on medical treatments or
doctors, and/or that the devil caused the cancer. These barriers, in
particular the thought that God causes or cures cancer, have been
demonstrated in prior research.
Research conducted by Clarke-Tasker (1993), for example, noted that
historically, spirituality has played a crucial role in the lives of
African American women. Consistent with the MSNBC study, African
American women, it seems, may associate a breast cancer diagnosis to
a failure to live according to God's will. In a similar study,
Jennings (1996) concluded that African American women may believe
that breast cancer screening is futile in the face of God's will
because God can still inflict breast cancer upon them. Hughes et al.
(1996) attributed differences in perceptions about breast cancer
risks to the influence of cultural factors such as the importance of
interpersonal relationships, spirituality, and time orientation.
Religious beliefs were found to play an important role in health care
practices and emotional adjustment of African American cancer
patients (e.g., Barber et al., 1998; Matthews et al., 2002), and in
the willingness of attending church-based health promotion programs
by African Americans (Lewis & Green, 2000). They consequently became
more reluctant to seek cancer information and relied more on the
belief that "God is going to take care of me" (Matthews et al., 2002, p. 213).
The above influence of religious beliefs on African American women's
attitudes toward cancer prevention in general led us to the
development of our second hypothesis:
H2: African American women scoring higher on religious beliefs will
report less fear of breast cancer than Caucasian women.
The role of knowledge in cancer prevention
As far as cancer prevention is concerned, knowledge is important
because it helps people not only better cope with cancer but also do
the right things to prevent the disease. With adequate knowledge
about breast cancer, for instance, women are able to look for
appropriate ways or strategies to deal with the disease; they are
also able to know what behavior is appropriate and what behavior is
inappropriate to do in maintaining their health (e.g., Bandura, 1997;
Denniston, 1981). In addition, adequate knowledge about breast cancer
helps women develop appropriate attitudes such as lower level of
breast cancer anxiety (Royak-Schaler et al., 1995; Price et al.,
1992; Frisby, 2002).
On the other hand, general cancer knowledge correlates positively
with both education and income (e.g., Stone and Siegel, 1986). When
education was used as a predictor variable in a regression analysis
of the cancer knowledge index, more of the variance in cancer
knowledge could be explained by education rather than by income. The
Stone and Siegel study further supports the idea that ethnicity
influences level of cancer knowledge. They discovered that even after
controlling for education, behavioral intentions, and cancer
anxieties, African Americans had significantly lower levels of
knowledge of cancer than Caucasians, Asians, and other groups.
Research conducted by the American Cancer Society (1996) lends
support to the notion that people from more deprived communities are
less likely to seek medical care because they may not have health
insurance or are not able to afford the cost of traveling to a
medical doctor. Because they are disproportionately represented
among the nation's poor (e.g., Squires, 2002), African American women
are at an increased risk of cancer incidence and mortality.
Further, direct or indirect personal experience in dealing with
breast cancer is another important factor that influences one's
knowledge about the disease and hence her motive to seek for
information related to breast cancer. One's cognitive capacity might
be diminished if he or she experiences extreme stress or anxiety
about an unknown illness or medical procedures. For example, some
African American cancer patients with close family members or friends
who experienced cancer demonstrated increased fear, suspicion, and
pessimism about their prognosis (Matthews et al., 2002).
Misperceptions and lack of factual knowledge about breast cancer have
been explored as barriers to breast cancer-related activities such as
mammography among African Americans (Royak-Schaler et al., 1995;
McDonald et al., 1999). In a recent study that assessed breast cancer
perceptions, knowledge, and screening behavior among 120 low-income
African-American women residing in public housing, knowledge about
breast cancer was found to be poor among the women (McDonald et al.,
1999). Lack of factual knowledge about the disease could contribute
to individuals' misperceptions of the risk factors for breast cancer
and affect their anxiety about the disease.
In a more recent study, Frisby (2002) conducted a survey of
ninety-two African American women and discovered that less than 2% of
women over the age of 40 had either experienced a mammogram or
conducted self-examinations. Moreover, her data revealed a startling
fact: African American women were unable to communicate knowledge or
a clear understanding of the risks associated with breast cancer,
suggesting that campaigns aimed at increasing awareness are clearly
important for such a deadly disease.
Thus, the above discussion helped us develop the third hypothesis as follows:
H3: Knowledge of breast cancer and its risks will significantly
increase perceived efficacy levels for both Caucasian and African
American women.
Method
Data and Sample
The above hypotheses were tested in a telephone survey of 446
interviews with U.S. women aged 40 or older conducted from February
until July 2004, with 240 being Caucasians and 206 African Americans.
Women aged 40 or older were selected because risk factors for breast
cancer in this age group tend to be significantly higher (CDC, 2004).
The respondents were randomly selected from eleven metropolitan areas
where disparities in cancer outcomes were known to exist. The areas
included Baltimore, Buffalo, Chicago, Dallas, Detroit, Houston,
Jacksonville, Kansas City (Missouri), Milwaukee, Nashville, and
Richmond. In each of the cities, African American women were over
sampled to make their proportion of the sample equivalent to Caucasians'.
The interviews were completed by trained interviewing and supervising
staff of an academic survey research center at a Midwestern
university using random digit dialing (RDD) techniques. The data
collection was aided within a computer-assisted telephone
interviewing (CATI) system. The response rate of the survey was 50%,
using the final codes and definitions for calculating response rate
provided by The American Association For Public Opinion Research
(AAPOR, 2000).
Measurement
Dependent variable
The dependent variables of the study were two factors that emerged
from an exploratory factor analysis of nine question items using the
maximum likelihood factor extraction method and varimax rotation. The
first factor, labeled as "fear of breast cancer," consisted of three
question items: "Getting breast cancer would make me feel isolated
from other people," "I feel that others would avoid me if I had
breast cancer," and "Having breast cancer might make being in public
uncomfortable." The factor displayed an eigenvalue of 2.11. The
second factor, "efficacy of breast cancer," contained three items: "I
avoid some behaviors because they may cause breast cancer," "There
are many things I can do to avoid breast cancer," and I follow some
behaviors because they may prevent breast cancer." It had an
eigenvalue of 1.54. Responses to all the six items were coded on a
7-point Likert scale with 1 being "strongly disagree" and 7 "strongly
agree." The two factors were not significantly correlated (r = .06, p
= .17). Combined, they accounted for 61 percent of the total
variance. Table 2 displays the question items and corresponding
factor loadings. Factor scores were computed and used in the analysis.
Independent variables
We used five variables to measure one's attention to breast cancer
information on various channels of television, radio, newspapers,
magazines, and the Internet. Response categories for the items were
(1) not at all, (2) a little, (3) some, and (4) a lot. Attention, as
a measure of one's mental effort (e.g., McLeod & McDonald, 1985;
Chaffee & Schleuder, 1986), indicates how much one engages his or her
mind when watching television programs and reading newspapers. When
making comparison among various media channels (for example,
television versus newspapers, television versus the Internet), one is
recommended to use measurement of attention because "adding media
attention measures to the comparison can reduce the spurious
influence of third variables on tests of cognitive effects" (Chaffee
& Schleuder, 1986, p.103).
Because of the importance of religious beliefs in African Americans'
health care practices and information seeking (e.g., Barber et al.,
1998; Matthews et al., 2002; Lewis & Green, 2000), we averaged the
scores of seven question items to calculate a composite score of
"religious beliefs." The seven items (alpha = .91 for Caucasians;
alpha = .82 for African Americans) included "If I lead a good
spiritual life, I will stay healthy," "If I stay healthy, it is
because I am right with God," "I rely on God to keep me in good
health," "Through my faith in God, I can stay healthy," "My spiritual
beliefs are the foundation of my whole approach to life," "I am often
aware of the presence of God in my life," and "I have a personal
relationship with God." The response categories for the items ranged
from 1 "strongly disagree" to 7 "strongly agree." A high value of the
score indicates high level of religious beliefs.
To measure one's knowledge about breast cancer, we used six question
items: "In terms of risk for breast cancer, age does not play a major
role," "In terms of risk for breast cancer, family background is
important," "The age at which a girl has her first period influences
her risk for breast cancer later in life," "Lifestyle, including diet
and exercise, is one of the risk factors related to breast cancer,"
"Breast cancer can occur to every adult woman at any time," and
"Getting a mammogram will keep me from getting breast cancer."
Respondents were asked whether these six items were true (1) or false
(0). The item of "In terms of risk for breast cancer, age does not
play a major role" was reverse coded because age is one of the key
risk factors for breast cancer (CDC, 2004; NCI, 2004). We then
created an index of "knowledge about breast cancer" by summing all
correct responses. A "6" in the index means that a respondent
answered all the six questions correctly, and a "0" indicates that
none of her responses was correct.
Demographic and control variables
The demographic variables included age, education, and income. In
addition, we controlled for the effects of both direct and indirect
experiences with breast cancer: "Have you ever been diagnosed with
breast cancer?" and "Does anyone in your family have breast cancer?"
Both variables were dummy coded, with 1 being "yes" and 0 "no."
Analytical procedures
We used multivariate hierarchical regression to test the hypotheses
in both Caucasian and African American women. In the full model
(Table 4), the demographic and control variables were entered into
the equation first. Then, measures of "religious beliefs" and
"knowledge about breast cancer" were entered in two steps,
respectively. The purpose was to identify unique contributions, if
any, that these two variables would make to the prediction of fear
and efficacy of breast cancer. In the fourth step, five measures of
attention to breast cancer information in various media were entered
as one group.
We assessed the significance of the regression models by examining
the standard incremental F-tests (Cohen & Cohen, 1983). In the
regression equation, the standardized coefficients were derived from
the final model and reported in Tables 3. For the incremental F- and
t-tests, alpha was set at 0.05.
Results
Table 1 displays the descriptive statistics for all the variables
used in the analysis. Because the survey screened for female
respondents aged 40 or older, the average age of Caucasian women was
59.4 years (SD = 12.4), and the African American women's was 55.1
years (SD = 10.8). On average, Caucasian women had an educational
level between "two-year junior or community college degree" and "some
university studies but no degree" (M = 15.7, SD = 2.6), slightly
higher than their counterparts' (M = 14.7, SD = 3.0). The average
annual household income of Caucasian women was slightly higher than
the category of "$30,000 but less than $50,000" (M = 3.2, SD = 1.5),
whereas African American women on average made between "$20,000 but
less than $30,000" and "$30,000 but less than $50,000" a year (M =
2.4, SD = 1.5). In addition, five percent of Caucasian women had been
diagnosed with breast cancer, compared to three percent of African
American women. Twenty percent of Caucasian women reported that some
of their family members had breast cancer, so did 21 percent of
African American women.
To see whether there would be any significant difference between the
two samples in religious beliefs, we ran an analysis of variance
(ANOVA) with post hoc tests. Our analysis revealed a significant
effect, F(1, 446) = 83.6, p < .000 (adjusted R2 = 16%), with African
American women (M = 5.97, SD = 0.11) scoring significantly higher on
religious beliefs than Caucasian women (M = 4.67, SD = 0.10).
Tests of hypotheses
In our first hypothesis, we anticipated that the information about
breast cancer from various sources of news media would be positively
related to fear and efficacy of the disease in Caucasian and African
American women. As shown in Table 3, when measures of demographics,
direct and indirect personal experiences, and religious beliefs were
statistically controlled, attention to radio messages about breast
cancer was significantly and positively related to efficacy of breast
cancer for African American women (ß = .32, p = .001), but not for
Caucasian women. Attention to breast cancer information on the
Internet was significantly and positively related to the outcome
variable for Caucasian women (ß = .15, p = .01), but not for the
African American sample. Television news, newspapers, and magazines,
were not predictors of efficacy of breast cancer in either of the two
groups of women.
As for the other dependent variable of the study, fear of breast
cancer, the only significant predictor was the use of magazines, as
it was evident for the African American sample (ß = -.26, p = .01,
Table 3). This suggests that the more attention African American
women paid to information about the disease in magazines, the less
fearful attitudes they developed. None of the other measures of news
media use was significant for the two groups of women examined here.
Nevertheless, it should be noted that the media use measures along
contributed a significant 7% in variance for Caucasian women and 9.7%
for African American women, respectively, to the variances of
efficacy of breast cancer accounted for by the equation. The same
measures contributed a marginally significant 5.2% in variance for
African American women to the variance of fear of breast cancer, and
an insignificant 1.2% in variance for Caucasian women. Thus,
Hypothesis 1 was only partially supported.
Hypothesis 2 suggested that African American women scoring higher on
religious beliefs would report less fear of breast cancer than
Caucasian women. As shown in Table 3, religious beliefs were
significantly and negatively related to fear of breast cancer for
African American women (ß = -.16, p = .01), when demographic
variables and media use measures were statistically controlled. This
suggests that for the African American women who held stronger
religious beliefs, they would be less fearful of breast cancer. As
for the Caucasian sample, the same finding was not evident.
Noticeably, the measure of religious beliefs along contributed a
significant amount of two percent in variance in accounting for the
outcome variable (Table 3), and the predicative power remained
significant in the full model (Table 4) when the effect of "knowledge
about breast cancer) was simultaneously examined. For Caucasian
women, the variance in "fear of breast cancer" accounted for by
"religious beliefs" was an insignificant 0.4%. Therefore, Hypothesis
1 was adequately supported.
It should be noted that, when the regression was repeated with
"efficacy of breast cancer" being the dependent variable, religious
beliefs were significantly and positively related to the outcome
variable for Caucasian women (ß =.17, p = .01, Table 3), controlling
for demographic and media use variables. This finding suggests that
Caucasian women having a stronger sense of religion feel more control
in their coping with breast cancer than those with less or no belief
in religion. In addition, religious beliefs were not a significant
factor in accounting for efficacy of breast cancer for African
American women. These findings remained almost unchanged in the full
model (Table 4).
In Hypothesis 3, we anticipated that knowledge of breast cancer and
its risks would significantly increase perceived efficacy levels of
breast cancer for both Caucasian and African American women. In other
words, women who are knowledgeable about the disease and its
prevention would report higher efficacy levels than those who are
less knowledgeable or have no knowledge at all. This hypothesis was
well supported. Knowledge of breast cancer was significantly and
positively related to efficacy of the disease for Caucasian women (ß
= .35, p = .001) as well as for African American women (ß = .19, p >
.01), when the demographic measures, religious beliefs, and media use
variables were statistically controlled. Both Caucasian and African
American women who scored higher on the knowledge index demonstrated
higher levels of control and efficacy of breast cancer. In addition,
knowledge of breast cancer by itself noticeably added a significant
9% and 6.4% in variance to the variances of the outcome variable
accounted for in the equation in Caucasian and African American
women, respectively.
Because of the significant predictive power of knowledge about breast
cancer as evidently mentioned above, we ran another hierarchical
regression to see what factors, if any, would explain women's
knowledge. The results are displayed in Table 5. Noticeably, for
Caucasian women, their knowledge about breast cancer was more
significantly predicted by the demographic variables as well as
direct and indirect personal experiences than that of African
American women, as evident as in the variances of 21.1% and 2.9%,
respectively. Income was significantly and positively related to
knowledge for Caucasian women (ß = .21, p = .01), and marginally
significant for African American women (ß = .11, p = .10). As for the
Caucasian sample, also significant were age (ß = -.28, p = .001),
education (ß = .10, p = .10), and direct personal experience in
dealing with breast cancer (ß = .20, p = .001).
In addition, attention to breast cancer information on the radio was
significantly and positively related to knowledge about breast cancer
for Caucasian women (ß = .20, p = .01, Table 5) as well as for
African American women (ß = .26, p = .01, Table 5), suggesting that
radio could be an effective source of information about the disease
for both groups of women. Individually, none of the other media was
significant. Together, the media use measures were more influential
to African American women (10.3% in explained variance) than to their
counterparts (4.3% in explained variance), as far as knowledge about
breast cancer is concerned. Religious beliefs were marginally
significant for African American women's knowledge (ß = -.12, p =
.10, Table 5), but not for Caucasian women's.
Among the demographic variables, age was significantly and positively
related to efficacy of breast cancer for African American women (ß =
.15, p = .05, Table 4); it was marginally significant and negatively
related to fear of breast cancer for Caucasian women (ß = -.11, p =
.10, Table 4). Income, on the other hand, was significantly and
negatively related to fear of breast cancer for the Caucasian sample
(ß = -.18, p = .01, Table 4) as well as for the African American
sample (ß = -.20, p = .01, Table 4), implying that those who made
less annual income were more fearful of the disease than those with
higher incomes. As for the outcome variable of efficacy of breast
cancer, income was a marginally significant and positive predictor
for African American women (ß = .12, p = .10, Table 4), but not a
factor at all for Caucasian women. In addition, the Caucasian women
who had been diagnosed with breast cancer were less fearful of the
disease (ß = -.24, p = .001, Table 4). And, the African American
women whose family members had been diagnosed with breast cancer felt
less in control of the disease (ß = -.18, p = .01, Table 4). In
addition, education was not a predictor at all in either Caucasian or
African American women in accounting for their fear and efficacy of
breast cancer.
Discussion
In the present study we used two hierarchical multiple regressions to
see whether knowledge about breast cancer, religious beliefs, and use
of various sources of news media would influence fear and efficacy of
the disease in Caucasian and African American women. Fear and
efficacy of breast cancer have been identified as two important
concepts in the literature of breast cancer prevention, reflecting
two opposing attitudes common in women. Because of the importance of
knowledge in its relations to attitude and behavior in cancer
prevention at the individual level, we ran another multiple
regression to identify factors, along with demographic measures, that
would affect knowledge about breast cancer. Overall, we found
adequate support for our expectations.
First, use of magazines was a strong predictor of fear of breast
cancer in African American women (Table 3), and its predictive power
remained unchanged when the effect of knowledge about breast cancer
was simultaneously examined (Table 4). This finding suggests that
magazines could be an effective source of information that can be
used to help African American women reduce their fear of breast
cancer. Previous research has shown that magazines are important
sources of cancer information (e.g., Jones, 2004; Frisby, 2002) and
particularly heeded by more women than men (Meissner et al., 1992;
Murray & McMillan, 1993). Unlike most newspapers that appeal to
broader and more general audiences, for example, magazines are often
designed for quite narrow audiences; they are more specialized and
can more easily be targeted to a niche market interested in certain
health issues. In addition, magazine articles can devote more space
to a health issue than can newspapers. Magazines targeting African
American women are also known for their content and context of
autobiographical, biographical, or fictionalized 'real life' cancer
stories (Hoffman-Goetz, 1999). Cognitively, the detailed contents and
tone of the cancer stories are likely to influence African American
women of popular magazines in developing less negative perceptions of
cancer-related information and reducing their fear about breast
cancer as found in the present study.
The other media outlets examined in the study also showed more or
less effect in influencing women's attitudes toward breast cancer.
Attention to messages about breast cancer on the radio was an
influential factor in explaining African American women's efficacy of
the disease, and use of the Internet helped increase Caucasian women'
efficacy (Tables 3 & 4). Our finding that radio was directly related
to African American women's anxiety about breast cancer suggests that
this format of broadcasting may deserve adequate attention in the
domain of health communication. Like television news and newspapers,
radio focuses on current events and is entirely auditory (Eveland,
2003). These attributes of the medium may be more attractive to
African American women than to their counterparts. If used
appropriately, radio could be an effective approach for public health
practitioners to reach African American women for breast
cancer-related issues.
Meanwhile, both television news and newspapers failed to show any
significant effect. One of the possible explanations is that
television news and newspapers usually appeal to broader and more
general audiences and readers and, as a result, they may not devote
enough space or coverage to a particular health issue more
interested, say, by women. Although health content is regularly
featured on local television newscasts, there is a wide variety of
topics in health news stories and most of the news stories tend not
to offer contrasting viewpoints (Gantz & Wang, 2004). Nevertheless,
these differences in media effects reported here suggest that more
attention needs to be paid to study the unique attributes of various
media because they may influence how people select and use the media
for their informational needs for health (e.g., Eveland, 2003; Jones,
2004; Witte et al., 1993).
The results of the present study also show that stronger religious
beliefs led to less fear of breast cancer in African American women
and to more efficacy of the disease in Caucasian women (Tables 3 &
4). This finding is consistent with previous research and the notion
that religious beliefs play an important role in emotional adjustment
about breast cancer in African American women (e.g., Barber et al.,
1998; Matthews et al., 2002; Clarke-Tasker, 1993; Jennings, 1996;
Hughes et al., 1996). It seems that the stronger religious beliefs
African American have, the less fearful they feel about breast
cancer. The positive impact of religious beliefs on Caucasian women's
efficacy of the disease is noteworthy as well, suggesting that
religion is not only an important factor that influences women's
attitudes toward breast cancer, but also churches and church-related
activities may be an important area for public health professionals
and alike to focus their educational and promotional resources in
order to effectively discuss issues related to breast cancer
prevention with the two groups of women.
Further, the present data provide strong support to the importance of
knowledge in breast cancer research as well as in its relationship to
women's attitudes toward control of the disease (Table 4). Knowledge
about breast cancer was significantly and positively related to
efficacy of the disease for both groups of women, when demographic
measures, religious beliefs, and media use variables were
statistically controlled. The more correct knowledge they had about
breast cancer prevention, the more in control of preventing the
disease the women felt. However, our data did not show that knowledge
was related to fear of breast cancer in both Caucasian and African
American women. This could suggest that fear of breast cancer may
have more to do with other factors such as personal experience in
dealing with the disease for Caucasian women, and religious beliefs
and use of news media for African American women, as evident in Table 4.
In addition, our results seem to suggest that because knowledge about
breast cancer has been shown to be relatively poor among African
American women (Royak-Schaler et al., 1995; McDonald et al., 1999;
Frisby, 2002), they may pay more attention to the information in the
media than Caucasian women, as suggested by the finding that the
amount of variance of knowledge explained by the five media variables
in the African American sample was 10.3%, compared to 4.3% of the
variance in the Caucasian women sample (Table 5). Our analysis here
also shows that demographic measures, including personal experience
in dealing with breast cancer, played a much more important role in
accounting for Caucasian women's knowledge than for African
Americans'. As mentioned earlier, radio was influential in predicting
knowledge about breast cancer in both samples, suggesting that
broadcast media could be an important source to disseminate correct
information about breast cancer and hence increase women's knowledge
about the disease.
By examining the relationships of religious beliefs, knowledge, and
use of various sources of news media to women's attitudes toward the
disease, our study contributes to the understanding of the processes
associated with acceptance and rejection of breast cancer messages at
the individual level. From a practical point of view, our results
imply that, in choosing a media channel or channels to promote
appropriate attitudes or behaviors, health communicators should
carefully consider the specific attributes of various media that
appeal to different audiences. However, due to the correlational
nature of our data, we could not be certain of the direction of
causality – whether knowledge influences attitudes toward breast
cancer or vice versa, for example. A multiwave panel design would
enhance the quality of the proposed model. In addition, it would
enable us to draw more sound conclusions regarding the effectiveness
of various sources of news media for dissemination of useful messages
about breast cancer had we had measures that tap into the specific
attributes of different media as suggested by Eveland (2003).
References
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of
behavioral change.
Psychological Review, 84, 191-215.
Bandura, A. (2001). Social cognitive theory of mass communication.
Media Psychology,
3(3), 265-266.
Barber, K. R. et al. (1998). Differences between African American and
Caucasian men
participating in a community-based prostate cancer screening program.
Journal of Community Health, 23(6), 441-451.
Berman, S. H., & Wandersman, A. (1990). Fear of cancer and knowledge
of cancer: A review
and proposed relevance to hazardous waste sites. Social Science of
Medicine, 31(1), 81-90.
Bobo, J. (1995). Black women as cultural readers. New York: Columbia
University Press.
Brashers, D. E., Goldsmith, D. J., & Hsieh, E. (April, 2002).
Information seeking and avoiding in
health contexts. Human Communication Research, 28(2), 258-271.
Brodie, M., Hamel, E. C., Altman, D. E., Blendon, R. J., & Benson, J.
M. (2003). Health news
and the American public, 1996-2002. Journal of Health Politics,
Policy, and Law, 28(5), 927-950.
Brown, R. S., & Lees-Haley, P. R. (1992). Fear of future illness,
chemical AIDS, and
cancerphobia: A review. Psychological Reports, 71, 187-207.
Cardwell, J. J., & Collier, W. V. (1981). Racial differences in
cancer awareness: What Black
Americans know and need to know about cancer. Urban Health, 12(3), 29-32.
Center for Disease Control and Prevention (CDC) (2004). 2003 fact
sheet – revised May
2004: The national breast and cervical cancer early detection program
– reducing mortality through screening. Retrieved October 2, 2004
from http://www.cdc.gov/cancer/nbccedp/about.htm.
Chaffee, S. H., & Schleuder, J. (1986). Measurement and effects of
attention to media news.
Human Communication Review, 13(1), 76-107.
Clark-Tasker, V. A. (1993). Cancer prevention and early detection in
African Americans. In M.
Frank-Stromborg & S. J. Olsen (Eds.), Cancer prevention in minority
populations: Cultural implications for health care professionals. St.
Louis, MO: V. C. Mosby.
Cohen, J., & Cohen, P. (1983). Applied multiple
regression/correlation analysis for the
behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.
Coleman, C. (1993). The influence of mass media and interpersonal
communication on societal
and personal risk judgments. Communication Research, 20(4), 611-628.
Dent, O. W., & Goulston, K. (1982). Community attitudes to cancer.
Journal of Biosocial
Science, 14, 359-372.
Denniston, R. W. (1981). Cancer knowledge, attitudes, and practices
among Black Americans. In
C. Metlin & G. P. Murphy (Eds.), Cancer among Black populations. New
York: Alan R. Liss.
Easerling, D. V., & Leventhal, H. (1989). Contributions of concrete
cognition to emotion: neutral
symptoms as elicitors of worry about cancer. Journal of Applied
Psychology, 74, 787-796.
Eveland, W. P. Jr. (2003). A "mix of attributes" approach to the
study of media effects and new
communication technologies. Journal of Communication, 53(3), 395-410.
Frisby, C. M. (2002). Messages of hope: Health communication
strategies that address barriers
preventing Black women from screening for breast cancer. Journal of
Black Studies, 32(5), 489 – 505.
Gantz, W., & Wang, Z. (2004, August). Health content in local
television news. Paper presented
at the annual convention of the Association for Education in
Journalism and Mass Communication (AEJMC) in Toronto, Canada.
Graber, D. A. (1997). Mass media and American politics. Washington,
D.C.: Congressional
Quarterly Inc.
Gram, I. T., & Slenker, S. E. (1992). Cancer anxiety and attitudes
toward mammography among
screening attenders, nonattenders, and women never invited. American
Journal of Public
Health, 82(2), 249-251.
Gutteling, J. M., Seydel, E. R., & Weigman, O. (1986). Perceptions of
cancer. Journal of
Psychosocial Oncology, 4(3), 77-91.
Hoffman-Goetz, L. (1999). Cancer experiences of African-American
women as portrayed in
popular mass magazines. Psycho-Oncology, 8, 36-45.
Holland, R., & Cullen, A. (1986). New insights and attitudes. In A.
I. Holleb (Ed.), The
American Cancer Society book: Prevention, detection, diagnosis,
treatment, rehabilitation, cure. New York: Doubleday.
Hughes, C., Lerman, C., & Lustbader, E. (1996). Ethnic differences in
risk perception among
women at increased risk for breast cancer. Breast Cancer Research and
Treatment, 40, 25-35.
Jennings, K. (1996). Getting black women to screen for cancer.
Incorporating health beliefs into
practice. Journal of the Academy of Nurse Practitioners, 8(2), 53-59.
Jones, S. C. (2004). Coverage of breast cancer in the Australian
print media – does advertising
and editorial coverage reflect correct social marketing messages?
Journal of Health Communication, 9, 309-323.
Kaiser Family Foundation, & Harvard School of Pubic Health (2002).
Health News Index Poll.
Retrieved October 26, 2004 from http://www.kaisernetwork.org.
Kellner, R., Abbott, P., Winslow, W. W. & Pathak, D. (1987). Fears,
beliefs, and attitudes in
DSM-III hypochondriasis. The Journal of Nervous and Mental Disease,
175(1), 20-25.
Lewis, R. K., & Green, B. L. (2000). Assessing the health attitudes,
beliefs, and
Behaviors of African Americans attending church: A comparison from
two communities. Journal of Community Health, 25(3), 211-224.
Long, E. (1993). Breast cancer in African American women: Review of
the literature. Cancer
Nursing, 16(1), 1-24.
Matthews, A. K., Sellergren, S. A., Manfredi, C., & Williams, M.
(2002). Factors influencing
medical information seeking among African American cancer patients.
Journal of Health Communication, 7, 205-219.
Mattson, M. (1999, September). Toward a reconceptualization of
communication Cues to action
in the Health Belief Model: HIV test counseling. Communication
Monographs, 16, 240-265.
McDonald, P. A., Thorne, D. D., Pearson, J. C., & Adams-Campbell, L.
L. (1999, Winter).
Perceptions and knowledge of breast cancer among African-American
women residing in public housing. Ethn Dis., 9(1), 81-93.
McLeod, J. M., & McDonald, D. G. (1985, January). Beyond simple exposure: Media
orientations and their impact on political processes. Communication
Research, 12, 3-33.
Meissner, H. I., Potosky, A. L., & Convissor, R. (1992). How sources
of health information
relate to knowledge and use of cancer screening exams. Journal of
Community Health, 17, 153-165.
Murray, M., & McMillan, C. L. (1993). Gender differences in
perceptions of cancer. Journal of
Cancer Education, 8, 53-62.
National Cancer Institute (NCI, 2004). Quick facts: Annual report to
the nation on the status of
cancer 1975-2000, featuring the uses of surveillance data for cancer
prevention and control. Retrieved October 3, 2004 from
http://www.cancer.gov/newscenter/pressreleasese/reportfactsheet
Rosenstock, I. M., Strecher, V. J., & Becker, M. H. (1988, Summer).
Social learning theory and
the health belief model. Health Education Quarterly, 15(2), 175-183.
Royak-Schaler, R., DeVellis, B. M., & Emerson, J. A. (1995). Breast
cancer in African-
American families: Risk perception, cancer worry, and screening
practices of first-degree relatives. Annals of the New York Academy
of Sciences, 768, 281-285.
Speckens, A. E. M., Spinhoven, P., Sloekers, P. P. A., Bolk, J. H., &
van Hemert, A. M. (1996).
A validation study of the Whitley Index, the Illness Attitude Scales,
and the Somatosensory Amplification Scale in general medical and
general practice patients. Journal of Psychosomatic Research, 40(1), 95-104.
Steptoe, A., Horti, J. & Stanton, S. (1986). Concern about cancer in
women undergoing elective
gynecological surgery. Social Science and Medicine, 23(11), 1139-1145.
Squires, C. (2002). Black audiences, past and present: Commonsense
media critics and activists.
In R. R. M. Coleman (Ed.), Say it loud! African-American audiences,
media, and identity (pp. 45-76). New York: Routledge.
The American Association for Public Opinion Research. (2000).
Standard definitions: Final
dispositions of case codes and outcome rates for surveys. Ann Arbor,
Michigan: AAPOR.
Witte, K., Stokols, D., Ituarte, P., & Schneider, M. (1993, August).
Testing the Health Belief
Model in a field study to promote bicycle safety helmets.
Communication Research, 20(4), 564-586.
Wood, J. F. (2002). House Negro versus field Negro: The inscribed
image of race in television
news representations of African-American identity. In R. R. M.
Coleman (Ed.), Say it loud! African-American audiences, media, and
identity (pp. 96-114). New York: Routledge.
1
Fear & Efficacy of Breast Cancer
TABLE 1 Descriptive Statistics for Demographic Controls and Media
Variables Used in the Analysis
Caucasian Women (n=240) African American Women (n=206)
Variables Mean s.d. Range Mean s.d. Range
Demographics & Controls
Age 59.36 12.35 40-87 55.09 10.78 40-91
Education 15.70 2.61 3-20 14.67 3.03 1-20
Income 3.15 1.53 0-6 2.37 1.51 0-6
Have you ever been diagnosed with breast cancer
b .05 .22 0-1 .03 .17 0-1
Does anyone in your family have breast cancer
c .20 .40 0-1 .21 .41 0-1
Attention to breast cancer information in the media
Attention to breast cancer information on
TV 2.86 .88 1-4 3.23 .87 1-4
Attention to breast cancer information on the
radio 2.20 1.04 1-4 2.39 1.18 1-4
Attention to breast cancer information in newspapers
2.67 1.06 1-4 1.68 1.14 1-4
Attention to breast cancer information in magazines
2.83 .98 1-4 2.95 1.04 1-4
Attention to breast cancer information on the Internet
1.90 1.08 1-4 1.79 1.13 1-4
Note. a & b. yes = 1, no = 0.
1
Fear & Efficacy of Breast Cancer
TABLE 3
Hierarchical Multiple Regression Predicting Fear and Efficacy of
Breast Cancer without Knowledge about Breast Cancer
Fear of Breast Cancer Efficacy of Breast Cancer
Caucasians African
Americans Caucasians African
Americans
(n = 240) (n = 206) (n =
240) (n = 206)
Step 1
Age -0.13* 0.04 -0.03 0.16*
Education -0.07 0.05 0.06 -0.03
Income -0.17* -0.18** 0.08 0.14*
Have you ever been diagnosed with breast cancer
a -0.22** -0.09 0.05 -0.03
Does anyone in your family have breast cancer b
0.04 -0.09 0.00 -0.18**
Incremental R2 (%) 8.8** 5.3* 4.1 6.2*
Step 2
Religious beliefs -0.06 -0.16** 0.17** 0.02
Incremental R2 (%) 0.4 2.0* 3.7** 0.7
Step 3
Attention to breast cancer info on TV 0.05 0.10 0.08 0.06
Attention to breast cancer info on the
radio -0.09 0.12 0.11 0.32***
Attention to breast cancer info in newspapers 0.02 0.15 -0.05 -0.09
Attention to breast cancer info in magazines 0.06 -0.26** 0.11 0.04
Attention to breast cancer info on the Internet
0.01 -0.01 0.15* -0.05
Incremental R2 (%) 1.2 5.2 # 7.0** 9.7**
Total R2 (%) 10.4 12.5 14.8 16.6
Note. a & b. yes = 1, no = 0.
Entries are standardized beta coefficients.
# p _ .10, *p _ .05, **p _ .01, ***p _ .001.
TABLE 4
Full Hierarchical Multiple Regression Predicting Fear and Efficacy of
Breast Cancer
Fear of Breast Cancer Efficacy of Breast Cancer
Caucasians African
Americans Caucasians African
Americans
(n = 240) (n = 206) (n =
240) (n = 206)
Step 1
Age -0.11 # 0.03 0.06 0.15*
Education -0.08 0.05 0.03 -0.02
Income -0.18** -0.20** 0.00 0.12 #
Have you ever been diagnosed with breast cancer
a -0.24*** -0.10 -0.02 -0.05
Does anyone in your family have breast cancer b
0.04 -0.10 0.01 -0.18**
Incremental R2 (%) 8.8** 5.0* 4.1 6.2*
Step 2
Religious beliefs -0.05 -0.15* 0.20** 0.05
Incremental R2 (%) 0.4 2.0* 3.7** 0.7
Step 3
Knowledge about breast cancer 0.08 0.10 0.35*** 0.19**
Incremental R2 (%) 1.2 1.4 9.0*** 6.4***
Step 4
Attention to breast cancer info on TV 0.05 0.10 0.08 0.06
Attention to breast cancer info on the
radio -0.11 0.09 0.04 0.27**
Attention to breast cancer info in newspapers 0.02 0.15 -0.06 -0.09
Attention to breast cancer info in magazines
0.07 -0.26** 0.16* 0.04
Attention to breast cancer info on the Internet
0.01 -0.01 0.16* -0.07
Incremental R2 (%) 1.5 4.6 7.1** 6.6*
Total R2 (%) 10.9 13.2 23.9 19.8
Note. a & b. yes = 1, no = 0.
Entries are standardized beta coefficients.
# p _ .10, *p _ .05, **p _ .01, ***p _ .001.
1
Fear & Efficacy of Breast Cancer
TABLE 2
Questionnaire Items & Factor Loadings for Perceived Fear and Efficacy
of Breast Cancer
Factor 1 Factor 2
Questionnaire item Fear of Breast Cancer Efficacy of Breast Cancer
1. Getting breast cancer would make me feel
isolated from other people. 0.73 0.06
2. I feel that others would avoid me if I had breast
cancer. 0.60 0.10
3. Having breast cancer might make being in public
uncomfortable. 0.63 0.06
4. I avoid some behaviors because they may cause
breast cancer. 0.16 0.60
5. There are many things I can do to avoid breast cancer. 0.04 0.52
6. I follow some behaviors because they may prevent
breast cancer. 0.11 0.74
Note.
1. n = 446.
TABLE 5
Hierarchical Multiple Regression Predicting Knowledge about Breast Cancer
Caucasians African
Americans
(n = 240) (n = 206)
Step 1
Age -0.28*** 0.07
Education 0.10 # -0.04
Income 0.21** 0.11#
Have you ever been diagnosed with breast cancer a 0.20*** 0.08
Does anyone in your family have breast cancer b -0.02 0.01
Incremental R2 (%) 21.1*** 2.9
Step 2
Religious beliefs -0.08 -0.12 #
Incremental R2 (%) 0.3 0.03
Step 3
Attention to breast cancer info on TV 0.00 0.01
Attention to breast cancer info on the radio 0.20** 0.26**
Attention to breast cancer info in newspapers 0.02 0.03
Attention to breast cancer info in magazines -0.14 -0.01
Attention to breast cancer info on the Internet -0.04 0.13
Incremental R2 (%) 4.3* 10.3**
Total R2 (%) 25.8 13.5
Note. a, b, c, & d. Yes = 1, No = 0.
Entries are standardized beta coefficients.
# p _ .10, *p _ .05, **p _ .01, ***p _ .001.
|
