Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Jun 1.
Published in final edited form as: J Commun. 2010 Jun 1;60(2):230–253. doi: 10.1111/j.1460-2466.2009.01474.x

Does Local Television News Coverage Cultivate Fatalistic Beliefs about Cancer Prevention?

Jeff Niederdeppe 1, Erika Franklin Fowler 1, Kenneth Goldstein 1, James Pribble 1
PMCID: PMC2885705  NIHMSID: NIHMS78582  PMID: 20563221

Abstract

A substantial proportion of American adults hold fatalistic beliefs about cancer prevention despite evidence that a large proportion of cancer deaths are preventable. Several scholars suggest that news media coverage is one source of these beliefs, but scant evidence has been brought to bear on this assertion. We report findings from two studies that assess the plausibility of the claim that local television (TV) news cultivates fatalistic beliefs about cancer prevention. Study 1 features a content analysis of an October 2002 national sample of local TV and newspaper coverage about cancer (n=122 television stations; n=60 newspapers). Study 2 describes an analysis of the 2005 Annenberg National Health Communication Survey (ANHCS, n=1,783 respondents). Study 1 indicates that local TV news stories were more likely than newspaper stories to mention cancer causes and scientific research and less likely to provide follow-up information. Study 2 reveals that local TV news viewing was positively associated with fatalistic beliefs about cancer prevention. Overall, findings are consistent with the claim that local TV news coverage may promote fatalistic beliefs about cancer prevention. We conclude with a discussion of study implications for cultivation theory and the knowledge gap hypothesis and suggest foci for future research.

Keywords: cultivation, knowledge gap, health communication, news coverage

Approximately thirty percent of US cancer deaths are attributable to smoking, while fourteen to twenty percent are attributable to overweight and obesity (Calle, Ridriguez, Walker-Thurmond, & Thun, 2003; Peto, Lopez, Boreham, Thun, & Heath, 1994). Although these data suggest that nearly half of all cancer cases are preventable, several studies document widespread fatalistic beliefs about cancer prevention (e.g., Slenker & Spreitzer, 1988). These beliefs are characterized by pessimism, helplessness, confusion and ambiguity about ways to avoid getting cancer. Nearly half of US adults agree that “It seems like almost everything causes cancer” (pessimism), more than a quarter agree that “There’s not much people can do to lower their chances of getting cancer” (helplessness), and almost three in four agree that “There are so many recommendations about preventing cancer, it’s hard to know which ones to follow” (confusion and ambiguity; Niederdeppe & Levy, 2007). People who agree with at least one of these statements are less likely than those who disagree to engage in a variety of behaviors that reduce cancer risk, and lower levels of formal education are linked to higher odds of agreement with all three fatalistic beliefs about cancer prevention (Niederdeppe & Levy, 2007; Han, Moser, & Klein, 2007).

The sheer volume of news coverage about cancer causes and prevention has led to broad speculation about its role in promoting fatalistic beliefs (Adelman & Verbrugge, 2000; Russell, 1999; Taubes, 1995). To date, however, little evidence has been brought to bear on these assertions. We do know that there is no shortage of news coverage about cancer (Adelman & Verbrugge, 2000; Viswanath et al., 2006), and journalistic norms may highlight a particular type of coverage that can be detrimental. News stories often call attention to new cancer risks at the expense of well-established causes, overemphasize new research findings, and omit follow-up information or details about what can be done to reduce cancer risk (Russell, 1999; Taubes, 1995). Over time, cumulative exposure to stories that highlight new cancer causes and research findings may lead to fatalistic beliefs about cancer prevention. Despite these assertions, few studies have systematically explored whether higher levels of exposure to cancer news coverage are associated with these beliefs. Moreover, scholars have not examined the effects of different sources or channels of cancer news despite good reason to suspect that local television (TV) coverage may be particularly harmful. Cultivation and knowledge gap research provide useful theoretical frameworks with which to test the relationship between local TV news viewing and beliefs about cancer prevention, with attention to possible differences by education. Building on these traditions, we summarize results from a content analysis of cancer coverage in local TV broadcasts and newspapers (Study 1) and an analysis of data from a large national survey (Study 2) to assess the plausibility of the claim that local TV cancer news coverage cultivates fatalistic beliefs about cancer prevention.

Cultivation Theory and Content-Specific Extensions

Cultivation theory posits that habitual exposure to TV influences beliefs about the general nature of the world (Gerbner, Gross, Morgan, & Signorielli, 1980). People who are heavily exposed to TV are more likely than those with less exposure to adhere to TV’s version of social reality (Gerbner, Gross, Morgan, Signorielli, & Shanahan, 2002). In general, most studies find a small but significant association between TV viewing and beliefs about a variety of topics, including crime, gender roles, and political views (Shanahan & Morgan, 1999).

Cultivation analysis refers to the research methodology typically applied in cultivation studies. While a few studies have focused on TV portrayals of doctors and health (e.g., Chory-Assad & Tamborini, 2003; Gerbner, Morgan, & Signorielli, 1982), cultivation analysis has been applied most frequently to the topic of TV violence and perceptions of the degree of violence in the world (Gerbner et al., 1980). Cultivation analysis has two primary components: (1) content analysis to identify overarching patterns of TV content (e.g., incidence of TV violence), and (2) survey data analysis to assess associations between TV viewing and, in the case of TV violence, perceptions of real-world crime. Perceptions of crime are also often compared to local crime rates in violence-related cultivation analyses (e.g., Hetsroni & Tukachinsky, 2006).

While Gerbner and colleagues (2002) acknowledge that different types of programs, genres, or channels are likely to have different types of short-term effects on audiences, they also articulate their belief that all TV programs contribute to “massive, long-term, and common exposure of large and heterogeneous publics to centrally produced, mass-distributed, and repetitive systems of stories” (p. 47). Nevertheless, several authors have focused on the specific cultivation effects of TV news broadcasts (Gross & Aday, 2003; Lowry, Nio, & Leitner, 2003; Romer, Jamieson, & Aday, 2003). Most of these analyses found significant associations between TV news exposure and crime perceptions. For instance, Romer et al. (2003) found that local TV news viewing increased fear of crime and concern about violence. Lowry et al. (2003) found that network TV news coverage of a 1994 crime scare accounted for most of an unprecedented increase, between 1992 and 1994, in perceptions of crime as the most important problem facing the US. Likewise, Gross and Aday (2003) found that local TV news increased the view that crime was the most important problem facing the Washington, D.C. area, although they found no association between local TV news and personal fear of crime victimization.

Recent extensions of cultivation theory, with a focus on differences between media source content and audience, are particularly relevant to understanding sources of fatalistic beliefs about cancer prevention. More Americans report getting information from local TV news than any other news source including national network newscasts, cable outlets, TV news magazines (such as Dateline or 60 Minutes), national and local newspapers, talk radio, and the internet (Fowler, Goldstein, Hale, & Kaplan, 2007; Pew Center, 2006). Furthermore, in part because of the focus on mass ratings rather than the affluent, educated demographics sought by newspapers (Kaniss, 1991), local TV news reaches a systematically different audience than many other news sources (Fowler, Goldstein, & Shah, 2008). In comparison to regular news consumers of national or print sources who tend to be older, more educated, and more knowledgeable than most of the population, habitual local TV news viewers (54 percent of the American public) tend to look more like the average American (Pew Center, 2006). In other words, local TV messages are not only viewed by more individuals but they also reach a socioeconomically diverse audience.

In light of the scope and breadth of their audience, local TV news stories have the potential to be very influential. Studies of local TV coverage of crime, politics, and health, however, have typically concluded that its content suffers from sensationalism and frequently contains little substance (Gilliam & Iyengar, 2000; Fowler et al., 2007; Pribble, Goldstein, Fowler, Greenberg, Noel, & Howell, 2006). While these studies suggest that local TV coverage of cancer may be detrimental, the specific role of local TV news in shaping beliefs about cancer prevention has not been addressed. Recent content analyses of cancer news have not differentiated between local and national TV (Slater, Long, Bettinghaus, & Reineke, 2008), have focused exclusively on print (Stryker, Emmons, & Viswanath, 2007), or have examined only a small set of local TV media markets without making comparisons to other types of media (Wang & Gantz, 2007). In light of differences in the audience of local TV relative to print (Potter & Kurpius, 2000) and intense competitive pressure in the broadcast news arena that may lead to further differences in content (e.g., Zaller, 1999), greater attention should be paid to how both local TV and newspaper coverage may contribute to fatalistic beliefs about cancer prevention.

Hypotheses related to cultivation theory

To this end, we offer three hypotheses about local TV cancer news content relative to local newspapers and a fourth hypothesis about the relationship between local TV news viewing and fatalistic beliefs about cancer prevention. News stories about cancer causes tend to be sensationalistic because they often focus on causes that (1) are new or controversial yet (2) are encountered by a sizeable proportion of the audience in their daily lives (Russell, 1999; Taubes, 1995). The intense competitive pressure in local TV is thought to be a culprit for particularly low quality news stories and high sensationalism relative to newspapers (e.g., Zaller, 1999). These factors lead to our first hypothesis:

H1: Local TV news coverage will be more likely than local newspaper coverage to report on the causes of cancer.

There are large resource constraints among local TV stations, producing fewer reporters for TV news relative to newspapers (Potter & Kurpius, 2000). Given the lack of resources, shorter deadlines for broadcasting stories and shorter time for presenting detailed information, TV journalists may also be more prone to covering research press releases about cancer research.

H2: Local TV news coverage will be more likely than local newspaper coverage to cover reports of new cancer research.

The limited timeframe for conveying information in a TV news segment relative to a newspaper article should also decrease the likelihood that local TV cancer news stories will provide follow-up information to the audience. Follow-up information provides details that could enable audiences to follow up on an action encouraged by the story. A lack of such information would prevent the viewer from being able to find source material behind the news stories in order to assess the credibility of scientific evidence about a cancer cause or research study.

H3: Local TV news coverage will be less likely than local newspaper coverage to include follow-up information about cancer or cancer research.

Cancer research studies that make the news often focus on novel or controversial findings (Russell, 1999; Stryker, 2002). Repeated exposure to stories that highlight uncertain cancer causes, particularly those identified by single research studies that have not yet been replicated, seems particularly likely to engender the belief that everything causes cancer. Cancer cause stories also carry implicit information about ways to prevent cancer (avoid the cause). Thus repeated exposure to stories about cancer causes may also contribute to a sense that there are so many recommendations about preventing cancer, it is hard to know which ones to follow. These effects would be exacerbated by a lack of follow-up information, where a viewer is not given adequate information to investigate the information’s source or credibility. Consequently, the extent to which a news source’s cancer coverage tends to focus on cancer causes, summarize cancer research (particularly studies about novel or controversial causes), and omit follow-up information should predict the extent to which exposure to that source engenders fatalistic beliefs about cancer prevention. Based on this proposition and our hypothesized patterns of coverage, we expect viewers of local TV news to be most likely to hold fatalistic beliefs.

H4: Local TV news viewing will be positively associated with an index of fatalistic beliefs about cancer prevention, controlling for socio-demographic factors.

Contributions to cultivation research

This paper shares cultivation analysis’s focus on cumulative effects of exposure to TV content with specific attention to possible effects of local TV news. Although we do not directly compare viewer perceptions to real-world prevalence of, say, crime, we do assume that views about cancer prevention need not be fatalistic. The reality is that everything does not cause cancer, and major institutions such as the National Cancer Institute (NCI) and American Cancer Society (ACS) have made clear recommendations about effective ways to reduce cancer risk. Thus, agreement with the fatalistic beliefs that everything causes cancer or that there are too many recommendations about cancer prevention constitutes a discrepancy between viewer perceptions and the reality of cancer prevention.

Knowledge Gaps and Local TV News about Cancer

The knowledge gap hypothesis proposes that individuals with greater education tend to acquire information provided by news coverage at a faster rate than individuals with lower education (Tichenor, Donohue, & Olien, 1970). As a result, gaps in knowledge between more and less educated groups may increase over time (Viswanath & Finnegan, 1996).

Hypotheses related to knowledge gaps

On its face, what we know about local TV content and fatalistic beliefs about cancer prevention might suggest patterns counter to those one would expect based on traditional knowledge gap research. If local TV news about cancer is detrimental and highly educated individuals acquire this information more quickly than less educated individuals, we might expect the relationship between local TV news exposure and fatalistic beliefs to be strongest among the most highly educated. Alternatively, some authors find that TV news does not increase knowledge and, in fact, may inhibit knowledge gain among audiences regardless of education (Patterson & McClure, 1976; Robinson & Davis, 1990). This might lead us to expect equivalent relationships between local TV news exposure and fatalistic beliefs by education.

Yet education is negatively associated with these beliefs, making it highly unlikely that local TV coverage of cancer increases fatalistic beliefs more among highly educated than less educated populations (Niederdeppe & Levy, 2007). The next logical question is to ask how less educated populations develop greater fatalism relative to more educated populations. Differential effects of local TV cancer news provide one plausible explanation. TV news provides information to the least educated in ways that are less cognitively demanding than print media (Eveland & Scheufele, 2000; Kwak, 1999; Neuman, 1976), and populations with low levels of formal education may lack the cognitive skills to make sense of complex scientific evidence related to cancer causes (e.g., Viswanath, 2006). To the extent that this information makes viewers prone to fatalistic beliefs, exposure to local TV coverage would be likely to increase educational gaps in these beliefs. Specifically, we should expect local TV cancer news coverage to have a larger impact on fatalistic beliefs among less educated compared to more educated populations.

H5: The expected positive association between local TV news viewing and fatalistic beliefs about cancer prevention will be stronger as education decreases.

Contributions to knowledge gap research

This paper builds on the tradition of knowledge gap research in our assessment of educational differences in response to local TV news exposure. We diverge from traditional knowledge gap research in our focus on the effects of aggregate TV news coverage (versus a discrete news event) to incorporate the cumulative effects central to cultivation theory. We also differ in our focus on fatalistic beliefs about cancer prevention, in contrast to discrete facts and knowledge, as the outcome of interest.

Overview of Study Objectives

This paper describes results from two studies, one of which uses 2002 data and the other of which uses data from 2005, to test the five hypotheses. We do not argue for a direct link between the specific content of 2002 news coverage (Study 1) and 2005 survey data (Study 2). Rather, Study 1 assesses the plausibility of the claim that local TV news has a tendency to cover cancer in a way that is more likely than other types of coverage, in the long term, to be detrimental. To do so, we compare cancer news stories from the largest national sample of local TV news ever collected to a large sample of local newspaper stories about cancer during the same time period. In light of significant differences between local TV and newspaper coverage about cancer, Study 2 uses national survey data to test associations between local TV news viewing and fatalistic beliefs about cancer prevention, paying specific attention to possible differences in these associations by education.

Study 1

Methods

We addressed H1, H2 and H3 by comparing (1) a national sample of local TV news coverage (n=122 TV stations) sampled in October 2002 to (2) a national sample of newspaper coverage (n=60 newspapers) sampled between October 2002 and March 2003.

Local TV news data

Local TV cancer news coverage data were drawn from the University of Wisconsin (UW) NewsLab’s sample of local TV news broadcasts from 2002, conducted in collaboration with the Norman Lear Center at the Annenberg School for Communication at the University of Southern California (USC). UW NewsLab collected evening news broadcasts from a random sample of 122 local TV stations drawn from a sampling frame of 200 stations (comprising the four major affiliates in each market) from the top 50 media markets, covering 67 percent of the nation’s population. Up to two half-hour news broadcasts were sampled from each station every evening from September 18 through November 4, 2002: the highest-rated half-hour of early-evening news and the highest-rated half-hour of late-evening news on each station. The 2002 dataset was originally collected to assess the quantity and quality of local TV political coverage (see Fowler et al., 2007, and Hale, Fowler, & Goldstein, 2007, for more information on the complete methodology). In partnership with the University of Michigan, all late-evening (usually 10 or 11pm) news broadcasts from October 2002 were reanalyzed for health coverage (see Pribble et al., 2006 for the complete methodology). Late-evening broadcasts were the focus of these analyses because they tend to include more health news than earlier broadcasts, often include segments originally aired earlier in the day, and attract a larger audience than morning or daytime local TV news (see Dean & Pertilla, 2007). Of 2,795 captured broadcasts, 1,799 health stories were identified, 258 (14%) of which were about cancer.

Newspaper data

We used the Lexis-Nexis database to identify cancer news stories in major newspapers from each of the top 50 media markets between October 2002 and March 2003. A validated search term, developed by Stryker, Wray, Hornik and Yanovitzky (2006) and replicated by Niederdeppe (2008), was used to automate the retrieval of relevant articles about cancer. We applied this search term to entire month of October 2002 (to match the local TV news sample) and a one-third sample of days between November 2002 and March 2003 (Evans & Ulasevich, 2005). The November 2002 through March 2003 sample was used to assess whether patterns of coverage in October 2002, Breast Cancer Awareness Month, were similar to those from other months. The search yielded a total of 1,156 articles for October 2002 and 1,333 articles for the 1/3 sample of days between November 2002 and March 2003. We did not conduct a parallel comparative analysis with local TV news because November 2002 through March 2003 data were not collected as part of the original UW NewsLab sample, and archival local TV news data were not available through Lexis-Nexis.

Content analysis procedures

Six undergraduate coders (all 6 for TV and 4 for print) content analyzed 258 local TV news stories and 2,489 newspaper articles about cancer. Coders were asked to determine the primary cancer type (e.g., breast, colon, lung, prostate, skin, etc.) and primary cancer topic (e.g., cause, prevention, screening, diagnosis, treatment, survivorship, death, funding, prevalence, etc.) of the story. In addition, coders were asked to identify whether or not each story (1) mentioned a cause of cancer, (2) discussed scientific research concerning cancer (including research focusing on causes, prevention, screening, or treatment), and (3) included follow-up information (e.g., phone number, email address, website, etc.) enabling viewers or readers to find additional information or resources from the report (Table 1).

Table 1.

Content Analytic Item Descriptions and Inter-Coder Reliability Kappa Scores

Variable Description Television Pre-Test (n=50) Newspaper Pre-Test (n=50) Newspaper Concurrent (n=417)
Primary Cancer Type
 Determine the primary type of cancer discussed in the story by assessing two factors: (a) prominence of the discussion of different types of cancer, and (2) quantity of coverage of different types of cancers. High prominence goes to the cancer type discussed in the headline and lead. Less prominence is given to types discussed later in the story. Determine quantity of coverage by estimating the amount of time/number of words devoted to the cancer types cancer discussed in the story. Categories include breast, colon, skin, lung, prostate, cervical, testicular, blood Leukemia) and others. 0.92 0.81 0.85
Primary Cancer Topic
 Determine the primary cancer topic discussed in the story by assessing the prominence of the discussion of different topics. The primary cancer topic is the one that is the most prominent. High prominence goes to the cancer topic discussed in the headline and the lead (the first 5 seconds of a TV story or the first two paragraphs of a print story). Categories include causes, prevention, screening, diagnosis, treatment, survivorship, death, funding, health care industry, prevalence, and other. 0.83 0.82 0.83
Cancer Cause Mentioned
 Causes of cancer include factors that increase a person’s or group’s risk of getting cancer. Examples include environmental toxins, genetic factors or family history, and behaviors (e.g., eating habits, smoking, exercise habits). Determine whether or not the story mentions causes of cancer 0.85 0.92 0.75
Cancer Research Mentioned
 Does the story report on scientific research about (1) cancer causes, (2) how to prevent cancer, (3) effective ways to screen for cancer, or (4) how to treat cancer? Determine whether the story reports on scientific research findings about cancer causes, cancer prevention, screening/early detection, or treatment. Use the definitions of cancer causes, cancer prevention, screening, and treatment outlined in the primary cancer topic variable 0.83 0.86 0.78
Follow-Up Information Provided
 Follow-up information (FI) contains complete details about people, places, and/or things that enable readers to follow up on an action encouraged by the story. To be considered FI, the details given must enable readers to complete the action incited in the story or be full source data to enable readers to do further research. It is not enough to simply mention the journal title in which an article is published. There needs to be at least one additional piece of information, such as the mention of the issue number (e.g., the September issue of the American Journal of Public Health), the author, or the article title. 0.71 0.79 0.84
Open in a new tab

Notes. Cells present kappa scores averaged between each pair of coders.

We reviewed codebooks from previous cancer news content analyses (Wang & Gantz, 2007; Nabi, 2007; Slater et al., 2008; Stryker et al., 2007) and identified relevant constructs to address study hypotheses and pilot tested these measures on a small set of TV and print stories. Through a series of iterative modifications, we arrived at a codebook that was ready for empirical reliability assessment. Before proceeding to code the full sample, each coder analyzed a sub-sample of 50 TV news stories (20%) and 50 newspaper stories (2%) for pretest inter-coder reliability. Due to the large volume of newspaper stories relative to TV stories, we double-coded an additional 417 newspaper articles (14%) after coding began to ensure that coders remained reliable in their assessments (Lombard, Snyder-Duch, & Bracken, 2002). Kappa scores for each variable and coder pair always exceeded .70 with one pair exception (.65). Across coder pairs, the average kappa scores ranged from .71 to .92 (TV pre-test), .79 to .92 (newspaper pre-test), and .75 to .85 (newspaper concurrent).

Analytic approach

Although exposure to a greater number of stories concerning cancer causes, research, and follow-up information may be consequential, direct comparisons in the volume of local TV news and newspaper stories would be misleading. The number of discrete stories on a typical 30-minute newscast is far smaller than the number of stories carried in a daily newspaper. Our analysis thus focused on the proportion of messages within each medium that discuss cancer causes, research, and follow-up. We examined the proportion of stories with a primary focus on a cancer cause or causes, those that mention a cancer cause or causes (even if not the primary focus), those that summarize scientific cancer research, and those that include follow-up information. To assess the extent to which local TV cancer coverage differed from newspaper coverage during October 2002 (H1 through H3), we conducted one-tailed unpaired t-tests. We also assessed the extent to which October 2002 coverage was representative of cancer coverage more generally throughout the year by conducting two-tailed, unpaired sample mean comparisons between print coverage in October 2002 and the one-third sample of print coverage from November 2002 through March 2003. Results from both comparisons (October 2002 TV vs. October 2002 newspaper, and October 2002 newspaper vs. November 2002 through March 2003 newspaper) are displayed in Table 2. For illustrative purposes, we also describe the content of local TV stories that discuss cancer causes.

Table 2.

Local Television and Newspaper Coverage of Cancer, Sample Mean Comparisons

Television (Oct. 2002) Newspaper (Oct. 2002) Newspaper (Nov.2002 – Mar. 2003)
Breast Cancer as Primary Cancer Type 0.67** 0.44++ 0.17
Cause as Primary Cancer Topic 0.21** 0.12++ 0.17
Cancer Cause Mentioned 0.27* 0.22 0.23
Cancer Research Mentioned 0.67** 0.36++ 0.31
Follow-Up Information Provided 0.17** 0.41 0.41
Total number of articles/stories 258 1,156 1,333
Open in a new tab

Notes. Cells indicate the percentage of stories that represent each category within each sample.

*

denotes significant differences between TV sample and October 2002 newspaper sample at p < 0.05;

**

denotes p < 0.01;

++

denotes significant differences between October 2002 newspaper sample and November 2002 to March 2003 newspaper sample at p < 0.01.

Results

Local TV cancer news stories addressed a variety of possible cancer causes. These included well-established causes like obesity, smoking, alcohol use, sun exposure, human papillomavirus (HPV) infections, genetic deficiencies and environmental factors, but also novel or controversial causes such as cellular telephone use, deodorant use, use of hormone replacement therapy, beef consumption, mold, diets rich with amino acids and with sugars, solvents used in dry cleaning, and excessive exercise. Newspaper stories described a similar set of cancer causes but covered these topics less frequently than local TV. H1 was thus supported for both primary focus (t = 3.35, p < 0.01) and for mentions of cancer causes (t = 1.79, p < 0.05; Table 2). It is also noteworthy that 62 of the 70 local TV news stories that mentioned cancer causes (88%) were reports of cancer research studies (not reported in tables). Conversely, only 160 of the 251 newspaper stories that mentioned cancer causes (64%) were reports of cancer research studies (not reported in tables).

H2 was supported. Local TV news stories were much more likely than local newspapers to focus on scientific research (t = 9.52, p < 0.01). H3 was supported. Local TV news included follow-up information less often than local newspaper coverage of cancer (t = −8.83, p < 0.01).

Comparing the two newspaper samples, breast cancer was far more likely to be the specific type of cancer covered in October compared to the other months (t = 9.98, p < 0.01). October stories were less likely than stories in November through March to primarily focus on cancer causes (t = −3.36, p < 0.01) but were equally likely to mention causes of cancer (t = −1.01, p = 0.31). October stories were more likely to focus on new research than stories in the other months (t = 2.89, p < 0.01), and there was no statistically significant difference in the inclusion of follow-up information (Table 2).

Summary of Findings

Study 1 establishes that local TV and newspaper coverage of cancer were substantively different in October 2002. Local TV stories about cancer were more likely to discuss cancer causes and cancer research but less likely to provide follow-up information than newspaper stories. These results suggest that local TV news covers cancer in ways that may be particularly likely to cultivate fatalistic beliefs about cancer prevention.

Study 2

Methods

We address H4 and H5 using data from the 2005 Annenberg National Health Communication Survey (ANHCS), the only national survey with detailed measures of both news media use and fatalistic beliefs about cancer prevention.

Survey data

ANHCS 2005, a publicly available dataset developed by researchers at the Annenberg Schools for Communication at the University of Pennsylvania and USC, was designed to capture national trends in media exposure, health cognitions, and behavior. The survey collected data from a nationally representative sample of US adults between January 20, 2005 and December 27, 2005 (ANHCS, 2007a; 2007b). ANHCS data were collected by Knowledge Networks (KN), which maintains a panel of respondents recruited via random-digit-dialing (RDD). If necessary, panel participants received internet access (via Web TV) to complete surveys online. A total of 3,625 panel members participated in the 2005 ANHCS. Our analysis used a subset of 1,783 respondents who were randomly selected to receive items pertaining to fatalistic beliefs about cancer prevention. Monthly panel recruitment response rates averaged 30.5 percent (range 26–34%) while the survey completion rate among panel members averaged 74.3 percent (range 69–77%). Study procedures and the survey instrument were approved by Institutional Review Boards at both sponsoring universities.

The unweighted analytic sample contained more women (50.3%, n = 897) than men, with a mean age of 46.70 years (range = 18–99, SD = 16.68). Most participants without missing values described themselves as Caucasian (75.3%, n = 1,783); 9.9 percent self-identified as African-American (n = 176), 10.1 percent as Hispanic (n = 180), and 4.7 percent as another race (n = 84). Data from the Current Population Survey (CPS) were used to create population weights so the data would reflect US population estimates for major demographic characteristics. We conducted all analyses with these population weights using STATA. Distribution estimates shown in the remainder of this paper use weighted percentages.

Dependent variable: Index of fatalistic beliefs about cancer prevention

ANHCS respondents were asked to report their level of agreement with two statements used in previous studies to gauge fatalistic beliefs about cancer prevention: “It seems like almost everything causes cancer” and “there are so many recommendations about preventing cancer, it’s hard to know which ones to follow” (Niederdeppe & Levy, 2007). A third item used in previous studies of fatalistic beliefs, “There’s not much people can do to lower their chances of getting cancer,” was not included in the survey. The two ANHCS items address different components of fatalism, including pessimism (“everything causes …”) and confusion (“hard to know …”). Each item was measured with a five-point Likert scale with response categories ranging from strongly disagree (1) to strongly agree (5). The two belief items were correlated (r = .37). We thus created a two-item index of fatalistic beliefs about cancer prevention (M = 7.43, range 2–10, SD = 1.89).

Independent variable: Local TV news viewing

Local TV news viewing was assessed by asking, “In the past seven days, on how many days did you watch the local news on TV?” Recent viewing was used as a proxy for habitual patterns of local TV news viewing. The average respondent watched local TV news most nights in the past week (M = 4.21, SD = 2.57).

Hypothesized moderating variable: Education

Nine response categories were provided for the question, “What is the highest level of school you completed?” Responses were coded into four categorical variables, identifying respondents that did not complete high school or equivalent (17%), completed high school but no more (31%), attended some college or technical school (27%), or graduated college (25%).

Control variables

Previous studies identify several demographic factors that are associated with fatalistic beliefs about cancer prevention (Niederdeppe & Levy, 2007; Han et al., 2007). These variables include age, sex, race/ethnicity, marital status (59% married vs. not married), working status (62% working full- or part-time vs. not working full- or part-time), household size (M = 2.7, range 1–6, SD = 1.3), frequency of religious service attendance (31% weekly, 47% rarely) and personal cancer history (whether or not has been diagnosed with cancer by a doctor; 6%). We included these variables in all multivariate models to account for their potential confounding of the relationship between local TV news viewing and fatalistic beliefs about cancer prevention.

We included a variety of health-related variables that might influence both beliefs about cancer prevention and TV viewing patterns. Self-rated health was worded as follows: “In general, would you say that your health is …” with responses including poor or very poor (5%), fair (17%), good (44%), and very good or excellent (34%). We also calculated body-mass index (BMI; a measure of overweight) by dividing self-reported weight (in kilograms) by height (in meters squared; M = 28.4, range 7–60, SD = 6.6). Finally, we accounted for smoking and drinking behaviors by asking whether a respondent had smoked a cigarette (26%) or drank five or more drinks in one sitting in the past 30 days (26%).

We also included four variables designed to measure news exposure from channels other than local TV news. These items were worded in a manner parallel to the wording of the local TV news viewing measure: In the past seven days, on how many days did you … [read a newspaper (M = 3.07, SD = 2.84); watch the national news on TV (M = 3.22, SD = 2.72); listen to radio talk shows or news (M = 2.14, SD = 2.58); and use the internet, other than email? (M = 3.44, SD = 2.81)]. We included these variables in all multivariate models to account for the possibility that local TV news viewing might be indicative of a broader pattern of news media exposure, a pattern that could confound the interpretation of any association between local TV viewing and fatalistic beliefs. We also controlled for overall TV viewing by combining responses to two questions: “On a typical weekday, about how many hours do you watch TV each day?” and “During a typical weekend, about how many total hours do you watch TV?” We created a measure to reflect the average hours of daily TV viewing (M = 5.05, range 0–16, SD = 3.51). This measure was used to control for the possibility that local TV news viewing might simply reflect a broader pattern of hours spent in front of the TV.

Analytic approach

We estimated a series of OLS regression models to test H4 and H5. We began with a series of separate OLS regression models to assess the bivariate relationships between local TV news, each control variable, and fatalistic beliefs about cancer prevention. Results from these models are shown in Table 3, second column. Next, we estimated a multiple OLS regression model, controlling for demographics, health, health behaviors, and media use, to test H4. A β coefficient for local TV news viewing that was greater than zero and statistically significant would be considered evidence in support of H4. Results from this model are displayed in Table 3, third column. Finally, we estimated a second multiple OLS regression model, including a series of interaction terms between local TV news viewing and education, to test H5. We included three interaction terms, including each of the three indicator variables for education (with college degree as reference group). A β coefficient for any of these interaction terms that was greater than zero and statistically significant would be considered evidence in support of H5. Results from this model are shown in Table 3, fourth column. We used results from this regression to calculate model-predicted values for the index of fatalistic beliefs about cancer prevention among four groups: college degree with no local TV viewing, college degree with daily local TV viewing, less than high school diploma with no local TV viewing, and less than high school diploma with daily local TV viewing. We present these values to illustrate the basic pattern of observed results related to H5.

Table 3.

Ordinary Least Squares Regression Models Testing Variables Associated With an Index of Fatalistic Beliefs about Cancer Prevention

Independent Variables Bivariate Models Multivariate Model Model with Interactions
Variables Related to Hypothesis 5
 # days watching local TV news in past week 0.07* (0.026) 0.10* (0.026) 0.11 (0.080)
Variables Related to Hypothesis 6
 Education less than high school diploma (vs. college degree) 0.09** (0.009) 0.06 (0.137) 0.11 (0.101)
 Education completed high school (vs. college degree) 0.12** (0.001) 0.09* (0.018) 0.07 (0.309)
 Education some college (vs. college degree) 0.07 (0.072) 0.05 (0.206) 0.06 (0.427)
 Education less than high school diploma X local TV news viewing -- -- −0.07 (0.342)
Education completed high school X local TV news viewing -- -- 0.02 (0.807)
 Education some college X local TV news viewing -- -- −0.02 (0.801)
Control Variables
 Age −0.05 (0.092) −0.03 (0.452) −0.03 (0.475)
 Female (vs. male) 0.02 (0.491) 0.01 (0.670) 0.01 (0.658)
 Non-Hispanic (NH) African American (vs. NH Caucasian) −0.02 (0.445) −0.06 (0.056) −0.06 (0.059)
 Hispanic (vs. NH white) −0.04 (0.202) −0.05 (0.086) −0.05 (0.075)
 NH other (vs. NH white) −0.06 (0.106) −0.03 (0.404) −0.03 (0.415)
 Working full- or part-time 0.03 (0.383) 0.08* (0.041) 0.08* (0.042)
 Married −0.03 (0.333) −0.01 (0.779) −0.01 (0.779)
 Household size 0.03 (0.356) 0.02 (0.616) 0.02 (0.624)
 Self-rated poor or very poor health (vs. very good or excellent) 0.09** (0.002) 0.08* (0.016) 0.08* (0.016)
 Self-rated fair health (vs. very good or excellent) 0.08* (0.018) 0.06 (0.091) 0.07 (0.082)
 Self-rated good health (vs. very good or excellent) 0.07* (0.035) 0.06 (0.069) 0.06 (0.081)
 Body-mass index (BMI) 0.04 (0.210) 0.02 (0.626) 0.02 (0.590)
 Smoking in past 30 days 0.09** (0.001) 0.05 (0.092) 0.05 (0.101)
 Binge drinking in past 30 days 0.08** (0.002) 0.05 (0.105) 0.05 (0.095)
 Average daily hours of TV viewing 0.09** (0.002) 0.08** (0.007) 0.08** (0.006)
 # days reading newspaper in past week −0.04 (0.187) 0.00 (0.896) −0.01 (0.866)
 # days watching national TV news in past week −0.01 (0.719) −0.06 (0.157) −0.07 (0.147)
 # days listening to radio news in past week −0.05 (0.074) −0.05 (0.101) −0.05 (0.105)
 # days using internet (non-email) in past week −0.01 (0.787) 0.00 (0.906) 0.01 (0.825)
R squared -- 0.06 0.06
Number of observations 1,783 1,633 1,633
Open in a new tab

Notes. Cells contain β coefficient estimates and p-values in parentheses.

*

denotes β coefficients different from zero at p < 0.05;

**

p < 0.01. Each model used population weights to adjust estimates for non-response and sampling non-coverage. Personally had cancer and weekly religious attendance were not significant predictors of the index and were dropped because their inclusion substantially increased the number of missing cases in the multivariate model.

Results

H4 was supported. Local TV news viewing was positively associated with the index of fatalistic beliefs about cancer prevention, in both a bivariate model (β = 0.07, p < 0.05) and in a multivariate model that controlled for socio-demographic factors (β = 0.10, p < 0.05). Other significant predictors of the index in the multivariate model included having only a high school diploma (β = 0.09, p < 0.05), working full- or part-time (β = 0.08, p < 0.05), having poor or very poor self-rated health (β = 0.08, p < 0.05), and average daily TV viewing (β = 0.08, p < 0.01; Table 3).

H5 was not supported. All three interaction terms between local TV news viewing and categories of education were non-significant, and two were negative (wrong direction; Table 3). Model-predicted values for respondents with a college degree ranged from 6.69 (no local TV viewing) to 7.44 (daily local TV viewing), while predicted values for respondents with less than a high school diploma ranged from 7.42 (no local TV viewing) to 7.68 (daily local TV viewing).

Discussion

Results from content and survey analyses make plausible the claim that local TV cancer news coverage may cultivate fatalistic beliefs about cancer prevention. Contrary to studies that assume that TV content or news media are monolithic, we found evidence that suggests local TV news coverage may be particularly harmful. More than one in five local TV cancer news stories focused on a cancer cause or causes, and many of the causes described were novel or controversial and yet likely to be relevant to many individuals (e.g., cell phone use, beef consumption, dry cleaning, exercise). Almost all local TV stories that mentioned cancer causes were reports of scientific research. More broadly, two out of three local TV cancer news stories discussed research findings. This is particularly concerning because newsworthy cancer studies tend to be those that report on novel or controversial findings (Stryker, 2002), perhaps at the expense of well-documented causes and known prevention methods (Russell, 1999). Follow-up information was rarely included. While newspapers also reported on cancer causes and research, they did so with far less frequency than local TV and provided information that would enable readers to assess the source of the information nearly half of the time. Specifically, local TV news stories were more likely than newspaper stories to focus on and discuss causes of cancer (supporting H1), more likely to discuss scientific research findings (supporting H2), and less likely to include information that would allow viewers to follow-up by seeking out additional resources, guidance, or advice regarding the coverage they watched (supporting H3).

These patterns suggest a tendency for local TV news to focus on aspects of cancer that are likely to cultivate the beliefs that everything causes cancer or that there are too many recommendations about cancer prevention. Informed by cultivation theory, we hypothesized that local TV viewing would be associated with these beliefs. National survey data analyses provided evidence consistent with this assertion. Local TV news viewing was positively associated with an index of fatalistic beliefs about cancer prevention, controlling for socio-demographic factors (supporting H4). Building on the knowledge gap hypothesis, we also hypothesized that local TV news viewing would be more strongly associated with fatalistic beliefs among less educated populations. Statistical tests failed to support this assertion (contrary to H5).

Revisiting Cultivation Theory

Findings suggest that content-specific extensions of cultivation theory, particularly those focused on local TV news, are useful additions (e.g., Lowry et al., 2003; Romer et al., 2003). At the same time, broader patterns of TV content may also be consequential in cultivating fatalistic beliefs about cancer prevention. We observed a significant relationship between overall TV viewing and fatalistic beliefs about cancer prevention, controlling for local TV viewing. There is no shortage of attention to health topics on TV, in the form of soap operas in medical settings (e.g., General Hospital), prime time dramas about doctors and hospitals (e.g., House, Grey’s Anatomy, E. R.), talk-shows that periodically interview cancer survivors (e.g., Oprah), or advertisements for products, services, or fundraisers related to cancer (e.g., cancer screening services at a local hospital). Content within these programs may shape beliefs about cancer prevention. Alternatively, TV viewing may simply be associated with broader personality traits that are also associated with these beliefs, such as locus of control (Potter, 1987). Future studies should examine references to cancer causes and research on entertainment programs or advertisements that appear during these programs or news broadcasts. These efforts would inform the question of whether broader patterns of exposure to TV’s general treatment of cancer may contribute to fatalistic beliefs about cancer prevention.

More broadly, our findings suggest that content-specific extensions to cultivation theory might be best conceptualized as a complement to, rather than a replacement for, traditional cultivation analysis. Overarching patterns of TV content likely shape social beliefs, but specific genres of content may be more or less consequential than others. Content-specific extensions do provide greater practical guidance for those in the public health arena who may wish to intervene to correct misperceptions about cancer prevention. For instance, one could envision an intervention to train local TV news reporters to provide follow-up information and place new research findings into the context of established recommendations for cancer prevention, or a scenario where health communication researchers work with TV writers and producers to accurately portray health content in TV dramas. Both types of projects are currently underway in the US, including projects at USC to train journalists on how to cover health (California Endowment Health Journalism Fellowships, 2008) and to provide entertainment industry professionals with accurate and timely information for health storylines (Health, Hollywood, & Society, 2008). The extension of cultivation analysis to focus on differential effects of specific genres of TV content (e.g., local TV news, TV dramas) may provide a useful theoretical backdrop for interventions designed to shape health content within these genres.

Revisiting the Knowledge Gap Hypothesis

This paper also built on the tradition of knowledge gap research by examining educational differences in response to news media content. Although some have suggested that TV news may reduce knowledge gaps by providing information to the least educated in ways that are comparably easy to process (Kwak, 1999), this assertion rests on the assumption that TV news provides useful information capable of attenuating knowledge gaps. The accumulated evidence, including content analytic results presented here, suggests that local TV news may not always provide information that is useful in this regard. At the same time, we found no evidence that local TV news viewing was more strongly associated with fatalistic beliefs about cancer prevention among lower versus higher educated populations. In fact, although not statistically significant, comparisons in model-predicted values for the index of fatalistic beliefs between college educated respondents and those without a high school diploma suggest that the relationship between local TV viewing and fatalism could be weaker among those with lower education. Either way, differential response to local TV news does not appear to account for the association between low education and higher fatalistic beliefs about cancer prevention. Future extensions of knowledge gap research, particularly those focused on TV as a channel with the potential to reduce or increase gaps, would benefit by distinguishing between types of content (e.g., local news, national news, talk-shows) and examining the effects of each content type among socioeconomically diverse groups.

Study Limitations

There are several study limitations worth noting. First, the timing of the local TV content analysis (2002) was not concurrent with the timing of the national survey (2005). It is possible that local TV news coverage about cancer has changed since 2002 to focus less on cancer causes and research. Such a change would undermine the argument that local TV news coverage is substantively different from newspaper coverage. The Project for Excellence in Journalism (2007) recently noted, “if TV stations are innovating or improving their journalism, most viewers say they haven’t noticed it” (p. 2), suggesting that local TV news coverage has not dramatically improved. This assertion has not, however, been empirically tested.

Second, our sample of local TV news coverage was restricted to a single month, October 2002. Cultivation analysis typically monitors longer-term trends in televised content (see Gerbner et al., 1980). October 2002 was also Breast Cancer Awareness Month and immediately preceded the 2002 midterm elections, raising questions about the degree to which the sample is representative of broader patterns of local TV news coverage. Newspaper coverage data suggest that, for at least one type of media, Breast Cancer Awareness Month did not drastically influence patterns of coverage most relevant for this study. While the proportion of newspaper stories focused on breast cancer was substantially higher in October 2002 relative to November 2002 through March 2003, the proportion of stories mentioning (but not focusing) on cancer causes and reporting follow-up information were equivalent. These results suggest that media channels may be consistent in their pattern of coverage of these issues. We did find differences in mentions of research, such that print coverage of research decreased after October 2002. This pattern would suggest an even greater discrepancy between local TV and print news on this dimension. We believe the local TV news coverage data, despite their limitations, are of unique value given the high costs of collecting national samples of local TV news content (see Slater et al., 2008; Wang & Gantz, 2007). To our knowledge, this study represents the largest sample of local TV coverage about cancer ever collected, in terms of population coverage.

Third, reliance on cross-sectional associations does not permit causal claims about the relationship between local TV news coverage and fatalistic beliefs about cancer prevention. While the inclusion of content analytic data showing systematic differences between local TV and newspaper coverage of cancer suggests the plausibility of a causal relationship, it is possible that differences between local TV news viewers and non-viewers explain the observed associations, regardless of local TV news content. Somewhere in between, Slater (2007) suggests such a relationship might be usefully characterized as a reinforcing spiral, where different types of audiences select different media, which in turn influence their beliefs, which in turn influence their selection of media, and so on. Our cross-sectional data are far too limited to make such a claim, but they do suggest promising avenues for future research involving longitudinal designs.

Future Directions

While concurrent survey and content analysis would enable us to more accurately assess cancer news exposure as a function of both individual viewing habits and actual content of messages being aired (see Freedman & Goldstein, 1999), our representative sample of local TV cancer coverage from the top 50 media markets reveals that traditional survey measures of media use may be insufficient to capture the variation in cancer messages actually airing. Specifically, although the media market within which an individual resides is essential information in determining what media messages the individual might receive, there is substantial heterogeneity in the volume and content of cancer messages airing on local TV stations both between and within the same media market (Figure 1). This suggests that market-level analysis may not be sufficient to adequately assess the volume of messages to which citizens may be exposed. In other words, future work on the effect of local TV news coverage should account for which station’s local TV news an individual watches in addition to the market within which he or she resides. This variation also highlights the need for research explaining the factors influencing variation in health and cancer coverage between and within markets (see Pollock, 2007).

Figure 1.

Figure 1

Open in a new tab

Variance in Volume of Local Television News Cancer Coverage Between and Within the Top 50 Media Markets

Notes. UW NewsLab captured between one and four of the four major affiliate stations (ABC, CBS, Fox, and NBC) per market during October 2002. Bars represent the range in volume of cancer coverage between all captured local television stations within the same media market for the month, and dots represent cancer coverage on each individual stations captured. Markets are order by size from the largest (1), New York, to the smallest of the top 50 (50), Louisville.

Future work should also consider the appropriate time lag necessary between actual airing of cancer messages and when such coverage is likely to affect individual beliefs. Longitudinal panel data, supplemented with content analysis, may be especially useful in this regard. Finally, in comparison to local newspaper archives, the cost of obtaining local TV news data has traditionally been prohibitively expensive, and although changing technology is beginning to make the capture and analysis of this content more accessible to scholars (Hale et al., 2007), studies of local TV remain few and far between. While the costs of local TV research still exceed other content, given local TV newscasts remain a major source of information in the U.S. (Pew Center for the People and the Press, 2006), more work should examine both the content and effects of local TV news.

Conclusions

This study provides evidence consistent with the claim that local TV news coverage may promote fatalistic beliefs about cancer prevention. While associations between local TV news viewing and fatalistic beliefs were cross-sectional in nature, content analytic results showing notable differences between local TV and newspaper coverage of cancer suggest the plausibility of a causal relationship. Findings extend cultivation theory to focus on the effects of particular types of TV content on beliefs about cancer prevention and add to a growing body of evidence about local TV news and knowledge gaps. Future studies should make use of between- and within-market variation in local TV coverage of cancer, longitudinal survey data, and content analysis of local TV news content to provide stronger causal tests of the relationship between local TV news and fatalistic beliefs about cancer prevention.

Acknowledgments

The authors are grateful to the Robert Wood Johnson Foundation Health and Society Scholars Program at the University of Wisconsin and the Robert Wood Johnson Foundation Health Policy Scholars Program at the University of Michigan for funding the study. Funding for the survey data analysis was also provided by NIH P50 Grant CA095856 to the Annenberg School for Communication at the University of Pennsylvania. We also thank the Annenberg Trust at Sunnylands for funding the survey data collection and the Annenberg Schools at the University of Pennsylvania and University of Southern California for developing the survey. We would also like to thank Paula Lantz and Peter Ubel for their comments on an earlier version of this paper, and Michael Cody and two anonymous reviewers for their insightful comments on improving the clarity of the manuscript.

Footnotes

Publisher's Disclaimer: This is the final peer-reviewed version of this manuscript that was accepted for publication but will still be subject to copyediting by Journal of Communication editors. The definitive version is available at www.blackwell-synergy.com.

References

  1. Aldeman RC, Verbrugge LM. Death makes news: The social impact of disease on newspaper coverage. Journal of Health and Social Behavior. 2000;41:347–367. [PubMed] [Google Scholar]
  2. Annenberg National Health Communication Survey. ANHCS 2005 user’s guide and codebook. 2007a Retrieved September 23, 2008, from http://anhcs.asc.upenn.edu.
  3. Annenberg National Health Communication Survey. ANHCS 2005 data set. 2007b Retrieved September 23, 2008, from http://anhcs.asc.upenn.edu.
  4. California Endowment Health Journalism Fellowships. California endowment health journalism fellowships. 2008 Retrieved September 23, 2008, from http://www.reportingonhealth.com.
  5. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of US adults. New England Journal of Medicine. 2003;348:1625–1638. doi: 10.1056/NEJMoa021423. [DOI] [PubMed] [Google Scholar]
  6. Chory-Assad RM, Tamborini R. TV exposure and the public’s perceptions of physicians. Journal of Broadcasting & Electronic Media. 2003;47:197–215. [Google Scholar]
  7. Dean WC, Pertilla A. The reality of local TV news. In: Rosenstiel T, Just M, Belt T, Pertilla A, Dean W, Chinni D, editors. We interrupt this newscast: How to improve local news and win ratings, too. Cambridge: Cambridge University Press; 2007. [Google Scholar]
  8. Donohue GA, Tichenor PJ, Olien CN. Mass media and the knowledge gap: A hypothesis reconsidered. Communication Research. 1975;2:3–23. [Google Scholar]
  9. Evans D, Ulasevich A. News media tracking of tobacco control: A review of sampling methodologies. Journal of Health Communication. 2005;10:403–417. doi: 10.1080/10810730591009781. [DOI] [PubMed] [Google Scholar]
  10. Fowler EF, Goldstein K, Hale M, Kaplan M. Does local news measure up? Stanford Law & Policy Review. 2007;18:410–431. [Google Scholar]
  11. Fowler EF, Goldstein K, Shah D. The challenge of measuring news consumption. In “Roundtable: New Media Measure for ANES? Political Communication Report. 2008 Winter;:2008. [Google Scholar]
  12. Freedman P, Goldstein K. Measuring media exposure and the effects of negative campaign ads. American Journal of Political Science. 1999;43:1189–1208. [Google Scholar]
  13. Gerbner G, Gross L, Morgan M, Signorielli N. The “mainstreaming” of America: Violence profile no. 11. Journal of Communication. 1980;30(3):10–29. doi: 10.1111/j.1460-2466.1979.tb01731.x. [DOI] [PubMed] [Google Scholar]
  14. Gerbner G, Gross L, Morgan M, Signorielli N. What TV teaches about physicians and health. Möbius. 1982;2:44–51. doi: 10.1002/chp.4760020209. [DOI] [PubMed] [Google Scholar]
  15. Gerbner G, Gross L, Morgan M, Signorielli N, Shanahan J. Growing up with TV: Cultivation processes. In: Bryant J, Zillman D, editors. Media effects: Advances in theory and research. Mahwah, NJ: Lawrence Erlbaum; 2002. pp. 43–67. [Google Scholar]
  16. Gilliam FD, Iyengar S. Prime suspects: The influence of local TV news on the viewing public. American Journal of Political Science. 2000;44:560–573. [Google Scholar]
  17. Gross K, Aday S. The scary world in your living room and neighborhood: Using local broadcast news, neighborhood crime rates, and personal experience to test agenda setting and cultivation. Journal of Communication. 2003;53:411–426. [Google Scholar]
  18. Hale M, Fowler EF, Goldstein K. Capturing multiple markets: A new method for capturing and analyzing local television news. Electronic News. 2007;1:227–243. [Google Scholar]
  19. Han PKJ, Moser RP, Klein WMP. Perceived ambiguity about cancer prevention recommendations: Associations with cancer-related perceptions and behaviors in a U.S. population survey. Health Expectations. 2007;10:321–336. doi: 10.1111/j.1369-7625.2007.00456.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Health, Hollywood, and Society. Overview. 2008 Retrieved September 23,2008, from http://www.learcenter.org/html/projects/?cm=hhs.
  21. Hetsroni A, Tukachinsky RH. TV-world estimates, real-world estimates, and TV viewing: A new scheme for cultivation. Journal of Communication. 2006;56:133–156. [Google Scholar]
  22. Kwak N. Revisiting the knowledge gap hypothesis. Communication Research. 1999;26:385–413. [Google Scholar]
  23. Lombard M, Snyder-Duch J, Bracken CC. Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research. 2002;28:587–604. [Google Scholar]
  24. Lowry DT, Nio TCJ, Leitner DW. Setting the public fear agenda: A longitudinal analysis of network TV crime reporting, public perceptions of crime, and FBI crime statistics. Journal of Communication. 2003;53:61–73. [Google Scholar]
  25. Moy P, McCluskey MR, McCoy K, Spratt M. Political correlates of local news media use. Journal of Communication. 2004;54:532–546. [Google Scholar]
  26. Nabi RL. News broadcast codebook. University of California; Santa Barbara: 2007. Unpublished manuscript. [Google Scholar]
  27. Neuman R. Patterns of recall among television news viewers. Public Opinion Quarterly. 1976;40:115–123. [Google Scholar]
  28. Niederdeppe J. Beyond knowledge gaps: Examining socioeconomic differences in response to cancer news. Human Communication Research. 2008;34:423–447. [Google Scholar]
  29. Niederdeppe J, Levy AG. Fatalistic beliefs about cancer prevention and three prevention behaviors. Cancer Epidemiology, Biomarkers, and Prevention. 2007;16:998–1003. doi: 10.1158/1055-9965.EPI-06-0608. [DOI] [PubMed] [Google Scholar]
  30. Patterson TE, McClure RD. The unseeing eye: The myth of TV power in national elections. New York: G. P. Putnam’s Sons; 1976. [Google Scholar]
  31. Peto R, Lopez AD, Boreham J, Thun M, Heath CJ. Mortality from smoking in developed countries, 1950–2000: Indirect estimates from national vital statistics. New York: Oxford University Press; 1994. [Google Scholar]
  32. Pew Center for the People and the Press. Online papers modestly boost newspaper readership: Maturing internet news audience broader than deep. 2006 Retrieved September 23, 2008, from http://people-press.org/reports/pdf/282.pdf.
  33. Pollock JC. Tilted mirrors: Media alignment with political and social change: A community structure approach. Cresskill, NJ: Hampton Press; 2007. [Google Scholar]
  34. Potter D, Kurpius DD. Public journalism and TV news. In: Eksterowicz AJ, Roberts RN, editors. Public journalism and political knowledge. Lanham, MA: Rowman & Littlefield; 2000. pp. 77–90. [Google Scholar]
  35. Potter WJ. Does television viewing hinder academic achievement among adolescents? Human Communication Research. 1987;14:27–46. [Google Scholar]
  36. Pribble JM, Goldstein KM, Fowler EF, Greenberg MJ, Noel SK, Howell JD. Medical news for the public to use? What’s on local TV news. American Journal of Managed Care. 2006;12:170–176. [PubMed] [Google Scholar]
  37. Project for Excellence in Journalism. Local TV. The state of the news media: An annual report on American journalism. 2007 Retrieved September 23, 2008, from http://stateofthemedia.org/2007/chapter%20pdfs/PEJ2007-LocalTV.pdf.
  38. Robinson JP, Davis DK. TV news and the informed public: An information-processing approach. Journal of Communication. 1990;40(3):106–119. [Google Scholar]
  39. Romer D, Jamieson KH, Aday S. TV news and the cultivation of fear of crime. Journal of Communication. 2003;53:88–104. [Google Scholar]
  40. Russell C. Living can be hazardous to your health: How the news media cover cancer risks. Journal of the National Cancer Institute Monographs. 1999;25:167–170. doi: 10.1093/oxfordjournals.jncimonographs.a024194. [DOI] [PubMed] [Google Scholar]
  41. Slater MD. Reinforcing spirals: The mutual influence of media selectivity and media effects and their impact on individual behavior and social identity. Communication Theory. 2007;17:281–303. [Google Scholar]
  42. Slater MD, Long M, Bettinghaus EP, Reineke JB. News coverage of cancer in the U.S.: A national sample of newspapers, TV and magazines. Journal of Health Communication. 2008;13:523–537. doi: 10.1080/10810730802279571. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Slenker SE, Spreitzer EA. Public perceptions and behaviors regarding cancer control. Journal of Cancer Education. 1998;3:171–180. doi: 10.1080/08858198809527935. [DOI] [PubMed] [Google Scholar]
  44. Straughan PT, Seow A. Fatalism reconceptualized: A concept to predict health screening behavior. Journal of Gender, Culture, & Health. 1998;3:85–100. [Google Scholar]
  45. Stryker JE. Reporting medical information: Effects of press releases and newsworthiness on medical journal articles’ visibility in the news media. Preventive Medicine. 2002;35:519–530. doi: 10.1006/pmed.2002.1102. [DOI] [PubMed] [Google Scholar]
  46. Stryker JE, Emmons KM, Viswanath K. Uncovering differences across the cancer control continuum: A comparison of ethnic and mainstream cancer newspaper stories. Preventive Medicine. 2007;44:20–25. doi: 10.1016/j.ypmed.2006.07.012. [DOI] [PubMed] [Google Scholar]
  47. Stryker JE, Wray R, Hornik RC, Yanovitzky I. Validation of database search terms for content analysis: The case of cancer news coverage. Journalism and Mass Communication Quarterly. 2006;83:413–430. [Google Scholar]
  48. Taubes G. Epidemiology faces its limits. Science. 1995;269:164–169. doi: 10.1126/science.7618077. [DOI] [PubMed] [Google Scholar]
  49. Tichenor PJ, Donohue GA, Olien CN. Mass media flow and differential growth in knowledge. Public Opinion Quarterly. 1970;34:159–170. [Google Scholar]
  50. Viswanath K. Public communications and its role in reducing and eliminating health disparities. In: Thomson GE, Mitchell F, Williams M, editors. Examining the health disparities research plan of the National Institutes of Health: Unfinished business. Washington, DC: Institute of Medicine; 2006. pp. 231–270. [PubMed] [Google Scholar]
  51. Viswanath K, Breen N, Meissner H, Moser RP, Hesse B, Steele WR, et al. Cancer knowledge and disparities in the information age. Journal of Health Communication. 2006;11:1–17. doi: 10.1080/10810730600637426. [DOI] [PubMed] [Google Scholar]
  52. Viswanath K, Finnegan JR. The knowledge gap hypothesis: Twenty-five years later. In: Burleson B, editor. Communication Yearbook 19. Thousand Oaks, CA: Sage; 1996. pp. 187–227. [Google Scholar]
  53. Wang Z, Gantz W. Health content in local TV news. Health Communication. 2007;21:213–221. doi: 10.1080/10410230701307527. [DOI] [PubMed] [Google Scholar]
  54. Zaller J. A theory of media politics: How the interests of politicians, journalists, and citizens shape the news. 1999 Retrieved September 23, 2008, from http://www.sscnet.ucla.edu/polisci/faculty/zaller/media%20politics%20book%20.pdf.

RESOURCES