AN ENTERTAINMENT-EDUCATION VIDEO AS A TOOL TO INFLUENCE MAMMOGRAPHY
COMPLIANCE BEHAVIOR IN LATINAS
By
Gail D. Love, Ph.D.
Assistant Professor
Department of Communications
California State University-Fullerton
P. O. Box 6846
Fullerton, CA 92834-6846
Email: [log in to unmask]
Phone: 714 278-3587
A/V Reqmts: PC/projector/VCR
ABSTRACT
Purpose of the study is to gain a greater understanding of the relationship
between culturally sensitive health-related messages and screening behavior
among Latinas. A communication intervention in the form of a two-minute,
Spanish-language, entertainment-education video was field tested in
conjunction with a reinforcement interview one week after respondents
viewed the video. Implications are increased knowledge and diffusion of
information may not necessarily lead to short-term attitudinal and
behavioral change, particularly when unpleasant consequences may result.
AN ENTERTAINMENT-EDUCATION VIDEO AS A TOOL TO
INFLUENCE MAMMOGRAPHY COMPLIANCE BEHAVIOR IN LATINAS
INTRODUCTION
Widespread health communication campaigns for breast cancer screening in
the U.S. have been effective in promoting screening behaviors by white,
African American and higher socioeconomic (SES) women. In 2000, compliance
with mammography guidelines for white women stood at 72.1 percent and at
67.9 percent for African American women, however for Latinas the rate is
61.4 percent, and for women below the poverty level it is a dismal 55.2
percent (CDC 2002). Cultural barriers are a factor for many minority women
(CDC 2002), particularly those who are marginally acculturated. This is
tragically illustrated with Latinas, whose breast cancer incidence rate is
lower than that for whites and African-Americans, but whose mortality rate
is higher because their cancers are diagnosed so late (CDC 2002).
Health educators have a responsibility to reach women, particularly
minority women and low-income women, with a breast education program that
does not talk at them about early detection, but rather reaches out to them
with a culturally appropriate message about a behavior that is perceived as
both unfamiliar and unreasonable to them.
As we have seen countless times in health education programs conducted in
Third World countries, teaching does not guarantee doing, and knowledge
does not guarantee practice. It is imperative that we remember that
knowledge of health information usually translates to action only for
affluent, well-educated women who routinely practice early diagnostic and
preventive behaviors. Most other women require a different approach.
For Latinas in particular bilingual is not necessarily bicultural. Simply
translating English-language breast education programs is not going to
guarantee compliance. Breast health is, first and always, a very personal
matter. Open discussion about self-touching and self-observation of certain
parts of the body, such as the breasts, requires sensitive communication
for women whose cultural heritage discourages such practices.
In addition to the cultural barriers, which discourage self-examination
and mammography, many Latinas face a common barrier to early diagnostic or
preventive behaviors, which questions why a person should spend money on
medical procedures or go to the doctor when she is not sick or in
pain. Additionally, Latinas' culture is very family-oriented, with women
often cast as the central caregiver. Most Latinas take care of everyone
else first, often neglecting their own welfare.
The emerging genre of the entertainment-education strategy holds a great
deal of promise as a powerful vehicle to influence beliefs and attitudes of
audiences. Entertainment-education is "the process of purposely designing
and implementing a media message to both entertain and educate, in order to
increase audience members' knowledge about an educational issue, create
favorable attitudes, shift social norms, and change overt behavior"
(Singhal & Rogers, 2002; Singhal & Rogers 1999).
In the past three decades, a number of efforts in both Third World
countries and some Western industrialized countries have combined the
entertainment format with educational messages to produce innovative new
mass media messages (Singhal et al, 2002, Singhal et al, 1990). Radio and
television soap operas in particular have been used to communicate a wide
variety of educational messages, from family planning to sexual abstinence
by teens to AIDS prevention.
Intuitively, it makes sense to utilize the entertainment genre as a
vehicle for educational messages. For decades, mass society has looked to
entertainment media for guidance on what to wear, what to say, and what to
think. In most countries, society looks to the stars of film, television
and popular music for cues about fashion, appearance and behavior. Thus we
receive power educational messages via entertainment media, even though the
educational aspects may be unnoticed and unintentional (Rogers et al, 1989).
The strategy of entertainment-education is straightforward: Show people
how they can be happier, healthier and safer through the universal appeal
of entertainment (Rogers et al, 1989).
The notion of modeling behavior, wherein viewers model their own behavior
after that of characters seen in television programming has roots in social
learning theory (Bandura, 1977). Bandura suggests that humans learn social
behaviors through modeling their own behavior after the behavior of people
they see in the mass media or with whom they interact
personally. Entertainment-education soap operas generally provide both
positive and negative role models for the health behavior being
promoted. Positive role models are generally attractive to the audience
and should be rewarded for their behavior, while negative role models are
punished for negative behavior, and sometimes undergo a transition.
Miguel Sabido's entertainment-education soap operas in Mexico, which serve
as the basis for most entertainment-education projects in developing
countries, were solidly rooted in a theoretical base (Nariman,
1993). Sabido developed his entertainment-education soap opera formula
around five theories: Shannon and Weaver's mathematical theory of
communication (1948); the dramatic theory of Eric Bentley (1967); universal
energies theory as developed by Jung; Bandura's (1977) social learning
theory; and MacLean's (1973) concept of the triune brain. Social learning
theory is at the heart of Sabido's strategy, as he developed soap opera
characters who served as positive and negative role models for the
educational theme, and who received rewards or punishments as they
practiced or rejected socially desirable behaviors.
Singhal and Rogers (2002) contend it is time for researchers to "pay
greater attention to the various types of entertainment-education
interventions, including differences in their scope, size, reach,
intensity, and other attributes."
Because the present study limits exposure of the Latina audience to a
single viewing of the communication intervention, a modification to the
standard entertainment-education soap opera format is necessary.
One of the shortcomings of the soap opera genre as a one-time
communication intervention is its inherent dependence on repetitive
exposure. In outlining 20 guidelines for further development of
entertainment-education soap operas for communicating behavior for social
change, Nariman suggested that the entertainment-education strategy is most
effective "when applied in a repetitious form of entertainment" (Nariman,
1993, p. 181).
While the present study is the first known attempt to use the
entertainment-education soap opera format in a one-time, non-mass media
exposure, Johns Hopkins University's Center for Communication Programs used
a non-mass media format, dramatic street theater, which reached only a few
hundred people per performance in India and in Bolivia to communicate
HIV/AIDS information and to promote family planning respectively. The
street theater experience is that the one-shot exposure produces effects if
a counselor or educator steps forward immediately after the play ends to
discuss the play's message with the audience, and answer questions (Valente
& Bharath, 1999; Valente et al, 1995).
Purpose and Design of the Study
Culturally sensitive communication can play a role in influencing
health-related behavior among minority populations (Rogers, 1991). The
focus of the present research is to better understand the relationship
between culturally sensitive communication messages and mammography
screening behavior among Latinas.
Specifically, a communication intervention to increase knowledge, change
attitudes, increase interpersonal discussion, and increase mammography
compliance among Latinas was developed and field tested. The intervention
is a unique, two-minute Spanish-language video entitled Hermanas (Sisters),
which looks like a segment from a television soap opera. The video
combines the concepts of (1) a "trigger" video, specifically designed to
"trigger" discussion of the topic; and (2) an entertainment-education
strategy, patterned after the successful entertainment-education television
soap operas aired in Mexico in the late 1970s and 1980s. These works,
which were produced by Miguel Sabido, were solidly rooted in a theoretical
base centered on social learning theory, and provide the basis for most
subsequent entertainment-education projects in other developing
countries (Singhal and Rogers, 1999).
Hermanas depicts two Latina sisters relaxing over coffee at the end of the
day. The older sister (Sandra) reluctantly confesses that she failed to
keep her mammography appointment earlier that day. Chastised by her
younger sister (Marci), Sandra poignantly offers the excuses most cited by
Latinas for failing to get a mammogram. When Marci counters the excuses,
Sandra cites barriers most often identified by Latinas for
non-compliance. Marci's emotional responses offer viewers information on
the importance of mammography screening, while Sandra represents a
sympathetic figure fortunate enough to have someone who loves her to prod
her into a behavior she does not want to perform. The video closes with
Sandra promising to get her mammogram soon, and a slow pan to an attractive
young Latina health educator who reminds viewers that a mammogram can
detect breast cancer when it is barely larger than a grain of sand. She
reminds the audience that they should get a mammogram for themselves and
for those they love, once a year for a lifetime.
The entertainment-education video intervention was field tested by
embedding it within a one-hour, Spanish-language breast health program
provided by a trained, bilingual, Latina educator to Latinas at their work
sites and in the ambulatory clinic waiting room of a large
university-affiliated hospital, which offered low-cost or no-cost
mammograms on site. A three-group experimental design with a pre-test,
post-test and one-week follow-up test, operationalized measures of the
video's effectiveness: (1) increased knowledge of breast cancer;
(2) increased knowledge of mammography; (3) a more positive attitude
toward breast cancer; (4) increased discussion about breast cancer with
female friends and/or family members; and (5) increased mammography
compliance, defined as obtaining a mammogram or reporting the intention to
obtain a mammogram.
The effects of the entertainment-education video were studied in
conjunction with a related intervention (an experimental treatment): A
follow-up reinforcement interview conducted with Latina respondents one
week after the educational program by the bilingual health educator who
conducted the educational program.
Theoretical Foundation and Literature Review
The present study draws upon three theoretical perspectives: The
entertainment-education communication strategy; social learning theory; and
Lewinian theory, each of which will be discussed below.
From the earliest tabloid-like penny newspapers through today's
instantaneous worldwide dissemination of images via satellite and
videophone, mass media content can generally be categorized as either
entertaining or educational. Entertainment is "a performance or a
spectacle which captures the interest or attention of an individual giving
him/her pleasure, amusement, or some form of gratification." Education is
"a formal or informal program of instruction and training which has the
potential to develop an individual's skill to achieve a particular end by
boosting his/her mental, moral or physical powers (Singhal, 1990).
The above categories and definitions of "entertainment" and "education"
suggest links to uses and gratifications theory. Uses and gratification
theory assumes that mass communication audience members initiate media
selection, are goal-directed in their media use, and actively select
differing media and content to fulfill their felt needs (Katz et al., 1973;
Rubin, 1986; Tan, 1985). Additionally, uses and gratifications have been
researched for a variety of behaviors (e.g., McLeod et al, 1982), though
none specifically related to cancer-related information seeking.
What is generally considered to be the first recognizable interventions
using the entertainment-education concept occurred in several
countries: the Australian Broadcasting Corporation's The Lawsons in 1944;
the BBC radio series The Archers in 1951, and the Peruvian telenovela
(television soap opera) Simplemente Maria in 1969. At that time the notion
of education messages embedded in entertainment programs did not exist, and
communication scholars were not a part of the design or evaluation of
effects (Singhal et al 1994).
Following the broadcast of Simplemente Maria in Mexico, Miguel Sabido, who
worked as a creative writer-producer-director at Televisa, the private
Mexican television network, set about to understand the theoretical basis
of the telenovela by deconstructing it. Sabido then produced a series of
six entertainment-education television programs for Televisa, whose impacts
were evaluated (Nariman, 1993; Singhal & Rogers, 1999).
Sabido drew heavily on Albert Bandura's (1977, 1997) social learning
theory, which later evolved into social cognitive theory, in designing his
entertainment-education telenovelas. Since that time theorizing and
research about entertainment-education was dominated by this theoretical
approach. Sabido's methodology for the design of entertainment-education,
in particular soap operas, heavily influenced much of the later work in
entertainment education.
This "natural fit" between Bandura's theory and the
entertainment-education interventions which were to follow comes in the
form of the interventions' use of positive and negative role models which
seek to influence audience behavior (Singhal & Rogers 2002).
During the 1980s and 1990s, there were improvements in understanding the
effects of entertainment-education interventions, particularly since the
mid-1980s when university-based scholars became involved in research on
entertainment-education. As entertainment-education has evolved,
particular attention is being paid to the role of several key concepts:
social modeling, self-efficacy, parasocial interaction, and
media-stimulated peer communication in stimulating behavior change (Singhal
& Rogers 1999).
In social modeling theory, Bandura proposed that individuals observe and
imitate the behavior of other individuals, who serve as both positive and
negative role models, resulting in learned behavior changes (Bandura 1977,
1997). Bandura has now become increasingly interested in applications of
his theory to entertainment-education concepts, and has taken steps to
sharpen the strategy, particularly in regard to the concept of
self-efficacy, which is an individual's perception of his or her capability
to deal effectively with a situation, and a sense of perceived control over
a situation (Bandura, 1977, 1997; Rogers, 1995).
Further evidence of Bandura's interest in entertainment-education is
evidenced by his contribution of a chapter on "Social Cognitive Theory for
Personal and Social Change by Enabling Media" to Singhal et al (eds.) 2003.
Another important key in entertainment-education effects is the concept of
parasocial interaction, that is, the development of a parasocial
relationship with a fictional character. Horton and Wohl (1956) proposed
the concept of parasocial interaction, and Rubin et al (1985) and other
scholars (such as Rubin & McHugh, 1987) have measured parasocial
interaction between audience members and television role models. Using a
scale composed of several items (such as, "Do you talk to your faviorite
television character during the broadcast?") a high degree of parasocial
interaction was found. It is suggested audience members of
entertainment-education soap operas are influenced to change their behavior
through this parasocial interaction.
Past research within the entertainment-education strategy suggests that
certain of its effects is as a catalyst for triggering interpersonal peer
communication leading to changes in the social discourse of the audience
(Storey, 1995), and in motivating receivers to talk to each other about
things they learned from the entertainment-education message (Rogers, 1995).
Group discussion between the health educator and the participants in the
present study is a key component of the communication process. The role of
the educator as facilitator or group discussion leader, as well as design
of the study as a field experiment, was guided by Lewinian theory.
A social psychologist, Kurt Lewin is noted for founding research and
training on group dynamics. Like the other "founding fathers" (Rogers,
1993) of communication science, Lewin did not perceive himself as studying
communication processes, although his work in group dynamics is considered
seminal in the field.
Lewin's field experiments at the University of Iowa are classics in
communication research. His "sweetbread studies," designed to increase
consumption of so-called glandular meats during World War II, revealed
startling differences between behaviors of members in lecture meetings
versus those in discussion groups led by a researcher particularly skilled
in group leadership. In a follow-up survey done several weeks after the
meetings, only 3 percent of Iowa housewives in the lecture meetings
actually served sweetbreads, whereas 32 percent of the discussion group
members served sweetbreads (Lewin, 1958). It was suggested that the
difference in the effects of the two experimental treatments was due to the
higher level of involvement within the discussion condition, where
housewives took an active part by asking questions and sharing experiences
with the group. The sweetbread study became a classic in communications
research in spite of its several shortcomings.
Generalizing results of the sweetbreads study and other researches
conducted at Iowa, Lewin developed a three-step procedure for changing
behavior which he referred to as (1) unfreezing; (2) moving, and
(3) freezing the new behavior. Because behavior which has been changed is
frequently reverted back to by individuals, the influence of other groups
members is extremely important in both changing and maintaining
behaviors. Lewin (1943) observed: "The group decision had a 'freezing'
effect for future action" (quoted by Rogers, 1993).
The design of the present study incorporates the essence of Lewin's group
discussion condition. All three groups received the breast care education
program in a warm, friendly atmosphere encouraging interaction and
discussion. The leader was a Spanish-speaking woman, trained to make
participants feel comfortable discussing a topic as gender-sensitive as
breast health. Because all discussions were conducted in Spanish, and the
questionnaires were in Spanish, Latina participants felt ownership of the
educational program, displaying no intimidation or hesitation to ask
questions, offer comments or share experiences. The leader asked
participants in all groups for a public commitment through a show of hands
to practice breast self-examination and to get their mammograms.
In designing the present study's two-minute video, the "trigger film"
concept suggested by Slote (1993) was utilized. Slote suggested the
usefulness of the short "trigger" video (three to five minutes) as an
attention-getting device only, not as the primary information vehicle. The
trigger video was designed to generate audience attention and participation
through an abbreviated but highly visual presentation, which acted as a
prelude to a live instructor, who then elaborated and expanded on the
information presented in the video. The trigger video structure fits well
with Johns Hopkins' strategy of presenting a live instructor or educator
immediately after the presentation to facilitate discussion and questions
from the audience.
The limits presented by the presentation structure in the present study
suggest construction of a video which incorporates concepts from three
strategies: (1) production components and scripting structure from
Sabido's soap operas; (2) editing and format from Slote's concept of a
brief, highly visual "trigger" presentation; (3) epithet designed to
facilitate immediate interaction between a credible educator and the
audience, from Johns Hopkins' one-shot strategy.
Drawing upon these theoretical perspectives to guide development and field
testing of communication, theoretical hypotheses were developed for the
present study.
Theoretical and Empirical Hypotheses
The present study hypothesized increase in five measures of the
effectiveness of the video and a one-week follow-up reinforcement interview
by the health educator. The theoretical and empirical hypotheses are:
(1) Theoretical: Exposure to the entertainment-education video message
increases knowledge about breast cancer.
Empirical: Members of Groups #2 (video only) and #3 (video plus follow-up)
will have a greater increase in levels of breast cancer knowledge from T-1
to T-3 than will members of Group 1 (control).
(2) Theoretical: Exposure to the entertainment-education video message
increases knowledge about mammography.
Empirical: Members of Groups #2 (video only) and #3 (video plus follow-up)
will have a greater increase in levels of breast cancer knowledge from T-1
to T-3 than will members of Group 1 (control).
(3) Theoretical: Exposure to the entertainment-education video message
increases positive attitudes about breast cancer.
Empirical: Members of Groups #2 (video only) and #3 (video plus follow-up)
will have a greater increase in positive attitudes toward breast cancer
from T-1 to T-3 than will members of Group #1 (control).
(4) Theoretical: Exposure to the entertainment-education video message
increases discussion of breast cancer with friends and family members.
Empirical: Members of Groups #2 (video only) and #3 (video plus follow-up)
will discuss breast cancer with more friends and family members than will
members of Group #1 (control).
(5) Theoretical: Exposure to the entertainment-education video message
increases mammography compliance behavior.
Empirical 5a: Members of Group #3 (video plus follow-up) will have a
greater increase in mammography compliance from T-1 to T-3 than will
members of Groups #1 (control) and #2 (video only).
Empirical 5b: Members of Group #3 (video plus follow-up) will have a
greater increase in mammography compliance from T-1 to T-3 than will
members of Group #1 (control).
Empirical 5c: Members of Group #3 (video plus follow-up) will have a
greater increase in mammography compliance from T-1 to T-3 than will
members of Group #2 (video only).
Empirical 5d: Members of Group #2 (video only) will have a greater
increase in mammography compliance from T-1 to T-3 than will members of
Group #1 (control).
Empirical 5e: Members of Group #2 (video only) and Group #3 (video plus
follow-up) will have a greater increase in mammography compliance from T-1
to T-3 than will members of Group #1 (control).
Figure 1 is a summary of the hypothesized effects of the
entertainment-education video and the follow-up reinforcement interview.
Method
The entertainment-education intervention was evaluated via an experimental
design involving 175 Latinas, age 35 to 71, randomly divided into three
groups, who attended a one-hour breast health education program presented
in Spanish by a bilingual Latina health educator (Figure 2). The groups
included:
Group 1: A control group of 58 Latinas who received the educational program
with no entertainment education video or reinforcement interview.
Group 2: A group of 57 Latinas who received the educational program with
the entertainment-education video only.
Group 3: A group of 60 Latinas who received the educational program with
the entertainment-education video plus a reinforcement interview via
telephone one week after the educational program
Each respondent was asked to complete a pre-test questionnaire and a
post-test questionnaire. Groups 1 and 2 were administered a follow-up
questionnaire by telephone one week after the educational program. Group 3
received a reinforcement interview by telephone one week after the
educational program, and were administered a follow-up questionnaire by
telephone one week after the reinforcement interview. A total of 15
programs were presented at four different sites in two Southern California
counties over a ten-day period.
A total of 374 respondents received the educational program. Of this
number, 199 were eliminated for the following reasons: 131 were under age
35; 30 did not have a telephone number written on their questionnaire; 11
did not answer the telephone after three attempts to complete a follow-up
questionnaire; 27 submitted incomplete written questionnaires. Respondents
under age 35 were eliminated from the present study because breast cancer
screening guidelines do not recommend mammograms for women under 36.
Data Analysis
All measures were calculated for the three groups (#1: control, #2: video
only, and #3 video plus follow-up interview) at three points in
time. Statistical significance of the differences in mean change scores
for the three groups, across the three points in time, were calculated by
t-tests, using an alpha of 5 percent. Each of the empirical hypotheses
were to be accepted or not accepted on the basis of statistical significance.
Findings
Because the video intervention and the reinforcement interview intervention
in the present study were embedded within a one-hour educational program,
effects were anticipated in all three groups, including the control group
(Group #1), which received neither of the two interventions of
study. Pre-post change values were calculated. The overall change, that
is the change from T-1 to T-3, was used as the variable for testing the
hypotheses. A t-test was calculated to determine the statistical
significance of the differences in mean change scores using an alpha of 5
percent. Empirical hypotheses were supported or not supported on the basis
of statistical significance.
Empirical Hypothesis 1: Members of Groups #2 (video only) and #3 (video
plus follow-up) will have a greater increase in levels of breast cancer
knowledge from T-1 to T-3 than will members of Group 1 (control).
Results revealed that the mean change scores for breast cancer knowledge
from T-1 to T-3 for Groups #2 and #3 (M = .45) were greater at a
statistically significant level than the mean change scores for breast
cancer knowledge for Group #1 (M = .21) (Table 1). The t value of 2.01 is
greater than the one-tailed tabular value of t of 1.6, with 173 degrees of
freedom. Empirical Hypothesis 1 was supported.
Empirical Hypothesis 2: Members of Groups #2 (video only) and #3 (video
plus follow-up) will have a greater increase in levels of breast cancer
knowledge from T-1 to T-3 than will members of Group 1 (control).
Results revealed that the mean change scores for mammography knowledge
from T-1 to T-3 for Groups #2 and #3 (M = .50) were greater at a
statistically significant level than the mean change score for breast
cancer knowledge for Group #1 (M = .16) (Table 2). The t value of 2.96 is
greater than the one-tailed tabular value of t of 1.658, with 173 degrees
of freedom. Empirical Hypothesis 2 was supported.
Empirical Hypothesis 3: Members of Groups #2 (video only) and #3 (video
plus follow-up) will have a greater increase in positive attitudes toward
breast cancer from T-1 to T-3 than will members of Group #1 (control).
Results revealed that the mean change scores for positive attitudes toward
breast cancer from T-1 to T-3 for Groups #2 (video only) and #3 (video plus
follow-up) (M = .17) were not greater at a statistically significant level
than the mean change score for positive attitudes toward breast cancer for
Group #1 (M = .17) (Table 3). The t value of .68 is less than the
one-tailed tabular value of t pf 1.658 with 170 degrees of
freedom. Empirical Hypothesis 3 was not supported.
Empirical Hypothesis 4: Members of Groups #2 (video only) and #3 (video
plus follow-up) will discuss breast cancer with more friends and family
members than will members of Group #1 (control).
Results revealed that the mean change scores for discussion with more
friends and family members from T-1 to T-3 for Groups #2 (video only) and
#3 (video plus follow-up) (M = .04) were greater at a statistically
significant level than the mean change score for discussions with more
friends and family members for Group #1 (M = .09) (Table 4). The t value
of 2.39 is greater than the one-tailed tabular value of t of 1.658, with
173 degrees of freedom. Empirical Hypothesis 4 was supported.
Empirical Hypothesis 5a: Members of Group #3 (video plus follow-up) will
have a greater increase in mammography compliance from T-1 to T-3 than will
members of Groups #1 (control) and #2 (video only).
Results revealed that the mean change score for mammography compliance from
T-1 to T-3 for Group #3 (video plus follow-up) (M = .52) was not greater
at a statistically significant level than the mean change scores for
mammography compliance for Groups #1 and #2 (M = .44) (Table 5). The t
value of .45 is less than the one-tailed tabular value of t of 1.658, with
167 degrees of freedom. Empirical Hypothesis 5a is not supported.
Empirical Hypothesis 5b: Members of Group #3 (video plus follow-up) will
have a greater increase in mammography compliance from T-1 to T-3 than will
members of Group #1 (control).
Results revealed that the mean change score for mammography compliance from
T-1 to T-3 for Group #3 (video plus follow-up) (M = .52) was not greater at
a statistically significant level than the mean change score for
mammography compliance for Group #1 (M - .31) (Table 6). The t value of
.96 is less than the tabular value of t of 1.658, with 110 degrees of
freedom. Empirical Hypothesis 5b is not supported.
Empirical Hypothesis 5c: Members of Group #3 (video plus follow-up) will
have a greater increase in mammography compliance from T-1 to T-3 than will
members of Group #2 (video only).
Results revealed that the mean change score for mammography compliance from
T-1 to T-3 for Group #3 (video plus follow-up) (M = .52) was not greater at
a statistically significant level than the mean change score for
mammography compliance for Group #2 (M =.56) (Table 7). The t value of .21
is less than the tabular value of t of 1.658 with 110 degrees of
freedom. Empirical hypothesis 5c is not supported.
Empirical Hypothesis 5d: Members of Group #2 (video only) will have a
greater increase in mammography compliance from T-1 to T-3 than will
members of Group #1 (control).
Results revealed that the mean change score for mammography compliance from
T-1 to T-3 for Group #2 (video only) (M = .56) was not greater at a
statistically significant level than the mean change score for mammography
compliance for Group #1 (M = .31) (Table 8). The t value of 1.29 is less
than the single-tailed tabular value of t of 1.658 with 109 degrees of
freedom. Empirical hypothesis 5d is not supported.
Empirical Hypothesis 5e: Members of Group #2 (video only) and Group #3
(video plus follow-up) will have a greater increase in mammography
compliance from T-1 to T-3 than will members of Group #1 (control).
Results revealed that the mean change scores for mammography compliance
from T-1 to T-3 for Groups #2 (video only) and #3 (video plus follow-up) (M
= .54) was not greater at a statistically significant level than the mean
change score for mammography compliance for Group #1 (M = .31) (Table
9). The t value of 1.28 is less than the single-tailed, tabular value of t
of 1.658, with 165 degrees of freedom. Empirical hypothesis 5e is not
supported.
Figure 3 is a summary of the findings for hypothesized effects of the
entertainment-education video and the follow-up reinforcement interview.
Conclusions
Data analysis revealed that while the entertainment-education video and
follow-up reinforcement interview produced significant effects in knowledge
of both breast cancer and mammography, they were not as effective in
producing significant positive attitudes about breast cancer and
mammography compliance.
Several implications are suggested by these findings. First, the
experimental treatments increased knowledge, however knowledge may not
necessarily lead to attitude change or behavior change, at least not
immediately. Rogers (1991) suggests preventive behaviors are difficult to
diffuse because rewards are distant in time and not assured. The same
principle may apply to early detection behavior, that is, successful
diffusion of breast cancer and mammography information may not be
sufficient to produce attitudinal and behavioral change, particularly when
the behavior may result in one or all of several consequences, all of which
are unpleasant:
(1) She may find that she has a dreaded disease
(2) She may find that the behavior is painful
(3) She may find that the behavior is embarrassing
(4) She may find that the behavior is inconvenient
(5) She may find that the behavior is costly
Evaluation research on attempts to diffuse family planning and preventive
health behaviors in many Third World countries reveal a similar situation,
that is, knowing what to do does not necessarily translate to doing it
(Rogers 1973).
Findings regarding increased knowledge and behavioral change in the present
study are consistent with typical findings in evaluation research for
entertainment-education programs. In general, sharp increases in knowledge
levels are achieved, however change in behavior does not necessarily
follow, particularly in the short term (Rogers 1991).
The inverse relationship between knowledge and behavior change in both the
present study and past entertainment-education studies suggests
reinforcement of the notion that cultural norms are stronger influences on
behavior than knowledge among minority members of society. This notion has
strong implications for content and focus of culturally sensitive
communication.
A more effective entertainment-education video might portray characters as
positive role models who receive rewards for their positive attitudes and
behavior, as suggested by social modeling behavior. That is, the Hermanos
character who was afraid to get her mammogram should instead by portrayed
as a positive role model who finally conquered her fear, got her mammogram,
and found that it was (1) merely uncomfortable, not painful; (2) not that
inconvenient since her sister, her mother and her friends provided support;
(3) not that costly, as she found a low-cost mammogram available; (4) that
the peace of mind she felt was very rewarding; and (5) that she now
realizes that what she should fear more is not knowing whether or now she
has breast cancer.
The significant increase in discussions about breast cancer with friends
and family members also suggests a possible first step toward diffusing
early detection behavior. A negative attitude toward breast cancer could
lead to negative behavior regarding its detection. In carrying the
positive role model notion a step further, perhaps the Hermanos character
should reveal that she finally gained the courage to get her mammogram
because she discussed it with her best friend or her sister or her
daughter, and that she is now going to discuss her positive experience with
another of her friends or perhaps her sister-in-law.
There were a number of limitations in the present study. The first
involved the size and homogeneity of the intact groups. Because specific
treatments were randomly assigned at each site to include the video
(respondents in Group 2), or not to include the video (respondents in Group
1), or to include the video and a follow-up reinforcement interview
(respondents in Group 3), the author could not control which type of
educational program would be offered at which time of day; that is, even if
the author knew the early-morning program would attract a much larger
number of respondents, or a much older group of respondents, she could not
divide the group or change the type of educational program to balance out
the number or respondents in the three groups. Programs at two of the
sites were advertising via flyers, and attendance was voluntary. While
flyers asked women to telephone for reservations, only two reservations
were received for one education program, and no reservations were received
for the other. While attendance was mandatory for respondents at the site
involving employees of a school district, the women could choose which of
the three educational programs offered they wished to attend. An
inordinately large number of these women were in the control group because
the program with no intervention was randomly selected for the early
morning session, and most women wanted to attend early-morning programs so
they could go home as soon as possible.
This intact group, whose members tended to be more affluent and
better-educated, made up the bulk of the control group not only because of
their large numbers, but also because few were eliminated due to age (most
were over 35) or on the basis of not having a telephone (all had
telephones) or because of difficulty in reaching them for the T-3 follow-up
reinforcement interview (the majority were available at the telephone
number and at the time indicated on their questionnaire). While an equally
large number attended the next class, just after lunch, that group was not
skewed because many of the mid-day program respondents were eliminated
because they were under 35.
The homogeneity of intact groups tended to exclude large numbers of
respondents also. For example, a good many respondents at two of the sites
tended to be younger, therefore they were eliminated because they were
under 35. Few were eliminated from the school district group because of
age, since most were over 35. Also, many respondents at the county
hospital site were eliminated because of low literacy levels. A number
could not read the questionnaire, and others took so long to complete the
pre-test questionnaire that they simply stopped working on it once the
educational program began. Many of the county hospital participants were
also eliminated because they had no telephone, or because they were no
longer available at the telephone number they wrote on the questionnaire at
the time of the T-3 follow-up, or because they were suspicious and afraid
to talk to the educator when she called for the T-3 reinforcement interview.
One of the most significant limitations of the present study was the
homogeneity of all groups, particularly the control group. The intact
group for the morning educations program at the school district dominated
the control group because of its large number. Additionally, because most
members of the control group came from the same sight, and tended to be
well-informed compliers with a positive attitude about breast cancer, their
scores at T-1 were higher than scores for one of the other groups at
T-3. The high scores on all measures for the control group at T-1 resulted
in a ceiling effect, wherein little variance within the group was possible
since so many members scored so high at T-1. This intact group phenomena
could have been addressed if logistics has permitted presentation of three
programs simultaneously at each site, allowing respondents to be randomly
assigned as individuals, not as groups.
Another limitation to be addressed as a contributor to the lack of
significance in behavioral change is the fact that the
entertainment-education video in the present study was a "one-shot"
application of the entertainment-education strategy. In nearly all
programs utilizing an entertainment-education strategy, interventions were
repetitive, broadcast daily and/or weekly. The repetition of the format
was a strong reinforcement to influencing behavior (Rogers 1991).
Secondly, the present study was limited by the necessity to word
questionnaire items in the simplest possible way in order to accommodate
respondents with relatively low literacy levels. The questions measuring
knowledge were so basic that more-educated women scored extremely high on
the pre-test, leaving little room for variance over time. By contrast, the
county hospital respondents struggled with their responses, and some
actually showed a decrease in knowledge.
Shortcomings of the present study include the tendency of many respondents,
particularly minority respondents, to report the answers they perceived
that the educator wanted. This response-set program was compounded when
the post-test questionnaire was completed, as many of the women became
quite fond of the health educator during the friendly, interactive
discussion. Additionally, many respondents exhibited anxiety and
embarrassment in completing the questionnaire, viewing it as a "test" of
sorts, for which they might now know the right answers.
Future Research
The present study reveals several areas for future research. More research
is needed on the use of the entertainment-education strategy in "one-shot"
applications such as that used in the present study. Researchers have
little experience in using this strategy in settings other than via the
mass media. Findings in the present study suggest the strategy applied in
a "one-shot" setting can be effective in increasing knowledge among
minority women. The next step is to explore different scenarios directed
to specific audiences. Additionally, research is needed to determine
whether use of recognizable soap opera stars in an entertainment-education
video would influence behavior, specifically in casting a well-known,
popular star as a positive role model who gets her mammogram regularly.
Additionally, innovative new ways should be explored to combine mediated
entertainment-education interventions and interpersonal communication of
culturally sensitive messages. Past efforts to communicate bilingual
rather than bicultural messages via mass media have proven to be
ineffective. Findings from the present study suggest more emphasis on the
key role played by a bilingual educator who is perceived as a warm,
friendly person who understands and cares, not an authoritative, impersonal
expert who simply communicates behaviors that should be performed.
More research is needed in understanding the power of media-stimulated peer
communication. The breast cancer educator/facilitator should encourage
interpersonal discussion both within and outside the group, particularly in
relation to the intervention. It is suggested that the title of the
intervention, Hermanas (sisters), is an excellent way to encourage audience
members to think about and to discuss mammography compliance with female
family members and friends. We are all sisters and we need to care about
each other and help each other to do all we can to protect ourselves from
breast cancer.
More research is needed to determine appropriate communication messages for
specific audiences, i.e., higher SES level women versus lower SES level
women. Specifically, should higher SES women, particularly those who are
compliers (women who get mammograms regularly) receive communication
messages urging them to disseminate the information they possess to other
women?
The notion of diffusing communication messages through friends and family
members should be explored. Relationship between compliers and the women
with whom they discuss breast cancer should be explored, keying on the
direction of the communication flow. For instance, do older compliers tend
to discuss breast cancer with their daughters, and thus influence their
decision to comply? Do younger compliers tend to discuss breast cancer
with mothers, grandmothers, aunts or friends who are not
compliers? Suggested research questions include whether the primary
audiences for culturally sensitive communication messages for minority
women should be opinion-leaders, or younger women. Compliance rates
clearly demonstrate that younger women tend to be more compliant, with
compliance rates declining with age. Older women's compliance must be
addressed, as breast cancer risk increases with age.
Researchers should also look to successful and unsuccessful preventive
behavior information campaigns. While messages created to stop people from
smoking may seem quite removed from messages created to motivate women to
get a mammogram, there are certain overlaps. Specific attention should
also be paid to extensive information programs for preventive behavior
among Latinos.
By all indications, breast cancer will remain a high-profile public issue
for some time to come, which means the disease will remain a frequent topic
of news stories in the mass media. Many of these news stories reinforce
negative attitudes toward breast cancer, through emphases on topics such as
celebrity women's diagnoses and new findings linking breast cancer
incidence to a full range of risk factors. This continual reinforcement of
negative information about breast cancer requires equally power
communication messages to cultivate positive attitudes toward breast health
and breast cancer and its early detection, particularly among lower SES and
minority women.
The time we finally convince all women that the major factor to fear about
breast cancer is not knowing you have it, is the time when communication
will contribute to saving thousands of women's lives.
Bibliography
Backer, T., Rogers, E., Sopory. P (1992) Designing Health Communication
Campaigns: What Works? Newbury Park, CA: Sage.
Bandura, A. (1977), Social Learning Theory, Englewood Cliffs NJ: Prentice-Hall
Bandura, A. (1997). Self-efficacy: The exercise of control. New
York: Freeman
Bently, E. (1967), The Life of Drama. New York: Antheneum.
Centers for Disease Control (2002) National Center for Health Statistics
Centers for Disease Control (2002) Use of Mammography, Health, United States
Horton, D., & Wohl, R. R. (1956). Mass communication and parasocial
interaction: Observation on intimacy at a distance. Psychiatry, 19(3),
215-229.
Katz, E., Gurevitch, M., and Haas, H. (1873), On the use of the mass media
for important things. American Sociological Review, 38:164-181.
Lewin, K. (1947) Frontiers in group dynamics: I. Challenge of group
life: social planning and action research Human Relations. 143-153
Lewin, K. (1958), Group decision and social change. In E. E. Maccoby, T.
M. Newcomb and E. L. Hartley (eds.), Readings in Social Psychology, 3rd
Ed., 197-211. New York: Holt, Rinehart and Winston.
MacLean, P. D. (1973), A Tribute Concept of the Brain and
Behavior. Ontario Mental Health Foundation. Toronto, Canada: University
of Toronto Press. P. 2-22.
Magnus, H. (Ed.) (1982) Trends in Cancer Incidence. Washington,
D.C.:Hemisphere.
McLeod, J., Bybee, C., and Durall, J. (1982), Evaluating media performance
by gratifications sought and received. Journalism Quarterly, 59:3-12.
Nariman, H. N., (1993), Soap Operas for Social Change: toward a Methodology
for Entertainment-Education Television. Westport, Conn: Praeger.
Rogers, E. M. (1973), Communication Strategies for Family Planning. New
York: Free Press.
Rogers, E. M. (1986), Communication Technology. New York: Free Press
Rogers, E.M., Aikar, S., Chang, P., & Sopory, P. (1989), Proceedings from
the conference on entertainment-education for social change. Los Angeles,
CA: University of Southern California, Annenberg School for Communication.
Rogers, E. M. (1991), Communication campaigns to change health-related
lifestyles; Paper presented to the XIV World Conference on Health
Education, Helsinki, Finland, June 16-21.
Rogers, E. M. (1993) Organizational Factors in Health Campaigns. Newbury
Park, CA.: Sage.
Rogers, E.M. (1995) Diffusion of innovations (4th ed.). New York: Free Press.
Rubin, A. M. & Rubin, R.C. (1985) Interface of personal and mediated
communication: A research agenda. Critical Studies in Mass Communication,
2, 36-53.
Rubin, A. (1986), Uses, gratification and media effects research. In J.
Bryant and D. Zillmann (Eds.), Perspectives on Media Effects. Hillsdale,
NJ: Lawrence Erlbaum.
Rubin, R.C., & McHugh, M. P. (1987) Development of parasocial interaction
relationships. Journal of Broadcasting & Electronic Media, 31, 279-292.
Shannon, C. & Weaver, W. (1949), The Mathematical Theory of
Communication. Urbana, University of Illinois Press.
Singhal, A. (1990) Entertainment-Education Communication Strategies for
Development. Unpublished doctoral dissertation. Annenburg School of
Communication University of Southern California, August.
Singhal, A., Obregon, R. & Rogers, E.M. (1994) Reconstructing the story of
Simplemente Maria, the most popular telenovela in Latin America of all
time. Gazette, 54, 1-15.
Singhal, A, Rogers, E.. (1999) Entertainment-Education: A Communication
Strategy for Social Change. Mahwah, NJ: Lawrence Erlbaum Associates.
Singhal, A., Cody, M., Rogers, E., & Sabido, M. (2003), (Eds.),
Entertainment-education and Social Change: History, Research, and
Practice. Mahwah, NJ, Lawrence Erlbaum Associates.
Slote, A. (1993) Personal telephone interview by the author, April.
Storey, D. (1995, May) Entertainment-education, popular culture, and the
sustainability of health communications: Lessons from Indonesia and
Pakistan. Paper presented at the International Communication Association
Conference, Albuquerque, NM.
Tan, A. (1985), Mass Communication Theories and Research. New York: Macmillan.
Valente, T.W., & Bharath, U. (1999). An evaluation of the use of drama to
communicate HIV/AIDS information. AIDS Education and Prevention, 11, 203-211.
Valente, T.W., Poppe, P.R., Alva, M.E., deBriceno, R. V. & Cases, D.
(1995). Street theater as a tool to reduce family planning
misinformation. International Quarterly of Community Health Education, 15,
279-289.
Treatment Outcomes
Intervention/Interventions
) EH1 Knowledge of breast cancer scale
Exposure to trigger video and) EH2 Knowledge of mammography scale
reinforcement interview ) EH3 Positive attitude toward
breast cancer scale
Groups #2 and #3 ) EH4 Discuss breast cancer with friends
v.s. Group #1 (control) and family members
Exposure to trigger video and )
reinforcement interview )
) EH5a
Group #3 v.s. Groups #1 )
and #2 (control) )
Mammography compliance
Exposure to trigger video and )
reinforcement interview )
) EH5b
Group #3 v.s. #1 (control) )
Exposure to trigger video and )
reinforcement interview )
) EH5c
Groups #3 v.s. 2 (video only) )
Obtain a mammogram or report
Intention to get a mammogram
Exposure to trigger video )
) EH5d
Group #2 v.s. #1 (control) )
Exposure to trigger video and )
Reinforcement interview )
) EH5e
Groups #2 and #3 v.s. )
#1 (control) )
Figure 1
Diagram of the hypothesized effects of an entertainment-education video
plus follow-up reinforcement interviews
Reinforcement T3 Follow-Up
Group T1 Pre-Measure T2 Post-Measure Interview Measure
1: Control O1 O2 O3
2: Video only O1, X1 O2 O3
3: Video & Follow-
Up Interview O1, X1 O2 X2 O3
O1 = Pre-test questionnaire
O2 = Post-test questionnaire
O3 = Follow-up telephone questionnaire, 1 weeks after the education program
for
Groups 1 and 2, and 1 week after the reinforcement interview for
Group 3
X1 = Entertainment-education video intervention
X2 = Reinforcement interview urging compliance
Figure 2
Study Design
Table 1
Mean change scores in breast cancer knowledge from T1 to T3 for
Groups 2 and 3 versus Group 1
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(breast cancer _ Grps 2/3 to in mean scores 5% level?
Knowledge) Grp 1
Group 1: 0.21 0.59
Control (N=58)
0.24 2.01 Yes
Group 2:
Video Only N=57
Group 3:
Video + Follow-up Interview N=60
0.45 0.80
Range: 2 to 8, df=173.
Table 2
Mean change scores in mammography knowledge from T1 to T3 for
Groups 2 and 3 versus Group 1
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(mammography _ Grps 2/3 to in mean scores 5% level?
knowledge) Grp 1
Group 1: 0.16 0.52
Control (N=58)
0.34 2.96 Yes
Group 2:
Video Only N=57
Group 3:
Video + Follow-up Interview N=60
0.50 0.80
Range: 3 to 12, df=173.
Table 3
Mean change scores in positive attitudes toward breast cancer
from T1 to T3 for Groups 2 and 3 versus Group 1
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(positive attitude _ Grps 2/3 to in mean
scores 5% level?
about breast cancer) Grp 1
Group 1: 0.17 0.54
Control (N=58)
0 0.068 No
Group 2:
Video Only N=57
Group 3:
Video + Follow-up Interview N=60
0.17 0.73
Range: 2 to 8, df=170.
Table 4
Mean change scores for discussions about breast cancer with family members
and friends
from T1 to T3 for Groups 2 and 3 versus Group 1
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(discussions of _ Grps 2/3 to in mean scores 5%
level?
breast cancer) Grp 1
Group 2:
Video Only N=57
Group 3:
Video + Follow-up Interview N=60
0.04 0.04 0.13
0.05 2.39 Yes
Group 1: 0.09 0.12
Control N=58
Range: 0 to 6, df=173.
Table 5
Mean change scores in mammography compliance
from T1 to T3 for Group 3 versus Groups 1 and 2
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(mammography _ Grps 1/2 to in mean scores 5% level?
compliance) Grp 3
Group 3: 0.52 1.22
Video plus follow-
up interview N=60
0.08 0.45 No
Group 2:
Video Only N=57
Group 1:
Control N=58
0.44 1.04
Range: 1 to 4, df=167.
Table 6
Mean change scores in mammography compliance
from T1 to T3 for Group 3 versus Group 1
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(mammography _ Grp 3 to in mean scores 5% level?
compliance) Grp 1
Group 3: 0.52 1.22
Video plus follow-
up interview N=60
0.21 0.96 No
Group 1: 0.31 1.05
Control N=58
Range: 1 to 4, df=110.
Table 7
Mean change scores in mammography compliance
from T1 to T3 for Group 3 versus Group 2
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(mammography _ Grp 2 to in mean scores 5% level?
compliance) Grp 3
Group 3: 0.52 1.22
Video plus follow-
up interview N=60
0.4 0.21 No
Group 2: 0.56 1.01
Video only N=57
Range: 1 to 4, df=110.
Table 8
Mean change scores in mammography compliance
from T1 to T3 for Group 2 versus Group 1
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(mammography _ Grp 2 to in mean scores 5% level?
compliance) Grp 1
Group 2: 0.56 1.01
Video only N=57
0.25 1.29 No
Group 1: 0.31 1.05
Control N=58
Range: 1 to 4, df=109.
Table 9
Mean change scores in mammography compliance
from T1 to T3 for Groups 2 and 3 versus Group 1
Mean Change Difference in t-test for
Group Score Std. Dev. Mean scores difference Significant at the
(mammography _ Grp 2/3 to in mean scores 5% level?
compliance) Grp 1
Group 1: 0.31 1.05
Control N=58
0.23 1.28 No
Group 2:
Video Only N=57
Group 3:
Video + follow-up interview N=60
0.54 1.12
Range: 1 to 4, df=165.
|
