News Framing of Autism

Abstract

This study explores agenda setting, framing, and the concepts of media advocacy and mobilizing information through content analysis of The New York Times and The Washington Post news coverage of autism from 1996 to 2006, the year the Combating Autism Act was passed. Findings revealed that science frames decreased over time, while policy frames increased. Medical, government, family, and nonprofit sources were most common in news coverage. Solutions were mentioned more frequently than causes; however, mobilizing information was limited. Theoretical implications and practical applications are discussed.

Keywords

autism content analysis framing health policy media advocacy

Introduction

On December 19, 2006, then President George W. Bush signed the Combating Autism Act (CAA) into law, authorizing nearly $1 billion over 5 years to fund autism research and related activities. The Act provided a 50% increase over previous federal funding for autism and has been called the most comprehensive legislation for a single disease or disorder ever passed by Congress (The White House Archives, 2006). Specifically, the Act includes appropriations for autism research, diagnosis, treatment, and surveillance services through the Centers for Disease Control and Prevention (CDC) and the National Institutes of Health (The White House Archives, 2006).

Autism is a complex neurobiological disorder that is part of a group of developmental disabilities known as autism spectrum disorders (ASDs). Today, one child in 88 is diagnosed with an ASD (CDC, 2012a). Autism can occur among any race, ethnicity, or socioeconomic class, but it is almost 5 times more likely to affect males than females (CDC, 2012a). Autism can cause significant social, communication, and behavioral challenges; however, symptoms can vary greatly in terms of their nature and when they start and can range from very mild to severe (CDC, 2012b). There are no effective means to prevent autism and no known cure. However, early diagnosis and intervention may lead to various treatments and therapies designed to help develop communication, social, and cognitive skills (CDC, 2012b).

Although autism was first identified in the early 1940s, it was considered rare until recent decades when the incidence of autism began increasing at an alarming rate (Fombonne, 2003; Wing & Potter, 2002). Changes in diagnostic criteria, broadening acceptance of the autistic spectrum concept, and growing awareness among parents and medical professionals, as well as a possible “true increase” in incidence, led to a spike in autism diagnoses in the mid-1990s (Wing & Potter, 2002). Research on the causes of autism has focused on genetic and environmental factors, including the vaccine for measles, mumps, and rubella (MMR); however, none of the environmental factors, including the MMR vaccine, has been identified as the cause of autism (Wing & Potter, 2002).

The possibility of a link between autism and the MMR vaccine has caused controversy among medical and academic communities, as well as the general public, and it is important to consider this information when examining news coverage of autism. In February 1998, an article suggesting a potential link between autism and the MMR vaccine appeared in the British medical journal, The Lancet (Wakefield et al., 1998). Subsequent research refuted this association (Gerber & Offit, 2009), however, and the article was retracted after the study’s lead author, Andrew Wakefield, went under investigation for unethical and misleading research practices (Godlee, Smith, & Marcovitch, 2011). Despite the retraction, the supposed autism-vaccine link attracted considerable attention from parents, policy makers, and the general public and cast doubt on vaccine safety (Petts & Niemeyer, 2004).

Not surprisingly, media attention to autism has increased in accordance with this public controversy, rising incidence levels, and awareness and advocacy efforts. Analysis of more than 3,500 health stories in 48 different U.S. news outlets showed that autism was one of the top five diseases or disorders included in health news (Pew Research Center, 2008). As an article in JAMA: The Journal of the American Medical Association noted, all of this controversy and attention “has put autism on the public agenda” (Fombonne, 2003, p. 87). The article also acknowledged that autism research has received legitimate increases in public funding in recent years, yet connections between science, social needs, and funding policies for health issues still need to be explored (Fombonne, 2003).

Several theories help explain this increased attention and funding for autism. Agenda setting, which posits that the media influence what people think about, may explain how increased media attention helped push autism onto the public and political agendas, resulting in increased federal funding (Kim, Scheufele, & Shanahan, 2002; Kosicki, 1993; McCombs & Shaw, 1972; Rogers, Dearing, & Chang, 1991). News framing, which involves selection and emphasis of certain ideas or aspects of issues in news coverage to promote salience among the audience, may also be relevant to this phenomenon (Entman, 1993, 2007). Of course, agenda setting and framing are related, and much has been written about these communication processes (e.g., Maher, 2001; McCombs & Ghanem, 2001; Scheufele & Tewksbury, 2007; Takeshita, 2006; Weaver, 2007). The current study aims to add to existing research on agenda setting and framing by exploring the concept of media advocacy.

Media advocacy is another important concept that may help explain increased media coverage and subsequent policy changes related to autism. According to Wallack and colleagues (Wallack, 1990; Wallack, Dorfman, Jernigan, & Themba, 1993), media advocacy involves groups and organizations defining, identifying, and framing a problem to stimulate media coverage of the problem as a public health issue. More recently, media advocacy has been described as consisting of three steps: (a) setting the agenda, (b) framing the issue, and (c) advancing a solution (Gibson, 2010). Thus, media advocacy encompasses the ideas of both agenda setting and framing by activating “forces in a social system (that is, media coverage) to help stimulate public concern and action” (Glanz, 2002, p. 393). Although this study does not focus on specific strategic communication materials, such as press releases produced by health or autism organizations, media advocacy is worth exploring because of the way these efforts may have influenced media coverage of autism. Health and science reporters have long relied on such resources to inform their reporting (Len-Ríos et al., 2009), and it is important for communication practitioners of all types to understand how such “frame sponsorship” may influence the public agenda and decision making surrounding issues such as funding (Carragee & Roefs, 2004).

How have some U.S. media framed autism over the years? And what might be the implications for understanding media advocacy (including agenda setting and framing) in terms of theory and practice? The purpose of this study is to examine a sample (N = 302) of The New York Times and The Washington Post news coverage of autism from 1996 to 2006 in order to determine if and how the amount and framing of news coverage may have helped push autism onto the national agenda prior to the passing of the CAA. Although limited in scope, by evaluating the content that makes the news, this study could help health and science communication professionals and scholars better understand the concept of media advocacy, including which frames might be most effective in influencing public agendas and helping facilitate policy change over time.

Literature Review

News Coverage of Autism

Existing mass communication research on autism seems to focus primarily on media coverage of the controversy surrounding the MMR vaccine as a possible cause of autism (Boyce, 2006; Clarke, 2008; Holton, Weberling, Clarke, & Smith, 2012). For example, Clarke (2008) conducted a content analysis of 279 articles from three U.S. and three U.K. newspapers from 1998 to 2006 to examine balance in media coverage of the proposed link between autism and the MMR vaccine. Clarke’s results revealed that 49% of U.S. articles and 60% of U.K. articles were “balanced.” Among articles that were not balanced, U.S. newspapers presented more “antilink” coverage (i.e., no link between the vaccine and autism) while U.K. newspapers presented more “prolink” coverage (i.e., implied link between the vaccine and autism).

Similarly, Boyce (2006) conducted a content analysis of the autism-vaccine controversy in British newspapers, television, and radio coverage in 2002 and found journalists balancing “expert” sources, such as scientists, with the “ordinary voices” of parents and some politicians. As Boyce noted, journalists select sources for inclusion in news stories for many different reasons. Sources may be chosen for their expertise or news value, for editorial agendas or influence, or to provide balance or objectivity, or they may simply be chosen based on time constraints or because they were available or put forth by an organization or public relations practitioner (Boyce, 2006). Boyce argued that journalists should not rely on “experts” as the sole source of information for stories on scientific issues such as autism, because doing so ignores the larger social context and complex human consequences. She also argued that expertise needs to be defined within the context of the story, stating:

By being more forthright about the expertise a source possesses, journalists can limit the influence of powerful or elite sources and potentially redefine the role of public relations, calling for their profession to provide sources with more specific and relevant expertise and knowledge. (Boyce, 2006, p. 903)

While these articles provide valuable information on framing and sources within autism coverage, they focus specifically on media coverage of the autism-vaccine controversy. They do not take into account other news stories about autism and framing or the possible relevance to funding and policy change. Of course, research on other health issues has considered sources used in scientific reporting, including the role of expertise, source credibility, and the power of celebrity to help push issues onto the public agenda (e.g., Conrad, 1999; Corbett & Mori, 1999; Len-Ríos et al., 2009; Marin & Van Oss Marin, 1990; Tanner & Friedman, 2011). And many studies have focused on media framing of other health issues, including cancer, HIV/AIDS, and obesity, related to policy and funding (e.g., Andsager & Powers, 2001; Kim & Willis, 2007; Lawrence, 2004; Major, 2009; Rogers et al., 1991; Smith & Wakefield, 2005). These studies provide theoretical guidance and additional rationale for the current study, as described below.

News Framing of Health and Social Issues

The media play an important role in defining and framing health and social issues for the public, including offering possible causes and solutions for various problems (Kim, Carvalho, & Davis, 2010; Kim & Willis, 2007; Lawrence, 2004). Media coverage draws attention to problems, thus helping set the agenda for policy makers and the general public (Cobb & Elder, 1972; Rogers et al., 1991). In short, media framing of health issues may help tell the audience not only which issues to think about but also how to think about them (Kim et al., 2002).

The concept of framing has been used extensively to explore the ways in which media present problems to the public. Entman (1993, 2007) stated that media framing essentially involves selection and emphasis of certain aspects of an issue to promote salience among the audience. He argued that framing has four functions: defining problems, diagnosing causes, making moral judgments, and suggesting remedies (Entman, 1993). Ultimately, news frames are designed to “encourage target audiences to think, feel, and decide in a particular way” (Entman, 2007, p. 165).

Similarly, Reese (2001) referred to framing as “the way events and issues are organized and made sense of, especially by media, media professionals, and their audiences” (p. 7). Iyengar (1991) argued that the primary factor determining public opinion of social issues is the assignment of responsibility, which matters because “individuals tend to simplify political issues by reducing them to questions of responsibility, and their opinions on issues flow from their answers to these questions” (p. 8). Applying this assumption to the current study, it may be that public opinion and policy makers’ actions toward autism flow from media framing of the issue. Of course, these notions are predicated on audiences actually seeing and paying attention to news stories, which is part of the challenge for communicators working to set the agenda through media advocacy and health/science reporting.

As mentioned, framing is sometimes tied to agenda setting in research exploring how health issues become the focus of media and public attention and, sometimes, policy change. A seminal study of media coverage of HIV/AIDS from 1981 to 1988 revealed four “eras” of news coverage that occurred in chronological order: initial, scientific, human, and political (Rogers et al., 1991). The authors used time series analysis to examine how real-world indicators and other major events, such as new scientific studies or celebrities announcing that they have AIDS, influenced media coverage and, eventually, health policy. As the authors explained, news focused on science in early coverage of HIV/AIDS, because this is when the disease was most mysterious and new information was constantly being released. As time went on and the disease affected more people, human and policy themes began to emerge in news coverage (Rogers et al., 1991). The current study of autism is loosely based on this research, although much smaller in scope, and may reveal similar findings for a very different health issue.

Framing of health and social problems matters because media coverage can affect the way the public reacts to issues such as poverty, pollution, obesity, and cancer (e.g., Andsager & Powers, 2001; Hoffner & Ye, 2009; Kensicki, 2004; Kim et al., 2010; Kim & Willis, 2007; Lawrence, 2004; Major, 2009; O’Keefe & Jensen, 2008; Slater, Hayes, Reineke, Long, & Bettinghaus, 2009; Slater, Long, Bettinghaus, & Reineke, 2008; Smith & Wakefield, 2005). Wallack et al. (1993) observed that Americans prefer to frame health issues in order to portray the overall social system as fundamentally sound, while attributing problems to irresponsible individuals. This can lead to ignoring systemic causes and solutions and to the medicalization of problems, where the focus is on individual treatment rather than on prevention or elimination of issues at the societal level. Hallahan (1999) referred to the media’s assignment of responsibility or blame as diagnostic framing.

Similarly, Iyengar (1991) referred to the prevalence of episodic framing, where stories are told from the perspective of an individual person or event, over thematic framing, where stories are told from a broader, societal perspective. He suggested that episodic framing makes individuals less likely to hold government agencies or public officials accountable for the alleviation of problems, thereby decreasing the audience’s perception that they can participate in the political process of helping to solve the problem (Iyengar, 1991). More recently, an experiment involving media framing of lung cancer and obesity confirmed that thematic framing is the most effective way to help people understand society’s role in solving problems, which might inspire collective public action, while episodic framing might inspire individual action (Major, 2009).

Media Advocacy and Mobilizing Information

Media advocacy combines the concepts of agenda setting and framing and involves framing public health problems to offer specific solutions (Gibson, 2010; Wallack et al., 1993). Although it has been described conceptually, media advocacy seems to be difficult to measure empirically, partially because it has been broadly defined as a “catch-all word for the set of skills used to create a shift in public opinion and mobilize the necessary resources and forces to support an issue, policy, or constituency” (Wallack et al., 1993, p. 27). Recent research has recognized the difficulty of measuring media advocacy while highlighting the importance of continuing to explore the concept. A study on media coverage of binge drinking suggested that media and communication campaigns need to focus more on the consequences of the problem as well as on providing specific solutions for parents (Askelson, Campo, Mastin, & Slonske, 2009). And a study on a consortium of California health clinics determined that while direct lobbying may be most effective to activate actual policy change, media advocacy is still necessary to increase public awareness and support, which may help lead to policy changes (Gardner, Geierstanger, Brindis, & McConnel, 2010). This may be the case for autism as well; more research on media advocacy is needed.

One simple yet important part of media advocacy may include providing calls to action or mobilizing information in communication materials and media coverage. Following a content analysis of 300 news articles covering pollution, poverty, and incarceration, Kensicki (2004) reported that media coverage rarely indicated a specific cause, responsible party, or means to act in order to solve the problems. Instead, Kensicki found few “calls to action” regarding what the public can do to help find or create solutions, and it has been argued that such “mobilizing information” is fundamental in enabling and empowering the public to act (Lemert, 1984). Kensicki (2004) suggested that when media do not discuss how problems can be solved, they promote public apathy due to a lack of connection between social problems, nonprofit or government organizations, and individual action. Media framing of the causes and solutions of autism may affect public perceptions of the problem, whereby focusing on solutions may influence people to actively seek or facilitate change. Similarly, calls to action may provide individuals with the specific information they need to act or get involved with a cause.

The concept of mobilizing information (MI) is similar to Kensicki’s (2004) “calls to action” and has been cited in mass communication research since the 1970s (Hoffman, 2006; Lemert, 1981, 1984; Lemert, Mitzman, Seither, Cook, & Hackett, 1977; Nicodemus, 2004); and it is increasingly being used in studies on disaster and health communication (Friedman & Hoffman-Goetz, 2003; Friedman & Kao, 2008; Tanner, Friedman, Koskan, & Barr, 2009; Weberling, 2010). In short, MI is media-provided information that presents the audience with a means to act on existing ideas and motivations. In a news story, MI might include names, phone numbers, websites, times and dates of meetings, titles of documents, and more.

In their early research, Lemert et al. (1977) found that 36% of newspaper items contained some type of MI, although it was typically found in positive stories rather than stories of a negative or controversial nature. Lemert (1981) later distinguished three types of MI: identificational MI, which might include names and contact information for people or groups; locational MI, which might include the time and place of an activity; and tactical MI, which might include specific instructions for certain behaviors. Other studies have found evidence of these types of MI, though it is usually limited, with the less explicit identificational MI being the most common (Hoffman, 2006; Tanner et al., 2009; Weberling, 2010). The current study also employed these definitions of MI to determine how media coverage may (or may not) encourage citizens to act related to autism.

Research Questions and Hypotheses

Using previous research as a guide, this study sought to answer and test several research questions and hypotheses. First, following research on the importance of framing health issues, including Rogers et al.’s (1991) study of HIV/AIDS in the 1980s, which found a science “era” in early news coverage of the disease and increased policy frames in later years, this study asks the following research question and proposes a two-part hypothesis:

Research Question 1: What are the most common frames presented in newspaper coverage of autism from 1996 to 2006? How does framing change over time?

Hypothesis 1a: Between 1996 and 2006, early newspaper coverage of autism is more likely to include science frames.

Hypothesis 1b: Between 1996 and 2006, later newspaper coverage of autism is more likely to include policy frames.

Although Boyce (2006) looked at sources in British news coverage of autism, Clarke’s (2008) study that included U.S. news coverage of autism did not examine sources as a variable. Thus, the current study also asks the following research question:

Research Question 2: What sources are included in newspaper coverage of autism from 1996 to 2006?

Finally, based on Entman’s (1993, 2007) four functions of framing, which include diagnosing causes and suggesting remedies; the concept of media advocacy, which includes advancing specific solutions (Gibson, 2010; Wallack et al., 1993); and the concept of MI (e.g., Lemert, 1981, 1984; Hoffman, 2006, Tanner et al., 2009; Weberling, 2010), this study asks a third research question and offers the following final hypothesis:

Research Question 3: How does newspaper coverage of autism from 1996 to 2006 present possible causes and solutions?

Hypothesis 2: Newspaper coverage of autism will omit calls to action or mobilizing information more often than it includes such information.

Method

The current study employs content analysis to examine news coverage of autism in The New York Times and The Washington Post from January 1, 1996, to December 31, 2006. The New York Times and The Washington Post were chosen because they have consistently high circulations (BurellesLuce, 2008) and are often recognized as agenda-setting media or “prestige press” read by policy makers and other elites (Lacy, Fico, & Simon, 1991; Merrill, 1999; Pew Center for Research, 2010). Additionally, newspapers remain an important source for complex stories including health and science information (Len-Ríos et al., 2009; Riffe, Lacy, & Reimold, 2007; Vasterman, Yzermans, & Dirkzwager, 2005). The time period (1996-2006) was chosen because U.S. autism incidence levels increased in the mid-1990s (Fighting Autism, 2009), and the CAA was passed in December of 2006.

Selection of Sample

Steps were taken to determine the best database and search terms for creating the study sample. For example, a search for the word “autism” revealed the same number of news items as searches using “autistic” and “autis*.” Additionally, the LexisNexis Academic database was chosen because it yielded the most articles (2,185) on autism during the specified time period when compared with the America’s Newspapers (1,546) and Google News (764) databases. Using LexisNexis and the word “autism” to search The New York Times and The Washington Post news coverage from 1996 to 2006 yielded 2,185 total news items. Analysis included all news items, including feature stories about autism, regardless of page, section, or length of article; however, letters to the editor were excluded because of the study’s focus on news framing, which is generally considered to be a function of reporting rather than public response. After eliminating letters to the editor, along with duplicate articles, obituaries, wedding announcements, and book reviews, 1,517 viable news items remained. This population was stratified by year, and approximately 20% of each year’s articles were randomly sampled. Systematic sampling was used to select every fifth story from each newspaper, yielding a final sample of 302 articles, which accounts for approximately 20% of each year’s articles as well as 20% of the total number of relevant articles. The sampling error at the 95% confidence level is approximately 5%. The unit of analysis is the complete news item or article. Articles found using the LexisNexis Academic database were printed for two graduate student coders to analyze.

Coding Variables

Coders determined presence or absence of frames in news coverage and also determined the dominant frame in each article. This research focused on three types of frames: the science frame, the human frame, and the policy frame. Generally, science frames focused on the biological, genetic, or environmental causes and effects of autism. Human frames focused on individual/personal or familial experiences, perspectives, and/or consequences of autism. Policy frames focused on political, legal, or funding issues related to autism. For this study, a news frame was defined as a theme or central organizing idea within an article. More specifically, to be considered a frame, there must have been at least one paragraph (minimum of two to three sentences) focusing on the human, scientific, or policy aspects of autism.

In addition to presence or absence of frames, coders determined each article’s dominant frame, that is, the predominant or most frequently mentioned theme or central organizing idea within the article. The codebook (available on request) stated that the dominant frame may be evident in the headline or lead paragraph, but coders should read the article all the way through to determine the primary focus of the article. Explanation and examples provided in the codebook included the following:

An article may begin with a story about a boy with autism and the struggles he faces in school, but then the rest of the article may focus on the science behind autism. This article should be determined to have both “human” and “science” frames, but the dominant frame should be “science” because science is the focus of the majority of the article.

Sources, defined as any individual or organization quoted or cited for providing information in autism articles, were also counted and coded into the following categories: government, nonprofit, individual, family, personal, science/medical, celebrity, and other. Sources could be specifically named (e.g., President George W. Bush) or they could be referred to more generally for providing information (e.g., “said a government official” or “according to the CDC”). Coders entered the number of each type of source appearing in each story, counting sources only once per article. The codebook provided definitions and examples of sources. While government, nonprofit, and science/medical sources may be fairly clear, other sources were defined as follows: “Individual” referred to a person who actually has autism or an ASD; “Family” sources included parents, siblings, grandparents, or other family members related to someone with autism; “Celebrity” sources referred to famous, public figures who are somehow connected to autism; “Personal” sources were defined as people other than individuals with autism and their family members who have firsthand experience with autism, such as teachers, counselors, or friends. Coders were to select “Other” only if sources did not seem to fit into any other defined category. If sources seemed to fit into two categories, such as celebrity and family (e.g., actors/actresses or sports figures who have children with autism), coders were instructed to code these sources for how they are best known or how they were introduced and described in the story (in this example, as a celebrity rather than as a family member).

Coders also looked for the presence or absence of possible causes and solutions related to autism. Causes could include the autism-vaccine link or discussion of other possible environmental or genetic causes, while solutions might include new treatments, therapies or technologies, or educational or policy solutions. Related to solutions, coders also looked for calls to action or for mobilizing information, defined as any information appearing in an article that enables audiences to act. This study relied on Lemert’s (1981, 1984) original definitions of MI and accounted for identificational MI (names of individuals or organizations), locational MI (phone numbers, web links, addresses), and tactical MI (specific instructions on how to get involved or act on existing motivations).

Coder Reliability

After a thorough training session, two graduate student coders analyzed 15 articles using the coding protocol. An early check for intercoder reliability revealed some discrepancies, and adjustments were made to the coding protocol. Coders then used the revised codebook to code 30 articles to represent 10% of the sample, as recommended by Wimmer and Dominick (2006). Stratified random sampling was used to select articles for intercoder reliability; articles were divided by year and randomly selected within each year in order to be representative of the study sample. Final intercoder reliability was α = .85 or higher for presence of frames, identification of dominant frames, causes and solutions, and MI. Krippendorf’s alpha was also used to assess reliability on the source variables, which involved exact counts of each type of source; reliability for the source variables ranged from α = .80 to α = .99. Thus, all variables achieved acceptable reliability scores consistent with other published content analyses (Matthes, 2009; Riffe, Lacy, & Fico, 2005).

Sample Stratification

In order to examine findings and frames over time, the 11-year time period (1996-2006) was divided into five time frames based on major events related to autism that influenced news coverage over time. The first time frame, 1996-1997, was not marked by any major incident affecting news coverage and could be considered the “initial” time frame in this sample. In February 1998, Wakefield et al. discussed the possible autism-MMR vaccine link at a press conference, and his research was published in the medical journal, The Lancet. As noted in other scholarly studies (Boyce, 2006; Clarke, 2008), this news made waves and prompted further scientific and medical research and subsequent media coverage throughout 1998 and 1999; thus, the second time frame for this study is 1998-1999.

The third time frame, 2000-2002, begins with Congress passing the Children’s Health Act, which authorized federal funding for autism research for the first time (CDC, n.d.). In the fourth time frame, 2003-2004, several published studies found no evidence of an autism-vaccine link, and there was breaking news that Wakefield et al.’s (1998) original study on autism had been discredited due to lack of evidence and conflicts of interest regarding research funding (O’Connor, 2004). The CAA was introduced in 2005 and passed in 2006, as noted in some news coverage (Associated Press, 2006), and these 2 years make up the fifth and final time frame in this study. These time frames were used to organize data in order to present findings about news coverage of autism over time.

Findings

Autism coverage in The New York Times and The Washington Post increased steadily over time from 1996 to 2006 in accordance with U.S. autism incidence rates, as reported by data from the CDC and the Individuals with Disabilities Education Act (Fighting Autism, 2009). Before conducting analyses to answer this study’s research questions and test hypotheses, the data were examined to determine equivalence between the two newspapers. Each variable was analyzed using pairwise statistical comparisons (z tests and t tests, depending on the variable); these tests revealed no statistically significant differences between the two newspapers.¹ Thus, data from the two newspapers were combined for all remaining analyses.

Research Question 1 asked about the most common frames presented in The New York Times and The Washington Post coverage of autism from 1996 to 2006, as well as how framing changed over time. Overall, the human frame was most prevalent (63% or 189 articles), followed by the policy frame (47% or 143 articles), followed by the science frame (44% or 132 articles). The differences between human and policy frames and between human and science frames were statistically significant, z > 1.96, p < .01. In looking at the presence of frames over time, it appears that the presence of science frames decreased over time, with science frames presented in 57% of articles in 1996-1997 and in 36% of articles in 2005-2006 (see Figure 1). Policy frames, on the other hand, increased over time, with policy frames presented in 33% of coverage in 1996-1997 and in 57% and 47% of coverage in 2003-2004 and 2005-2006, respectively. However, these differences between time frames were not statistically significant when tested to control for sample size and Type I error rate, z < 1.96, p > .01. The human frame was offered more frequently than the other two frames consistently over time.

Figure 1.

Presence of frames over time.

Dominant frames showed some but not all of the same patterns as simple presence of frames (see Figure 2). For instance, among the total sample, human frames were dominant in 41% of articles (124), science frames were dominant in 32% of articles (96), and policy frames were dominant in 27% of articles (82). However, the dominance of human frames seemed to decrease over time, with a significant difference between the proportion of articles with dominant human frames in 1996-1997 (61%) and 2003-2004 (33%), z > 1.96, p < .01. Meanwhile, the dominance of policy frames increased over time, with a significant difference between the proportion of articles with dominant policy frames in 1996-1997 (11%) and 2000-2002 (30%), as well as between 1996-1997 (11%) and 2003-2004 (31%), z > 1.96, p < .01. The proportion of articles with dominant science frames stayed fairly consistent over time, and differences between time frames were not statistically significant.

Figure 2.

Dominant frames over time.

These findings indicate partial support for Hypothesis 1a, which predicted that early newspaper coverage would be more likely to include science frames. Indeed, science frames were more prevalent in early coverage of autism than policy frames. As predicted, the presence of science frames decreased over time when one compares Time 1 (1996-1997) to Time 5 (2005-2006), but these differences were not consistent or statistically significant (p < .01) across all time periods. For example, there was a spike in the presence of science frames in Time 4 (2003-04). Also, the proportion of dominant science frames stayed fairly flat over time, ranging from 28% to 37% across all time periods (see Figures 1 and 2). Thus, while early coverage of autism appeared to be more likely to include science frames, the differences were not necessarily significant or consistent over time.

Hypothesis 1b, which predicted that later newspaper coverage would be more likely to include policy frames, also received partial support according to the data. Indeed, policy frames were more prevalent in later coverage of autism than science frames; however, again, these differences were not statistically significant (p < .01) across all time periods. As predicted, the presence of policy frames increased over time when one compares Time 1 (1996-1997) to Time 4 (2003-2004), but this increase was not statistically significant. The proportion of dominant policy frames also increased over time, with statistically significant differences between Time 1 (1996-1997) and Time 3 (2000-2002) and between Time 1 (1996-1997) and Time 4 (2003-2004). Thus, Hypothesis 1b, which predicted an increase in policy frames over time, received more robust and consistent support than Hypothesis 1a, which predicted a decrease in science frames over time (see Figures 1 and 2).

Research Question 2 asked about the types of sources included in The New York Times and The Washington Post news coverage of autism. Coders counted the number of each type of source identified in each article. Table 1 shows sources in terms of both the proportion of articles that included each type of source, as well as the proportion of total sources provided across all articles. Perhaps not surprisingly, scientific/medical “experts” were included most frequently in news coverage of autism. Government and family members were the next two most common types of sources, followed by nonprofit sources. “Other” sources, which included pharmaceutical company representatives and others who did not seem to fit into any of the established source categories, were also fairly common. Personal sources, including teachers, friends, and neighbors, as well as individuals with autism, were not frequently included in news coverage. Finally, celebrities were the least common source type included in autism news coverage (see Table 1 for details).

Table 1.

Presence and Proportions of Sources in Articles.

	Presence of Sources in Articles	Proportion of Total Sources
Science/medical	47% (143)	34% (468)
Government	41% (125)	19% (268)
Family	42% (128)	16% (222)
Nonprofit	35% (106)	11% (154)
Other	21% (64)	8% (114)
Personal	18% (54)	6% (87)
Individual	14% (41)	5% (67)
Celebrity	2% (7)	1% (9)
Total	302 (articles)	1389 (sources)

Sources were also examined for trends over time. Figure 3 shows the top four types of sources in terms of proportions as described above and shown in Table 1 (science/medical, government, family, and nonprofit). As the line graph shows, science/medical and government sources started at the same point, each appearing in 39% of articles in 1996-1997. Family members were actually the most frequently cited source during this time period, appearing in 50% of articles, while nonprofit sources were mentioned in only 17% of early news coverage. However, in the final years of this study (2005-2006), government sources were the most common, appearing in 46% of coverage, followed by science/medical sources (41%), family members (35%), and nonprofit sources (33%). As the graph shows, there was a spike in nonprofit sources in 1998-1999, with 52% of articles citing nonprofit sources.

Figure 3.

Proportions of top four sources over time.

Research Question 3 asked how The New York Times and The Washington Post presented possible causes and solutions of autism. The proportion of articles presenting possible solutions was significantly higher than the proportion of articles presenting possible causes. Overall, 48% of articles discussed possible solutions or remedies for autism, while 35% of articles discussed possible causes of autism, z > 1.96, p < .01. Although the proportions of articles featuring possible causes and solutions varied over time, solutions were offered consistently more frequently than causes (see Table 2).

Table 2.

Presence of Causes, Solutions, and Mobilizing Information Over Time.

	Causes	Solutions	Mobilizing Information
1996-1997	22% (4/18)_a	61% (11/18)_b	22% (4/18)
1998-1999	33% (9/27)_a	52% (14/27)_a	11% (3/27)
2000-2002	35% (28/81)_a	40% (32/81)_a	20% (16/81)
2003-2004	49% (40/81)_a	56% (45/81)_a	12% (10/81)
2005-2006	27% (26/95)_a	45% (43/95)_b	15% (14/95)
Total	35% (107/302)_a	48% (145/302)_b	16% (47/302)

Note: Proportions not sharing a common subscript are significantly different by the z test at the p ≤ .01 level. Comparisons are between proportions in cause and solution columns only.

Finally, Hypothesis 2 predicted that newspaper coverage of autism would omit MI or calls to action more often than it mentions such information. This hypothesis was supported. Indeed, only 16% of the sample contained some type of MI. Identificational MI and locational MI, which generally include contact information such as names and phone numbers or organizational websites, were found in 15% of the sample. Generally, the articles that included MI included multiple names (M = 2.86, SD = 2.99, range = 1-10) as well as locations (M = 3.66, SD = 3.51, range = 1-13). Furthermore, the more nuanced tactical MI, including specific instructions that may promote action among individuals with existing motivations, was found in only 8% of the sample. Again, articles that included tactical MI typically presented more than one tactic (M = 1.58, SD = 1.14, range = 1-6). Table 2 shows changes in MI over time; these differences were not statistically significant. Figure 4 shows a line graph of causes, solutions, and MI over time.

Figure 4.

Presence of causes, solutions, and mobilizing information over time.

Discussion and Conclusions

This study reveals several important findings surrounding news framing of autism and provides a foundation for additional research related to media advocacy, funding, and policy change for important health issues. Human frames were consistently the most common and dominant frames found in news coverage of autism appearing in The New York Times and The Washington Post from 1996 to 2006. This finding contrasts with previous research on more stigmatized health issues such as HIV/AIDS (Rogers et al., 1991), where it may be harder to find a human face willing to be identified in media coverage. Indeed, the majority of autism coverage focused on the human aspects of ASDs, generally by telling the story of a particular individual with autism or of families with children with ASDs. When it comes to health issues, this kind of episodic framing is not typically associated with holding the government, organizations, or society responsible for large-scale policy change (e.g., Iyengar, 1991; Kensicki, 2004; Lawrence, 2004; Major, 2009). However, these human interest stories may help call attention to diseases or disorders, even among policy makers, and this seems especially true with autism.

For example, one article told the story of former president George W. Bush visiting a 17-year-old boy with autism, simply because Bush had seen the boy’s story on television and it made Bush cry (“Bush Visits,” 2006). Other articles detailed highly personal, human interest stories as the impetus for celebrity or policy makers’ involvement in autism advocacy efforts, such as former professional football player Doug Flutie starting a foundation in honor of his autistic son, or Congressmen Dave Weldon and Dan Burton becoming involved in the fight for autism funding partially because of experiences in their own families. Clearly, human framing or putting a human face on a disease or disorder helps with media advocacy efforts and may ultimately affect policy change. The fact that policy makers, including Bush, Weldon, and Burton, were touched by and/or involved with autism and autism advocacy efforts (and were not afraid to say so) could have affected the passing of the CAA, and their involvement likely had more of an impact on policy change than media advocacy efforts. However, the increased media coverage over time certainly did not hurt funding for the health issue and likely helped increase awareness among policy makers who were not as personally familiar with autism.

While it appears that this finding contradicts what some scholars (e.g., Iyengar, 1991; Major, 2009) have suggested about thematic framing being more powerful for holding government entities responsible and possibly influencing policy change, we must remember to consider factors outside of media coverage, such as political lobbying, and the impact that individuals and organizations can have on the policy-making process. As some scholars have noted (Gardner et al., 2010), direct advocacy, including lobbying, typically has more of an effect on policy change; however, media advocacy and the resulting media coverage are still necessary to drive public and political attention to issues in order to help facilitate change. Clearly, more research is needed on framing effects as well as on the efficacy of specific tactics involved in media advocacy. Experimental research can help continue to examine the possible effects of media framing on individuals, but larger scale, longitudinal studies are also needed to help explore the media advocacy process and the influence that various individuals, organizations, communication materials, and media may have on the public and political agendas, particularly as the media environment continues to change. Gibson (2010) referred to groups of stakeholders that may influence the media advocacy and policy-making processes as “subaltern public spheres” (p. 57), and it is important that health communicators interact with these groups no matter where they may be gathering (e.g., on parenting blogs or social media sites such as Facebook or Twitter). Policy makers and media need to pay attention to these voices to keep up with issues, and in turn, these groups may be able to influence the process by continuing to share their personal stories. For practitioners and scholars, the whole process can be understood only through continued research.

Although science frames were more prevalent in early autism coverage, they were presented fairly consistently over time, most likely because the exact cause of autism is still unknown and research is ongoing. This differs from an issue such as HIV/AIDS or obesity, where causes are now understood and prevention and/or treatment measures exist. Indeed, from 1996 to 2006, there were multiple medical studies on the possible autism-vaccine link, not to mention other scientific research related to autism. Media coverage highlighted this steady stream of research, with some articles citing as many as 14 medical experts or scientific studies (Harris & O’Connor, 2005). Indeed, the mystery and controversy surrounding autism probably helped make the issue more newsworthy, driving media coverage along with reporters’ search for multiple sources to explain studies and provide context for audiences.

As predicted, policy frames increased over time and were more prevalent in later coverage of autism, which is consistent with research on other health issues (e.g., Rogers et al., 1991) and likely consistent with media advocacy efforts related to autism. As incidence levels rise, nonprofit organizations form, and there is more of a demand for research and services related to health issues. Meanwhile, media coverage and policy frames within that coverage also likely increase, as revealed in this study. However, in addition to focusing on funding for more research, many of the articles focused on funding for education and other educational policy changes, which may reflect the age range of individuals being diagnosed with autism, as well as the scope and relative “newness” of autism in the educational system.

The sources included in autism news coverage seem appropriate when one considers the issue. Science/medical and government sources are often deemed “experts” by the media and are therefore sought after for inclusion in health news. Sources that provide information freely and/or frequently in a format that audiences might find easily digestible are especially likely to be included in media coverage about scientific topics (Conrad, 1999; Len-Ríos et al., 2009; Tanner & Friedman, 2011). Additionally, many families were included as sources in news coverage of autism. As Boyce (2006) noted, families could be considered the real “experts” when it comes to an issue such as autism, and it seems that at least some U.S. newspaper reporters saw them as such and included them in coverage accordingly. However, individuals with autism, who could also be considered experts, were included in only 14% of articles; these are some of the “ordinary voices” that Boyce (2006) noted seem to be missing in autism news coverage. While nonprofit organizations were included in a fairly substantial proportion of articles (35%), they made up only 11% of total sources. This could reflect the limited number of autism nonprofits that existed or made themselves known to media at the time, which could be evidence of the need or opportunity for greater media advocacy efforts on the part of nonprofit organizations. Alternatively, there could be reluctance on the part of journalists to rely on nonprofit organizations as sources for an issue such as autism. The spike in nonprofit sources included in news coverage in 1998-1999 coincides with media coverage of the Wakefield et al. (1998) study and controversy surrounding the autism-vaccine link. It could be that nonprofit organizations became more willing to speak with media during this time, while government and science/medical responses may have been lacking as the controversy emerged and was sorted out, as one recent article suggested (Holton et al., 2012).

News coverage focused on solutions more than causes, which is likely the result of parents, educators, nonprofit organizations, and others pushing for solutions to help handle the difficult and still somewhat mysterious issue. Many of the articles focused on educational changes, and some offered community-building solutions, including after-school activities and websites or forums for parents or individuals with autism. Not surprisingly, specific causes of autism were not mentioned frequently; rather, the “mysteriousness” of autism was often the focus of news coverage, especially related to new research and information related to the autism-vaccine controversy.

Although many articles discussed solutions broadly, few provided specific MI that would enable audiences to act or to get involved in advocacy efforts related to autism. This is consistent with Kensicki’s (2004) findings about news coverage of social problems lacking specific “calls to action,” as well as several studies on the provision or exclusion of MI (Lemert, 1981, 1984; Hoffman, 2006; Holton et al., 2012; Tanner et al., 2009; Weberling, 2010) Again, this may indicate a need or opportunity for greater media advocacy on the part of health advocates and nonprofit organizations, or it could be related to journalists’ or news organizations’ processes and preferences in reporting. If the third part of media advocacy is advancing a specific solution, providing MI seems like a necessary part of the process to enable and empower audiences to act. However, the question still remains whether this is the result of a lack of information being provided by professional communicators and health advocates or whether the omission takes place on the part of reporters and/or specific media outlets. While Lemert’s (1984) early research found that reporters are likely to cut MI when facing time or space constraints, the media landscape has changed dramatically since that time, as Hoffman (2006) noted, and it is difficult to know whether MI is being omitted on the part of communication professionals or the media without further research. Conducting similar experiments today could yield very different results and help provide a better understanding of MI and modern media advocacy processes.

Limitations and Future Research

This study has limitations. First, the sample (N = 302) was small when one considers the longitudinal nature of the research. Only 20% of The New York Times and The Washington Post news coverage of autism was included in this study from 1996 to 2006, and those are only two of many media outlets. The sample did not include television, radio, blogs, or other forms of media, which may also influence public perceptions and policy surrounding autism. However, as noted, these two large newspapers are arguably still important to policy makers and help guide content across other platforms (Pew Research Center, 2010). Newspapers also remain an important source for complex stories and health information (e.g., Len-Ríos et al., 2009; Riffe et al., 2007; Vasterman et al., 2005). While the sample could have been larger, 20% per year and 20% of total coverage is fairly typical and sufficiently representative for content analysis (Riffe et al., 2005). In hindsight, it may have been helpful to code subframes within articles in order to draw more specific conclusions about particular frames. For instance, it would be helpful to know how many articles focused on education policy versus health policy.

Future research could explore more media, including television coverage, websites, blogs, and/or social media. Analyzing press releases or other communication materials over time to see if/how they compare with news coverage could reveal additional findings related to the media advocacy process. Surveys or interviews with the general public, policy makers, representatives from nonprofit organizations, and/or media would add yet another layer to this research. And experimental research could help confirm ideas about individual-level effects of framing on perceptions of responsibility, causes, solutions, the provision of MI, and willingness to take subsequent action. Ultimately, this study sought to add to current research on media advocacy and news framing of health issues, and it revealed some important findings related to autism. However, additional research is needed to better understand the media advocacy process related to health issues, public agendas, funding, and policy change.

Footnotes

Author’s Note

This article received the Eason Prize for Best Student Paper when it was presented to the Communicating Science, Health, Environment, and Risk (ComSHER) division at the 2010 Association for Education in Journalism and Mass Communication Annual Conference in Denver.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Brooke Weberling McKeever (PhD, University of North Carolina–Chapel Hill) is an assistant professor in the Public Relations sequence of the School of Journalism and Mass Communications at the University of South Carolina. Her research interests focus primarily on nonprofit public relations and health communication, including issues surrounding advocacy, funding, and strategies to increase stakeholder support.

References

Andsager

Powers

(2001). Framing women’s health with a sense-making approach: Magazine coverage of breast cancer and implants. Health Communication, 13, 163-185.

Askelson

Campo

Mastin

Slonske

(2009). The missing role of parents: A content analysis of newspaper coverage of parenting practices and communication strategies for addressing binge drinking. Communication Research Reports, 26(1), 50-61.

Associated Press. (2006, December 8). National briefing Washington: Autism research financing is approved. The New York Times, p. A33.

Boyce

(2006). Journalism and expertise. Journalism Studies, 7, 889-906.

BurellesLuce. (2008). Top newspapers, blogs & consumer magazines. Retrieved from http://www.burrellesluce.com/top100/2008_Top_100List.pdf

Bush visits with autistic basketball manager. (2006, March 16). The Washington Post, p. E2.

Carragee

K. M.

Roefs

(2004). The neglect of power in recent framing research. Journal of Communication, 54, 214-233.

Centers for Disease Control and Prevention. (2012a). Data & statistics. Retrieved from http://www.cdc.gov/ncbddd/autism/data.html

Centers for Disease Control and Prevention. (2012b). Facts about ASDs. Retrieved from http://www.cdc.gov/ncbddd/autism/facts.html

10.

Centers for Disease Control and Prevention. (n.d.). “Autism Community Report.” Retrieved from www.cdc.gov/ncbddd/autism/documents/AutismCommunityReport.pdf

11.

Clarke

(2008). A question of balance: The autism-vaccine controversy in the British and American elite press. Science Communication, 30(1), 77-107.

12.

Cobb

R. W.

Elder

C. D.

(1972). Participation in American politics—The dynamics of agenda-building. Baltimore, MD: Johns Hopkins University Press.

13.

Conrad

(1999). Uses of expertise: Sources, quotes, and voice in the reporting of genetics in the news. Public Understanding of Science, 8, 285-302.

14.

Corbett

J. B.

Mori

(1999). Medicine, media, and celebrities: News coverage of breast cancer 1960-1995. Journalism & Mass Communication Quarterly, 76, 229-249.

15.

Entman

R. M.

(1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51-58.

16.

Entman

R. M.

(2007). Framing bias: Media in the distribution of power. Journal of Communication, 57, 163-173.

17.

Fighting Autism. (2009). Autism statistics: Incidence, prevalence, rates. Retrieved from http://www.fightingautism.org/idea/autism.php

18.

Fombonne

(2003). The prevalence of autism. Journal of the American Medical Association, 289(1), 87-89.

19.

Friedman

D. B.

Hoffman-Goetz

(2003). Cancer coverage in North American publications targeting seniors. Journal of Cancer Education, 18, 43-47.

20.

Friedman

D. B.

Kao

E. K.

(2008). A comprehensive assessment of the difficulty level and cultural sensitivity of online cancer prevention resources for older minority men. Preventing Chronic Disease, 5(1), A07.

21.

Gardner

Geierstanger

Brindis

McConnel

(2010). Clinic consortia media advocacy capacity: Partnering with the media and increasing policymaker awareness. Journal of Health Communication, 15, 293-306.

22.

Gerber

J. S.

Offit

P. A.

(2009). Vaccines and autism: A tale of shifting hypotheses. Clinical Infectious Diseases, 48, 456-461.

23.

Gibson

(2010). The limits of media advocacy. Communication, Culture & Critique, 3(1), 44-65.

24.

Glanz

(2002). Perspectives on group, organization and community interventions. In Glanz

Rimer

B. K.

Lewis

F. M.

(Eds.), Health behavior and health education: Theory, research and practice (pp. 289-403). San Francisco, CA: Jossey-Bass.

25.

Godlee

Smith

Marcovitch

(2011). Wakefield’s article linking MMR vaccine and autism was fraudulent. Lancet, 342, 64-66.

26.

Hallahan

(1999). Seven models of framing: Implications for public relations. Journal of Public Relations Research, 11, 205-242.

27.

Harris

O’Connor

(2005, June 25). On autism’s cause, it’s parents vs. research. The New York Times, p. A1.

28.

Hoffman

L. H.

(2006). Is Internet content different after all? A content analysis of mobilizing information in online and print newspapers. Journalism & Mass Communication Quarterly, 83(1), 58-76.

29.

Hoffner

(2009). Young adults’ responses to news about sunscreen and skin cancer: The role of framing and social comparison. Health Communication, 24, 189-198.

30.

Holton

Weberling

Clarke

C. E.

Smith

M. J.

(2012). The blame frame: Media attribution of culpability about the MMR-autism vaccination scare. Health Communication. Advance online publication. doi:10.1080/10410236.2011.633158

31.

Iyengar

(1991). Is anyone responsible? How television frames political issues. Chicago, IL: University of Chicago Press.

32.

Kensicki

L. J.

(2004). No cure for what ails us: The media-constructed disconnect between societal problems and possible solutions. Journalism & Mass Communication Quarterly, 81(1), 53-73.

33.

Kim

S.-H.

Carvalho

J. P.

Davis

A. G.

(2010). Talking about poverty: News framing of who is responsible for causing and fixing the problem. Journalism & Mass Communication Quarterly, 87, 563-581.

34.

Kim

S.-H.

Scheufele

Shanahan

(2002). Thank about it this way: Attribute agenda setting function of the press and the public’s evaluation of a local issue. Journalism & Mass Communication Quarterly, 79, 1-25.

35.

Kim

S.-H.

Willis

L. A.

(2007). Talking about obesity: News framing of who is responsible for causing and fixing the problem. Journal of Health Communication, 12, 359-376.

36.

Kosicki

G. M.

(1993). Problems and opportunities in agenda-setting research. Journal of Communication, 43, 100-127.

37.

Lacy

Fico

Simon

T. F.

(1991). Fairness and balance in the prestige press. Journalism Quarterly, 68, 363-370.

38.

Lawrence

R. G.

(2004). Framing obesity: The evolution of news discourse on a public health issue. Harvard International Journal of Press/Politics, 9(3), 56-75.

39.

Lemert

J. B.

(1981). Does mass communication change public opinion after all? A new approach to effects analysis. Chicago, IL: Nelson-Hall.

40.

Lemert

J. B.

(1984). News context and the elimination of mobilizing information: An experiment. Journalism Quarterly, 61, 243-259.

41.

Lemert

J. B.

Mitzman

B. N.

Seither

M. A.

Cook

R. H.

Hackett

(1977). Journalists and mobilizing information. Journalism Quarterly, 54, 721-726.

42.

Len-Ríos

M. E.

Hinnant

Park

S.-A.

Cameron

G. T.

Frisby

C. M.

Lee

(2009). Health news agenda building: Journalists’ perceptions of the role of public relations. Journalism & Mass Communication Quarterly, 86, 315-331.

43.

Maher

M. T.

(2001). Framing: An emerging paradigm or a phase of agenda setting. In Reese

S. D.

Gandy

O. H.

Grant

A. E.

(Eds.), Framing public life: Perspectives on media and our understanding of the social world (pp. 83-94). Mahwah, NJ: Lawrence Erlbaum.

44.

Major

(2009). Break it to me harshly: The effects of intersecting news frames in lung cancer and obesity coverage. Journal of Health Communication, 14, 174-188.

45.

Marin

Van Oss Marin

(1990). Perceived credibility of channels and sources of AIDS information among Hispanics. AIDS Education and Prevention, 2, 154-161.

46.

Matthes

(2009). What’s in a frame? A content analysis of media framing studies in the world’s leading communication journals, 1990-2005. Journalism & Mass Communication Quarterly, 86, 349-367.

47.

McCombs

Ghanem

S. I.

(2001). The convergence of agenda setting and framing. In Reese

S. D.

Gandy

O. H.

Grant

A. E.

(Eds.), Framing public life: Perspectives on media and our understanding of the social world (pp. 67-81). Mahwah, NJ: Lawrence Erlbaum.

48.

McCombs

M. E.

Shaw

(1972). The agenda-setting function of the mass media. Public Opinion Quarterly, 36, 176-187.

49.

Merrill

(1999). The global elite. IPI Report, 4, 13-15.

50.

Nicodemus

(2004). Mobilizing information: Local news and the formation of a viable political community. Political Communication, 21, 161-176.

51.

O’Connor

(2004, March 2). Researchers retract a study linking autism to vaccine. The New York Times, p. A15.

52.

O’Keefe

Jensen

(2008). Do loss-framed persuasive messages engender greater message processing than do gain-framed messages? A meta-analytic review. Communication Studies, 59(1), 51-67.

53.

Petts

Niemeyer

(2004). Health risk communication and amplification: Learning from the MMR vaccination controversy. Health, Risk & Society, 6, 7-23.

54.

Pew Research Center. (2010). How news happens: A study of the news ecosystem of one American city. Retrieved from http://www.journalism.org/analysis_report/how_news_happens

55.

Pew Research Center’s Project for Excellence in Journalism. (2008, November 24). Health news coverage in the U.S. media. Retrieved from http://www.journalism.org/node/13770

56.

Reese

S. D.

(2001). Prologue: Framing public life: A bridging model for media research. In Reese

S. D.

Gandy

O. H.

Grant

A. E.

(Eds.), Framing public life: Perspectives on media and our understanding of the social world (pp. 7-31). Mahwah, NJ: Lawrence Erlbaum.

57.

Riffe

Lacy

Fico

F. G.

(2005). Analyzing media messages: Using quantitative content analysis in research (2nd ed.). Mahwah, NJ: Lawrence Erlbaum.

58.

Riffe

Lacy

Reimold

(2007). Papers lead TV in covering complex environmental issues. Newspaper Research Journal, 28, 77-87.

59.

Rogers

E. M.

Dearing

J. W.

Chang

(1991). AIDS in the 1980s: The agenda-setting process for a public issue. Columbia, SC: Association for Education in Journalism and Mass Communication.

60.

Scheufele

Tewksbury

(2007). Framing, agenda setting, and priming: The evolution of three media effects models. Journal of Communication, 57, 9-20.

61.

Slater

M. D.

Hayes

A. F.

Reineke

J. B.

Long

Bettinghaus

E. P.

(2009). Newspaper coverage of cancer prevention: Multilevel evidence for knowledge-gap effects. Journal of Communication, 59, 514-533.

62.

Slater

M. D.

Long

Bettinghaus

E. P.

Reineke

J. B.

(2008). News coverage of cancer in the United States: A national sample of newspapers, television, and magazines. Journal of Health Communication, 13, 523-537.

63.

Smith

Wakefield

(2005). Textual analysis of tobacco editorials: How are key media gatekeepers framing the issues? American Journal of Health Promotion, 19, 361-368.

64.

Takeshita

(2006). Current critical problems in agenda-setting research. International Journal of Public Opinion Research, 18(3), 11-32.

65.

Tanner

Friedman

D. B.

(2011). Authorship and information sourcing for health news on local TV news Web sites: An exploratory analysis. Science Communication, 33, 3-27.

66.

Tanner

Friedman

Koskan

Barr

(2009). Disaster communication on the Internet: A focus on mobilizing information. Journal of Health Communication, 14, 741-755.

67.

Vasterman

Yzermans

C. J.

Dirkzwager

A. E.

(2005). The role of the media and media hypes in the aftermath of disasters. Epidemiological Reviews, 27, 107-114.

68.

Wakefield

A. J.

Murch

S. H.

Anthony

Linnell

Casson

D. M.

Malik

. . . Walker-Smith

J. A.

(1998). Ilead-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet, 351, 637-641.

69.

Wallack

(1990). Media advocacy: Promoting health through mass communication. In Glanz

Lewis

F. M.

Rimer

B. K.

(Eds.), Health behavior and health education: Theory, research and practice (pp. 370-386). San Francisco, CA: Jossey-Bass.

70.

Wallack

Dorfman

Jernigan

Themba

(1993). Media advocacy and public health: Power for prevention. Newbury Park, CA: Sage.

71.

Weaver

D. H.

(2007). Thoughts on agenda setting, framing, and priming. Journal of Communication, 57(1), 142-147.

72.

Weberling

(2010). Mobilizing disaster relief: U.S. media coverage and public response to the Asian tsunami and Pakistan earthquake. International Communication Research Journal, 45, 50-80.

73.

The White House Archives. (2006). Fact sheet: Combating Autism Act of 2006. Retrieved from http://georgewbush-whitehouse.archives.gov/news/releases/2006/12/20061219-3.html

74.

Wimmer

R. D.

Dominick

J. R.

(2006). Mass media research: An introduction (8th ed.). Belmont, CA: Thomson Wadsworth.

75.

Wing

Potter

(2002). The epidemiology of autistic spectrum disorders: Is the prevalence rising? Mental Retardation and Developmental Disabilities Research Reviews, 8, 151-161.