Sheila Murphy

The primary focus of her research has been investigating how message factors (such as framing, channel, affective versus noneffective information), individual level factors (such as ethnicity, gender, self-efficacy, identification, and involvement) and cultural level factors (e.g., social norms and beliefs) impact decision-making. For the past 10 years, her work has primarily focused on health-related decisions and on the role of narrative or storytelling in shaping the public's knowledge, attitudes and practices. Below describes several related areas of her ongoing research:

Transformative R01: "Transforming Cancer Knowledge, Attitudes and Behavior through Narrative."

Humans are innate storytellers. This statement is supported by research on cognitive processing, anthropology, and our own experience. For instance, if asked, "What made Pinocchio's nose grow longer?" most of us can easily recall the correct answer, namely lying. We can easily access this information despite the intervening decades since we learned or even thought about this particular tale. But when asked to recall non-narrative information from that same time period that may have been much more central to our lives such as "What was the name of your second grade teacher?" we are often at a loss. The power and perseverance of a narrative or story structure has been recognized and utilized for thousands of years. Yet when it comes time to craft health messages designed to convey crucial, potentially life saving health information, Western medicine all but ignores the use of narrative. Rather this vital information is typically presented in a list of risk factors, recommended prevention steps, symptoms, and treatment options.

In 2010, Lourdes-Baezconde-Garbanti (from the Institute of Health Promotion and Disease Preventive Medicine at USC's Keck School of Medicine) and she applied for and were awarded a prestigious Transformative R01 grant from the National Institutes of Health. Transformative R01s are designed to "accelerate the current pace of discovery" and "challenge the status quo with innovative ideas." The purpose of this 5-year grant is to challenge the underlying assumption that the traditional recitation of the facts is the optimal way to convey health-related information. To do so, she and others brought together an interdisciplinary team from across the University of medical researchers, script writers, cinematographers, physicians, psychologists, communication scholars, and public health professionals in an effort to re-examine and reinvent how health-related information is conveyed.

To empirically test whether utilizing a narrative format might produce a more substantial and more sustained impact on knowledge, attitudes and prevention behavior compared to a non-narrative format, her team produced two short films each 11 minutes in length and both containing the same 18 facts regarding cervical cancer prevention, detection and treatment. The Tamale Lesson conveys facts regarding the cause of cervical cancer, as well as how to prevent it (via the HPV vaccine) and detect it (via Pap tests) using the Romeo family's preparation for their youngest daughter's Quinceanera or 15th birthday as the narrative vehicle. The non-narrative film, It's Time, contains the same 18 facts but uses a more traditional approach featuring doctors, patients, facts and figures. Click here for more information on the study.

The relative efficacy of these two films is currently being tested by first surveying 1200 randomly selected women (300 African America, 200 European American, 200 Korean American and 300 Mexican American) to establish a pretest baseline level of cervical cancer-related knowledge, attitudes and behavior, randomly assigning these same women to receive either the narrative or non-narrative film and then resurveying them 2 weeks and then 6 months later. This large scale quasi-experimental study will allow them to detect not only a main effect of narrative vs. non-narrative on women's cervical cancer-related knowledge, attitudes and behavior, but also how these are moderated by key demographic factors such as age, income and ethnicity as well as more theoretical factors such as identification with specific characters and feeling "transported" or engrossed in the narrative. Importantly, this research also re-examines the "one-size-fits all" assumption that guides most health messages and instead looks at the role of culture and acculturation in acceptance of cancer messages. The data collection comparing the narrative and non-narrative films will be finished by Fall 2012. Next they will examine the effect of communication modality by replacing the film study described above using a print version of the narrative and non-narrative materials.

Although this research focuses on cervical cancer, the results have clear implications for virtually all health care communication. If successful, this research could radically change how health messages are conveyed. To quote one NIH reviewer, "This research isn't just thinking outside the box, it's blowing up the box."

Theoretical Exploration of Narrative Impact.

Popular primetime shows each reach a domestic audience of between 10 to 20 million viewers each week (and ultimately far more in syndication and internationally). To test the impact of specific health-related storylines she has worked with Hollywood, Health and Society and the television networks to identify health issues depicted in upcoming episodes. Having advance notice of health storylines allows Murphy and her team to use pretest/post-test surveys to assess the impact of such narratives on diverse audiences and identify characteristics of both the stories and the viewers that increase or attenuate their impact.

For the past nine years, Murphy and her students have tried to identify the factors that make a narrative more or less persuasive. Recently they have focused on three theoretical constructs - identification with a specific character, transportation into the narrative, and the emotion evoked by the narrative. Research demonstrates that lasting behavior change relies on the use of characters with whom viewers can identify. In short, viewers appear to learn more from models - in this case, fictional television characters - with whom they identify, like, feel as if they know, or perceive to be similar to themselves (Bandura, 2004). In addition, research by Green and Brock (2000, 2002) has revealed greater attitude change among those who reported being "transported" into the narrative world. It is argued that because a person in a transported state is engrossed, having devoted his or her cognitive resources to the events playing out in the narrative, he or she may be less likely to counter argue or to critically assess the messages in the narrative (Slater, 2002).

Their analysis of 167 viewers of Desperate Housewives (Murphy, Frank, Moran, & Woodley, 2011) found that transportation of the sense of being absorbed in Lynette's lymphoma storyline was more important than identifying with the character of Lynette in increasing knowledge about lymphoma including symptoms, prognosis, treatment options, chemotherapy and screening behavior. Subsequently, she and her students looked at the impact of an episode of Law and Order: SVU "Witness" dealing with rape in the Congo on viewers' perceptions of Africa, the Congo, and attitudes about U.S. aid to those regions. Again, they found that transportation or being absorbed by the narrative was the primary determinant of having positive attitude toward U.S. aid to Africa (Murphy, Hether, Felt & de Castro Buffington, 2012).

Finally, as part of her Transformative R01 grant from the National Cancer Institute, they were able to encourage writers of a Spanish language telenovela, El Clon, to include a storyline dealing with the importance of screening for cervical cancer. Using a cross-sectional baseline/endline survey design, they measured the impact or storyline involving the fictional character of Dora, a woman in her late 40s, who experienced worrisome symptoms including pain and irregular vaginal bleeding. As the story unfolded, viewers were exposed to a number of key messages related to cervical cancer and the importance of screening. Baseline/endline comparisons revealed significant gains in both cervical cancer-related knowledge and behavior. Once again, the best predictor of behavior (interpersonal discussion, information seeking, planning to have a Pap test) was feeling "transported" or highly involved with the narrative (Murphy, Hether, Frank, de Castro Buffington & Baezconde-Garbanati, 2012). El Clon was viewed by 1.4 million viewers in the United States in 2010 and is currently being shown in over 20 countries internationally.

Entertainment Education in International Settings

Entertainment education or "the international placement of educational content in entertainment" (Singhal & Rogers, 2002, p.117) is practiced primarily in countries where the government has control over media content. For the past nine years Murphy has worked pro bono with the BBC World Service Trust [now known as BBC Media Action]-- the international non-profit charity of the British Broadcasting Corporation -- who "uses the power of popular media to reduce poverty and promote human rights". From 2002 to 2007, the BBC WST produced an entertainment education program, Jasoos Vijay, as the centerpiece of a larger campaign to raise awareness about HIV/AIDS and promote behavior change in India. Jasoos (Detective) Vijay was a weekly crime drama telecast on India's national television channel. The show aired 130 episodes over five years and was among the 10 most watched programs in India reaching over 70 million viewers. Two Annenberg PhD students, Jayee Chatterjee, Lauren Frank and Murphy conducted a pretest/post-test analysis of Jasoos Vijay (Chatterjee, Bhanot, Frank, Murphy & Power, 2009). Their analysis revealed that interpersonal discussion played a key role in changing HIV-related attitudes and behavior.

As a result of this research, the BBC WST designed a national campaign in India to "normalize" condom use and condom users by stimulating conversation among the general population. The "Condom Normalization" campaign resulted in an 8% increase in condom sales across India. Not only was the campaign effective in normalizing condom use and users but in 2010 it also won CNN's Public Service Announcement of the Year Award and a Bronze Palm at the Cannes Film Festival. Thus far this project has yielded three book chapters in a recently released 2011 book from Sage Drama, development, and cross-cultural translation and a peer-reviewed journal article in the Journal of Health Communication (Frank, Chatterjee, Chaudhuri, Lapansky, Bhanot & Murphy, 2012).

Barriers to Cervical Cancer Prevention in Hispanic Women: A multilevel approach.

Her second NIH-funded grant (Sandra Ball-Rokeach, Co-PI) takes the use of narrative in health communication to the next level by looking at naturally occurring health-related storytelling and the factors that may inhibit or facilitate health communication in a woman's social networks and community. This research recognizes that individuals do not operate in a vacuum but rather that our lives are inextricably intertwined with our family, friends and community. The goal of this 4-year study is to identify barriers and conduits to cervical cancer prevention, detection and treatment among Hispanic women at the individual, interpersonal (with an emphasis on social networks) and community (with an emphasis on storytelling networks and communication ecology) levels. To achieve this goal Murphy and her team are currently conducting 200 quantitative interviews (The Clinic Survey) with Hispanic women between 21-50 years of age recruited from Los Angeles County/USC women's clinic. Based on chart reviews of prior patients they anticipate that this sample should include approximately 450 women who receive adnormal Pap test results, chart reviews of prior patients suggest that approximately 300 will return for followup while the remaining 150, despite having been informed of their abnormal result, will not return to the clinic. The benefit of surveying women prior to their Pap tests is that they will be able to identify individual, interpersonal and community level predictors of abnormal Pap test results, followup and lack of followup. In addition we will also survey 300 women who have never been screened. Individual, interpersonal and community level factors will be analyzed in isolation, and more importantly, working together within a system.

One innovative aspect of the proposed research is that it extends Fishbein and Cappella's Integrated Model of Behavioral Prediction (which incorporates key elements of earlier behavioral prediction models, see Fishbein & Cappella, 2006) by including interpersonal networks and the community in which the individual lives as predictors of cervical cancer-related beliefs, knowledge and behavior. Moreover, replicating a procedure used successfully by The Metamorphosis Project, they will identify the strength of specific communities' health storytelling networks as played out through conversations generated by local or geoethnic media, community organizations, and interpersonal networks and assess both the current and potential level of health communication generated by each. Murphy and her team anticipate that their findings will inform practitioners and researchers not only about cervical cancer but also about the importance of taking a mutilevel approach to health communication more generally.

The USC Annenberg/ Hollywood, Health and Society Television Monitoring Project

Despite the proliferation of new forms of media, television continues to be the dominant medium in most people's lives. Moreover, in a random national survey two-thirds of Americans reported television as a primary source of health information (Beck, 2005) Because viewers rely heavily on television programs for health information, it is important to know what health messages are being conveyed either intentionally or unintentionally. In January 2003, Vicki Beck, the then director of Hollywood, Health and Society (HH&S) at the Norman Lear Center, Michael Cody, and Murphy launched a project to content analyze the health-related content of popular primetime television shows. It would have been relatively easy to identify, record and analyze the 10 primetime shows with the largest overall audiences during each Spring television season. But program preferences are far from random. Due to the ethnic differences in program preferences, Murphy and her team made the decision to stratify shows on the basis of programs for Nielsen's General, Hispanic and African American audiences (see Murphy, Hether & Ridout, 2008) The Television Monitoring Project continues to provide a yearly snapshot of the frequency of health-related content, specific health issues depicted, accuracy and educational value of the portrayals, presence of prevention and risk information and foods and beverages shown and consumed by characters, just to name a few of the over 100 variables coded.

Annenberg National Health Communication Survey (ANHCS)

Since January 2006, along with her Annenberg East colleagues, Bob Hornik and Joe Cappella, and her USC colleagues Margaret McLaughlin and Michael Cody, Murphy has been a CoPI on the Annenberg National Health Communication Survey (ANHCS), a nationally representative web-based survey of between 250 and 300 U.S. adults each month. They designed ANHCS to capture national trends in media exposure, health behavior, knowledge and beliefs. There are two parts to the ANHSC survey: the CORE is first 20 minutes of the ANHCS survey that contains items measuring media use and health behaviors, knowledge and beliefs as well as health policy and remains fairly constant over time to allow for analysis of change over time. The last 5 minutes of the ANHCS survey is essentially up for grabs. Faculty and graduate students at either Annenberg School can submit proposals for health-related items to be included as a module. Thus far, this data set has produced over two dozen faculty and student publications (e.g., Chen & Murphy, 2011; Murphy, Cody, Glik, Ang & Frank, 2009). Moreover, after two years the ANHCS is made public so that other researchers who are interested in health communication could use the data in their own research.