Modifying perceptions of hostility and credibility of news coverage of an environmental controversy through media literacy

Abstract

This study tests the proposition that hostile interpretations of media content can be reduced through news media literacy training. Within the context of the controversy over the adoption of biofuels as an energy source, we employ a web-based experimental design that manipulates subjects’ exposure to media literacy training and then presents them with news coverage on the issue of biofuels. We find strong support for the notion that media literacy affects individuals’ perceptions of media credibility. Exposure to a media literacy video led to increased ratings of story credibility, as well as increased trust in the media to cover both the issue and the news more broadly. Implications of these results are discussed.

Keywords

credibility perceptions hostile media effect media literacy

Claims of media bias are the order of the day. Diverse social commentators, advocacy groups, and political organizations decry that news coverage is not only biased, but biased against their side. Americans’ assessments of news accuracy and fairness have reached the lowest levels in decades. The Pew Research Center for the People and the Press estimates that in 2009 only 29 percent of Americans trusted the media to ‘get the facts straight’, and only 18 percent felt the media ‘deals fairly with all sides’ – the lowest figures in a downward trend since 1985 (Pew, 2009).

Part of this frustration with news media may be related to the reconstitution of the media environment created by new communication technologies and the impact of these technologies on news ownership, production, and distribution. However, another component of this reduced trust in media can be traced not to biased news reporting, but rather to citizens’ biased processing of news (Gunther, 1992).

While news outlets are often accused of being biased toward one side or another, empirical evidence shows that individuals’ perceptions, experiences, and involvement with an issue affect perceptions of ‘bias’ even in an ‘unbiased’ news story. This tendency for people, particularly those with strong feelings on an issue, to perceive articles on that issue as biased against their view has been conceptualized in the literature as the hostile media effect, or HME (Gunther and Schmitt, 2004; Perloff, 1989; Vallone et al., 1985).

Extant research has broadened our understanding of when and why these hostile interpretations of media occur, but little is known about how these biased perceptions can be prevented. The normative importance of reducing these perceptions of bias cannot be overstated. Democracies depend on people being able to rely on the news for information, but if audiences dismiss information in the media because it is hostile to their views, content that could increase knowledge and acceptance can instead lead to greater misunderstanding and polarization (Lord et al., 1979; Perloff, 1989). As a result, individuals may reinforce their own viewpoints while criticizing the other side and the media for hostility and bias.

To address this problem, the current study explores how hostile interpretations of media content can be reduced. Specifically, this research tests whether exposure to a media literacy video, which highlights the news process, its integrative efforts, and its importance for democracy, can reduce perceptions of bias and enhance news media credibility.

Media literacy education has taken on many forms in the United States. One strand in particular has focused on the importance of media literacy for participation in democratic practices (Kellner and Share, 2005), suggesting that a better understanding of how the media operate and of their role in society should increase perceptions of media credibility by ‘teaching’ audiences that coverage of a variety of views, even those opposed to their own, is necessary for a functioning democracy. Further, if a short media literacy training video can limit perceptions of hostility and boost perceptions of credibility, issue-specific media trust, and general media trust, then more sustained efforts, such as formal and informal education, for example, could lead to a more lasting understanding of and trust in the media (Hobbs, 2010).

To test this proposition, we examine the issue of biofuels, fuels derived from plants, which advocates have heralded as a clean energy solution that would reduce humans’ impact on climate change. However, critics question the efficiency of existing biofuel technologies, ethanol in particular, and point to the potential negative impacts of massive biofuel production on the food supply, as land and agricultural products become ‘bio-crops’.

The controversy over the utility of widespread biofuel adoption is particularly well suited for our purposes, considering this is an issue for which most people have limited access to non-mediated information and one that gained national prominence when energy became an important issue in the 2008 presidential campaign. But it is also an issue for which traditional heuristics for issue positions, such as political party identification, are not well developed. Thus, the biofuels controversy provides us with an opportunity to focus more closely on the relationship between media literacy and subsequent information processing. In doing so, this study does not advocate for an uncritical interpretation of news, but rather investigates how an improved understanding of the news process and its role in society can reduce biased processing.

Literature review

The hostile media effect

Research into the hostile media effect began with the idea that people demonstrate biases in processing information. Early research into biased assimilation, or the ability of exposure to a two-sided message to produce polarization, indicated that individuals engage in motivated reasoning (Kunda, 1990; Lord et al., 1979) and actively counter-argue against information that disagrees with their attitudes (Kunda, 1990; Taber and Lodge, 2006).

Research into the HME developed around this idea of biased processing, looking specifically at how people respond to news coverage of controversial issues. With media, we see a reversal of this trend to interpret content as supportive of our own views: partisans, defined as those with strong views about or deep involvement with an issue, consistently rate news coverage as biased against their position, even when neutral observers rate the same coverage as unbiased (Gunther, 1992; Perloff, 1989; Vallone et al., 1985).

The endurance and strength of HME has been demonstrated through the variety of topics and techniques by which it has been observed. Including both experimental and survey research and spanning different contexts and topics, partisans rate news coverage as hostile to their position (D’Alessio, 2003; Gunther, 1988, 1992; Gunther and Schmitt, 2004; Vallone et al., 1985). Furthermore, it applies regardless of whether individuals are asked to rate a particular story (D’Alessio, 2003; Perloff, 1989; Vallone et al., 1985) or to give their perceptions of the media more generally (Eveland and Shah, 2003; Gunther, 1988, 1992). That HME has been observed in such a range of settings speaks to its robustness and pervasiveness.

This widespread occurrence makes it necessary to understand the mechanisms and conditions that produce HME. Selective categorization of information appears to be the central mechanism by which HME occurs, as people with different perspectives rate the same information as hostile to their views (Gunther and Liebhart, 2006; Schmitt et al., 2004). In addition, HME is contingent on specific message characteristics: a more professional source (i.e. a journalist compared to a student) or a larger expected audience for the message both produce greater HME (Gunther and Liebhart, 2006; Gunther and Schmitt, 2004). In other words, individuals are more likely to perceive bias in a message when they assume others will see it, suggesting that expected influence is another mechanism producing HME (Perloff, 1989).

Beyond message characteristics, individual differences play a role in predicting HME. Research has consistently demonstrated that strength of involvement with an issue is one of the strongest predictors of hostile perceptions (Gunther, 1992; Perloff, 1989). Involvement encourages categorization of information as hostile to one’s position, based on differing perceptions of ‘neutrality’. Personal predispositions that foster hostile media perceptions include attitude extremity, interest, partisanship, and political ideology (Eveland and Shah, 2003; Gunther, 1992; Perloff, 1989; Schmitt et al., 2004; Vallone et al., 1985). Many HME studies seek out respondents who have opposing positions on an issue, demonstrating that partisans on both sides rated the messages as biased (Gunther and Schmitt, 2004; Perloff, 1989; Vallone et al., 1985). A strong position and emotional engagement produce hostile interpretations of media messages and their source (Gunther, 1992).

News credibility

The HME literature focuses on perceptions of hostility in media content, but hostility is only one component in a range of media perceptions. Extensive research into perceptions of the media has developed the concept of media credibility, which combines ratings of bias, trustworthiness, balance, accuracy, completeness, and fairness (Fico et al., 2004; Meyer, 1988). A story is considered ‘credible’ when respondents trust the content because it tells the whole story fairly and without bias.

Although much of the research into the hostile media effect has dealt specifically with ratings of bias, some research has expanded this definition to deal with concepts like trust and fairness. This research suggests that these concepts fit nicely with HME: partisans tend to rate the media as less trustworthy and fair (D’Alessio, 2003; Eveland and Shah, 2003; Gunther, 1988, 1992; Perloff, 1989; Vraga et al., 2009). These findings suggest that when considering the HME, it is important to look beyond perceptions of hostility toward the specific content because multiple components of credibility may be implicated in the HME.

While perceptions of hostility and credibility have largely been examined separately (D’Alessio, 2003; Eveland and Shah, 2003; Fico et al., 2004; Gunther, 1988; Meyer, 2004), it seems plausible that these two ideas are more closely linked than previous literature has generally acknowledged (Choi et al., 2006; Tsfati and Cohen, 2005). Some studies have begun exploring this possibility. For example, Tsfati and Cohen (2005) demonstrate that hostile media perceptions about the Gaza settlements are negatively linked to trust in media, which further impacted trust in democracy. Furthermore, while ratings of issue-specific media trust can differ from ratings of the media as a whole, they remain closely linked (Gunther, 1992; Hwang et al., 2008). Thus, before testing the effects of media literacy, we are interested in establishing whether perceptions of story bias, story credibility, trust in the media to cover an issue and general media trust are indeed empirically different constructs (Research Question 1).

In looking both at perceptions of story-specific credibility and at broader media credibility, we can also begin to explore whether story-specific perceptions lead to more general perceptions of media trust, or, on the contrary, whether broader media level perceptions serve as conduits to improved ratings of a particular story (Research Question 2). Previous research has suggested that this relationship can occur in either direction: that perceptions of a specific story can feed into perceptions of the media at large (Perloff, 1989; Rouner et al., 1999) or that perceptions of the media can alter our views about a particular story, particularly for issues in which people are deeply involved (Chia et al., 2007; Giner-Sorolla and Chaiken, 1994).

Social judgment theory

To explain how exposure to media literacy training could reduce the biases that cause HME, we draw on social judgment theory (Gunther, 1988; Stamm and Dube, 1994).¹ According to this theory, individuals judge messages depending on whether they fall into the latitude of acceptance, non-commitment, or rejection (Eagly and Chaiken, 1993; Sherif and Hovland, 1961). Ego involvement plays a fundamental role in determining where these latitudes occur, shifting people’s response to a message. If HME functions because partisans differently categorize information as hostile to their beliefs (Gunther and Liebhart, 2006; Schmitt et al., 2004), this closely aligns with social judgment theorists’ understanding about acceptance versus rejection of a message.

Therefore, the media literacy training could function by adjusting the latitudes of acceptance, non-commitment, and rejection for partisans on an issue. By highlighting the role of the media and the audience in promoting engaged citizenship and informed political debate, the training should render the news story, including its incongruent viewpoints, more acceptable by lessening the latitude of rejection for the message because it is when the message falls within this latitude of rejection that people should perceive the message as hostile or not credible.

Media literacy

If people’s perceptions of media credibility and bias are related to their position on an issue, it would appear that journalists can do little to reduce hostile interpretations of their work. However, a better understanding of how the media function and of their integrative role in society should increase perceptions of media credibility by highlighting that coverage of a variety of views is necessary for democracy and per se does not entail media bias. In fact, research suggests this type of training can reduce perceptions of bias in a subsequent news story (Vraga et al., 2009).

Although media literacy is not an easily defined concept (Hobbs and Jensen, 2009), media literacy education often focuses on teaching students how to analyze and deconstruct media messages and understand the process of message construction (Brown, 1998; Hobbs and Frost, 2003; Kellner and Share, 2005). Research suggests that a combination of general and issue-specific media literacy training can have immediate and enduring effects (Austin and Johnson, 1997; Livingstone, 2004), leading to more analytical evaluation of media messages (Tobias, 2008).

Some media literacy education deals specifically with news construction and consumption, developing the notion that critically evaluating the news is required for audiences to assess the credibility of information. In an early work on media literacy, Masterman argued that ‘Widespread media literacy is essential if all citizens are to wield power, make rational decisions, become effective change-agents, and have an effective involvement with the media’ (1985: 13). More recently, Mihailidis (2008a, 2008b) has argued that media literacy education should focus on informed and engaged citizenship. The media literacy approach in this study follows this line of thinking, highlighting the relationship between media, democracy and engaged citizenship and extending our previous research into the potential of media literacy for reducing HME. Mihailidis (2008b) makes the case for media literacy education that stresses the value of an independent and free press as an essential part of democracy, and, similarly, we highlight these elements in the media literacy stimulus in this study.

Thus, we argue that citizens need to understand the news process and imperatives behind journalistic work to analyze the news effectively. We propose that the lack of understanding of news processes can contribute to decreased perceptions of media credibility for the wrong reason: not based on how the media actually fare in their reporting, but rather through personal biases. Informing individuals of journalists’ goals and roles in a democratic society should increase their knowledge of the context in which media content is produced and this, in turn, may reduce hostile media perceptions and increase credibility perceptions. Previous research suggests media literacy training can achieve this potential: in one study, a media literacy presentation extolling the importance of a diverse press and the role of its audiences limited perceptions of hostility in a subsequent news story (Vraga et al., 2009). Therefore, we expect that exposure to media literacy training can play a role in reducing hostile interpretations of the specific media content and boost ratings of story credibility (Hypothesis 1).

But media literacy training will be less useful if it only affects perceptions of a particular story, because the hostile media phenomenon applies to perceptions of the media generally (Eveland and Shah, 2003; Gunther, 1988, 1992). Because our media literacy training focuses on the role of the media more broadly and because perceptions of a particular story are linked to perceptions of the media (Chia et al., 2007; Perloff, 1989; Rouner et al., 1999), we expect that exposure to a media literacy video will increase issue-specific and general media trust (Hypothesis 2).

Methods

To test these hypotheses and research questions, this study utilized an experimental design embedded in a web-based questionnaire. Participants were recruited from groups of individuals who are highly involved in the debate over biofuel implementation.

To assemble a sample of people involved with the issue of biofuels, we recruited members of biofuels-related interest groups and college students from a large Midwestern university enrolled in topically relevant classes. Starting in spring 2009, participants were recruited from several national and state energy special interest groups.² Publics representing a range of interests in the biofuels debate were recruited, such as Midwest Agnet (a Midwestern-based agricultural website) and Biofuels Digest (a biofuels news site).³ In early fall 2009, additional participants were recruited from advanced undergraduate environmental studies courses covering issues of energy, environment, and agriculture, such as ‘The Global Warming Debate’ and ‘Renewable Energy Solutions’.⁴ Of the participants (N=83), 63 percent were male and the mean age was 34 (SD=15.5; Min=18; Max=76); 22 percent were Republican, 18 percent were Independent, 59 percent were Democrat; and 49 percent were students (N=41). Our analysis suggests our recruitment of engaged partisans was effective: a pre-test measure of agreement with the statement ‘biofuels should be used as an alternative energy source’ was folded so that a higher score indicated a stronger position on the issue, producing a mean of 3.13 on a scale from 0 to 5 (SD=1.83).

Subjects were randomly assigned to an experimental group or a control group.⁵ In the experimental group, participants were exposed to a three-minute presentation on media literacy and the news process. The presentation was a video that combined a text-based Powerpoint presentation with key points highlighted, with a female narrator lecturing on this topic (see Appendix A for a transcription of the narrator’s script). Although it is difficult to find a consensus on what constitutes proper media literacy education, this presentation focused on the value of the press to democracy and the construction of news stories, pulling from two key strands of most media literacy education. The control group did not receive the presentation.

Immediately after the control group completed the pre-test and after the experimental group watched the media literacy video, participants were asked to read a brief news article attributed to the Associated Press that covered the advantages and disadvantages of wide-scale biofuel implementation. The Associated Press was chosen as a source because it is a widely used wire service contributing to both liberal and conservative media outlets. The article presented arguments for and against biofuel implementation (see Appendix B). After reading the article, participants were asked a series of questions about their perceptions of bias and credibility of the news story, trust in media related to this specific issue, and general media trust.

Dependent variables

For manipulation checks, we used two single-item indicators: asking respondents to rate how strongly they agreed with the statements: ‘I have a good understanding of the concept of media literacy’ (M=6.15, SD=2.14) and ‘An independent press will present a variety of views’ (M=6.93, SD=2.53).

To examine perceptions of hostility of the particular story, we used a single item asking respondents to rate their agreement on an 11-point scale with the statement: ‘The news story I just read is hostile to my stand on the issue of biofuels’ (M=3.90, SD=2.72).

To test the perceived credibility of the story, we used a standard media credibility battery, which involves rating six concepts – fairness, balance, accuracy, bias, trustworthiness, and completeness – on 11-point scales (Fico et al., 2004; Meyer, 1988). These items were averaged into a credibility index (Cronbach’s alpha=0.87, M=5.38, SD=2.09).

To measure issue-specific media trust, respondents were asked two questions about how fairly the media covers the biofuels debate and if the media present all sides of the issue. Participants rated their agreement with each statement on an 11-point scale, which was averaged to create an index (r=0.64, p<.001, M=3.96, SD=1.97).

To evaluate general media trust, respondents were asked to rate their agreement with three statements regarding the trustworthiness, balance, and completeness of media coverage generally. Participants ranked their agreement with each statement on an 11-point scale (Cronbach’s alpha=0.85, M=4.28, SD=2.00).

Controls

Participants were asked two questions about their political ideology in terms of social issues and economic issues. The questions used a 7-point scale, from ‘Very Liberal’ to ‘Very Conservative’ and were averaged (r=0.54, p<.001, M=3.50, SD=1.40).

To measure issue-specific media use, individuals answered a single item which asked them how much attention they paid to news stories about energy/environmental issues on an 11-point scale (M=8.20, SD=1.98).

To measure issue strength, a single item was used asking respondents to rate on an 11-point scale their agreement with the statement, ‘biofuels should be used as an alternative fuel source’ (M=7.41, SD=2.83). This item was folded, so that a higher number indicated a stronger position on the issue, ranging from 0 to 5 (M=3.18, SD=1.91).

Results

Manipulation checks

Before testing our research questions and hypotheses, we performed manipulation checks to ensure that respondents paid attention to the media literacy video. In this, and in all subsequent analyses, we included two controls: ideology and environmental media use. The first control is meant to ensure that the effects of our manipulation occur independently of ideological leaning, given the differences between liberals and conservatives in their perceptions of media (Eveland and Shah, 2003; Gunther, 1988, 1992; Vraga et al., 2009). Issue-specific media use is also important to control when considering media perceptions. Even in an experimental design, controls ensure that an experimental condition is not confounded with other factors in accounting for criterion variables (Keppel, 1991). These controls are necessary to correct for randomization errors and because they are significantly correlated with two of our dependent variables: story hostility and credibility.⁶

Using a one-way ANCOVA, we found a significant effect of exposure to the media literacy video on respondents’ ratings of their understanding of media literacy, (F(1, 75)=11.489, p<.01, η²=0.139). Those who saw the media literacy presentation (M=7.24, SE=0.44) reported having a greater understanding of media literacy than those in the control condition (M=5.31, SE=0.34). Similarly, we saw a significant main effect on agreement that the press should present a variety of views, (F(1, 75)=4.704, p<.05, η²=0.062), with those who saw the presentation agreeing more strongly (M=7.57, SE=0.48) than those in the control group (M=6.21, SE=0.38).

Research question 1

Before examining the effects of our experimental manipulations on perceptions of the story and the media, we turn to answer research question 1 and explore whether perceptions of bias, story credibility, media credibility and general media trust are indeed empirically distinct constructs.

As Table 1 demonstrates, factor analysis suggests that it is more appropriate to consider these as two factors, rather than four separate variables: one dealing with perceptions of the story (both hostility and credibility) and the second indicating media trust (both issue specific and general).⁷ Therefore, we use the indicators for each of these variables to create two factors: ratings of the story, with a higher score indicating the story was more credible (and less hostile) (Cronbach’s alpha=0.84, M=5.27, SD=1.87) and ratings of the media, with a higher score indicating more trust (Cronbach’s alpha=0.89, M=3.96. SD=1.93).

Table 1.

Factor analysis

	Factor 1	Factor 2
Eigenvalues	2.095	1.280
Percent of variance	52.351%	31.992%
Story hostility	.090	−.916
Story credibility	.098	.859
Issue-specific trust	.941	.021
General media trust	.945	−.023

Hypothesis testing

To test our hypotheses, we performed a series of one-way ANCOVAs. Our first hypothesis predicted that those exposed to media literacy training would rate the story more highly than those who did not see the training. Our data support this hypothesis: a significant main effect of exposure to the media literacy training emerged (F(1, 82)=6.673, p<.05, η²=0.079); those who saw the training rated the story as significantly more credible (M=5.87, SE=0.31) than those in a control condition (M=4.82, SE=0.25).

Our second hypothesis suggested that exposure to the media literacy video would heighten ratings of media trust. Our results supported this hypothesis: we found a significant main effect for exposure to the training (F(1,75)=11.649, p<.01, η²=0.141), with those who saw the training expressing greater trust in the media (M=4.95, SE=0.34) than those in the control condition (M=3.43, SE=0.27).

Additional analyses

While we selected our sample from among an engaged audience, we performed additional analyses to ensure that our effects were not limited to the least-partisan of this group. Therefore, we re-ran our analyses for both hypotheses using only the most partisan respondents – those who had a score of 3 or higher out of 5 on our measure of issue strength, a total of 65 percent of the sample. Among this group, our results are replicated: strong partisans exposed to the media literacy training showed significantly higher ratings of story credibility (F(1, 52)=7.717, p<.01, η²=0.139) and media trust (F(1,46)=14.281, p<.001, η²=0.254).⁸ Thus, we feel confident that our media literacy training was effective for even the most partisan of this highly engaged group.

Research question 2

Our results suggest that the media literacy training functioned as predicted, heightening perceptions of story credibility and increasing ratings of trust in the media more generally. However, we were also interested in disentangling these relationships, especially among story-specific and general media ratings. Therefore, our final research question examines the potential for mediation as a result of exposure to the media literacy training.

We tested these competing models using path analysis performed with the MPlus software package. Because we are uncertain which – if either – model would provide a better explanation for possible mediation between the literacy training, ratings of story credibility, and perceptions of media trust, we begin by examining the model fit statistics for each model. Model 1 tested whether perceptions of the story mediated the relationship between the training and ratings of media trust; Model 2 tested the inverse: whether ratings of media trust mediated the effects of the training on perceptions of the story. Neither model provides an adequate fit for the data (see Table 2): the chi-square statistic is significant, the RMSEA does not reach the critical value of 0.05, and neither the TLI nor the CFI reach the desired level of 0.95.⁹ Therefore, it appears that mediation did not occur between exposure to the media literacy training, ratings of story credibility, and perceptions of general media trust.

Table 2.

Model fit statistics for path analysis

	Model 1	Model 2
Chi-square	3.904*	3.786*
CFI	.784	.802
TLI	.383	.406
AIC	699	698
BIC	713	712
RMSEA	.187	.184

Discussion

The public’s reduced trust in the news they consume is no longer a novel finding, but few solutions have been offered to address the normative problems for democracy that distrust in media can create. However, research into the application of media literacy suggests that giving people a more thorough understanding of the functioning of the news and inviting them to become critical consumers may encourage them to control their biases (Mihailidis, 2008; Tobias, 2008; Vraga et al., 2009). This study examined whether exposure to a media literacy video could affect individuals’ perceptions of hostility and credibility of a particular story and trust in the media overall.

Our study produced support for these hypotheses. Exposure to a media literacy video increased perceptions of story credibility and trust in the media. Furthermore, the shifts in perceptions were sizeable: averaging more than a one-point increase on an 11-point scale.

These shifts in media perceptions occurred despite the fact that the audience was selected for their engagement with the issue of biofuel use. It is these involved partisans who should believe news media are hostile to their point of view (Eveland and Shah, 2003; Gunther, 1992; Vallone et al., 1985). This study suggests that media literacy training may be able to counteract these biases by encouraging individuals to consider the processes by which news is made and the democratic functions it is designed to serve.

Reviewing the rationale by which the media literacy video may work clarifies this argument. Partisans should have the most extreme and broad latitudes of rejection (Eagly and Chaiken, 1993; Sherif and Hovland, 1961), making it difficult for them to accept a message with contrary viewpoints. But if the media literacy training functioned by increasing the latitude of acceptance – or at least non-commitment – for messages including diverse viewpoints, it should be powerful among partisans. However, we remain unsure whether the media literacy training increased acceptance of the incongruent arguments themselves or whether it worked by highlighting the appropriateness of the media to include these arguments. We suggest that the context in which individuals are exposed to media literacy training needs to be carefully considered when gauging response to media content. Further, research into the HME and its functioning would benefit from more careful consideration of the contributions of social judgment theory (Gunther, 1988; Stamm and Dube, 1994).

This study also provides a bridge between HME and research into perceptions of media credibility and trust. Much of the previous research into the HME has focused on evaluations of the story and the hostility of the content (Eveland and Shah, 2003; Gunther and Schmitt, 2004). But we suggest, and our findings support, that perceptions of hostility may be one component of a more general view of media credibility. Our media literacy training worked to enhance ratings of the story both in terms of its lack of hostility and its credibility, as views of story hostility and credibility were indistinguishable among this audience (Fico et al., 2004; Meyer, 1988). Although some previous research into the HME has examined other constructs related to credibility, such as trust and fairness (D’Alessio, 2003; Gunther, 1988, 1992; Perloff, 1989), little research has bridged these two fields. This study is a step in this direction, a direction that holds great promise for a more insightful understanding of how media perceptions fit together, what affects those perceptions, and what can be done to encourage the audience to consider media fairly and critically.

Our study also attempted to disentangle the relationship between ratings of a specific story and views of the media more generally. Our analyses confirmed that exposure to the media literacy training boosted both ratings of the story and trust in the media separately. With this confirmation, we tested competing models to determine whether mediation occurred, examining both the possibility that ratings of the story influenced views of the media (Perloff, 1989; Rouner et al., 1999) or that shifting views of the media led respondents to rate the story more favorably (Chia et al., 2007; Giner-Sorolla and Chaiken, 1994). Neither model provided an adequate fit to the data, suggesting that mediation did not occur. Instead, it appears the media literacy training functioned to boost perceptions of the story and of the media separately.

Naturally, this study has limitations that need to be addressed. The measurement of some of our variables is not ideal. One concern is our use of a single-item measure of hostility. Although our factor analysis in response to research question 1 led us to combine ratings of hostility and credibility into a single measure of story ratings, it is possible that with a more nuanced measurement of hostility, perceptions of story credibility and hostility would be distinct. Future research needs to re-examine the linkages between perceptions of hostility and credibility to clarify this relationship. Further, this study included exposure to a single news story on a single topic and particular care was taken to ensure the story maintained journalistic norms of balance and objectivity. Not all stories adhere to this norm and the ideological cue of a source remains influential. Future research should examine other topics and contexts in which media literacy training could be effective.

Our results for the mediational analysis also remain limited by our sample size. Given the small sample, only relatively simple models were feasible using path analyses. A larger sample or a more complex model may illuminate new pathways between media literacy and ratings of media credibility and trust, a possibility that future research should explore.

In addition, it remains to be seen if media literacy training goes beyond reducing the perceptual biases in the news consumer. The balance we strove to maintain in the story led individuals to view the story as not particularly hostile; even in the control condition, the story only received a rating of 4.40 out of 10 in terms of its hostility. While the media literacy training reduced perceptions of hostility even further, it is possible that the training would not be as effective with a story perceived as more hostile. As previous research suggests, even a story seen as balanced and objective by neutral others can provoke hostile media perceptions among partisans (Perloff, 1989; Vallone et al., 1985); it is possible that this study did not fully stimulate hostile media perceptions. However, respondents with stronger attitudes on the issue (i.e. those who had a score of 3 or higher out of 5 in terms of attitude strength) did rate the story as substantially more hostile (M=4.37, SD=2.95) than those with less strong attitudes (M=3.07, SD=2.05) – a difference that is statistically significant (t=2.097, p<.05). Therefore, while the story was not rated as especially hostile, those with more extreme attitudes on the issue do rate the story as more hostile, as predicted by HME.

Further, we would not want media literacy training to blind people to actual biases that may exist in news coverage. If the training discourages critical consumption of news that truly displays a bias, it would raise concerns about its value. In this study, we do not have the means to test this. Future research needs to evaluate critical thinking after media literacy training and consider multiple news stories, including biased stories, to answer these questions.

That our sample included both college students and community members is an important aspect of this study suggesting that the potential for media literacy to contribute to the societal integration that media trust enables is not limited to college populations. That said, all of our respondents had relatively high levels of education and our media literacy training was tailored for this audience, including sophisticated terminology. We suspect that media literacy training would need to be customized, based on the knowledge of the audience, and encourage future research to build presentations suitable for a variety of audiences, including middle, high school, and college students – those most likely to be exposed to media literacy in a formal setting.

In addition to formal classroom education, in both primary and secondary education and college, media literacy education can be incorporated into more informal settings. To date, most media literacy education has occurred in the classroom or other formal settings, but we believe that content could be developed for both formal and informal training (Hobbs, 2010; Knight Commission, 2009). For example, developing public service announcements (PSAs) promoting media literacy that can run on television and as sidebar content online is one way to target individuals outside the classroom. A recent digital and media literacy report suggests that PSAs and other informal educational approaches should be used to promote media literacy outside of the classroom (Hobbs, 2010). The content from our media literacy presentation could be used as a PSA and placed on relevant websites. In addition, similar content focusing on the same themes could be developed. While these methods may not function as effectively as extensive classroom training, creating media literacy education content for both formal and informal education extends the range of opportunities for promoting media literacy and exposes a wider range of people to the message. This study is a first step in testing media literacy training in an experimental condition. Understanding the impacts of this type of training is important before embarking on large-scale efforts to create media literacy materials and curricula.

A limitation of our study is that we cannot test whether these beneficial effects persist beyond the experimental session. We encourage future research to investigate the potential for literacy education to influence perceptions of news on a variety of issues, as well as the endurance of these perceptions.

We are not arguing that citizens should place blind trust in the news, but simply rejecting coverage that does not agree with their viewpoints is also dubious. Given the importance of a free press to a functioning democracy, it is important to understand the underpinnings of audience perceptions of media content. This study suggests that media literacy training focusing on the benefits of a free press, the need for representation of diverse views, and the role of the critical consumer could be an important way to improve citizens’ understanding of the journalistic process, reducing perceptions of bias and hostility. By encouraging critical and fair consumption of news, we are promoting the functioning of a healthy democracy. Crafting media literacy training that does not fall victim to the audience’s prior beliefs, while maintaining a balance between healthy skepticism and overly critical consumption, may prove difficult; nevertheless, it is a task that seems increasingly necessary.

Footnotes

Appendix A: Media literacy presentation text

Media literacy is an emerging field that has gained momentum in the last decade. One area of media literacy deals specifically with the news process. Today’s short media literacy tutorial explores the steps involved in how a news story begins, to how it is interpreted by you, a news consumer.

It is often said that democracy is about a plurality of ideas and opinions that are in competition with one another. In a democratic society, people are free to express their opinions and ideas, and there is an accepted tolerance for this exchange. The press is one place where a variety of ideas can be expressed. In most news stories, a person can expect to find things that he or she doesn’t agree with or doesn’t like, because it is the role of the press to present multiple viewpoints. Journalists, for the most part, strive to present all sides of a story.

There are many ways journalists go about getting different sides of a story. First, when journalists are given a story, they must go out and find relevant sources that represent various viewpoints on the issue. They then have to balance the responsibilities of their profession and yet do this while working under the constraints of a deadline. Despite having their own perspectives, diligent journalists strive to present an accurate, well-rounded story. In addition, there are some checks and balances in the system, including an editing process with fact checking and review by multiple editors. Also there is a code of ethics that journalists operate under that stresses reliable and credible reporting. However, no matter what journalists do, in the final analysis, your viewpoints and experiences as a news consumer are equally relevant.

There are three key concepts in media literacy education that are important to consider when you read and interpret a news story. The concepts are: (1) all media messages are constructed; (2) different people experience the same message differently; and (3) media professionals, including journalists, conform to a set of industry norms.

When you watch the news or read the newspaper, you bring your own interpretations. These personal preferences influence your understanding of a story. For example, you choose what news stories to read based on your own taste and interests, while avoiding other stories. Two people reading the same story can focus on different aspects and may interpret the story differently. As a reader, you should weigh arguments carefully and understand that, in the press, a variety of views will be presented. Engaging with the various sides of an issue is an important step in forming your own opinion.

So it’s a relationship: on the one hand the journalist striving to provide accurate information and, on the other hand, you the news consumer striving for a fuller understanding of specific issues and concerns.

Freedom of the press and freedom of expression are two of the founding principles of democracy. The public’s right to understand and know what’s happening in their world or with their leaders is essential to a functioning democracy. Ultimately it is the job of the press and the public to work together to foster inclusive debate necessary for democracy.

Appendix B: News article text

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Biographical notes

Emily Vraga is an Assistant Professor in the Department of Communication at George Mason University. Her research focuses on the effects of new media formats on the processing of content and perceptions of media credibility, as well as how personal predispositions, particularly political orientations, influence individuals’ understanding of media content.

Melissa Tully is an Assistant Professor in the School of Journalism and Mass Communication at the University of Iowa. Her current research focuses on the use of new media technologies by civil society organizations and user expectations about new technologies.

Heather Akin is a PhD student in Life Sciences Communication at the University of Wisconsin-Madison. Her research investigates cross-cultural understanding of environmental issues and how media use and political context affect perceptions of global problems.

Hernando Rojas is an Associate Professor in the School of Journalism and Mass Communication at the University of Wisconsin-Madison. He received his doctorate in mass communication in 2005 from the University of Wisconsin, and has an MA in mass communication from the University of Minnesota, and a law degree from Universidad Externado de Colombia. Rojas’ research focuses on: (a) the deployment of new communication technologies for social mobilization in a variety of contexts; (b) the influence of audience perceptions of media (and audience perceptions of media effects) on both public opinion and the structure of the public sphere; and (c) the conditions under which media support democratic governance.

References

Austin

Johnson

(1997) Effects of general and alcohol-specific media literacy training on children’s decision making about alcohol. Journal of Health Communication 2(1): 17–42.

Brown

(1998) Media literacy perspectives. Journal of Communication 48: 44.

Chia

Young

SYW

Wong

ZWD

Koh

(2007) Personal bias or government bias? Testing the hostile media effect in a regulated press system. International Journal of Public Opinion Research 19(3): 313–330.

Choi

Watt

Lynch

(2006) Perceptions of news credibility about the war in Iraq: Why war opponents perceived the internet as the most credible medium. Journal of Computer-Mediated Communication 12: 209–229.

D’Alessio

(2003) An experimental examination of readers’ perceptions of media bias. Journalism and Mass Communication Quarterly 80(2): 282–294.

Eagly

Chaiken

(1993) The Psychology of Attitudes. Fort Worth, TX: Harcourt Brace Jovanovich.

Eveland

Jr. Shah

(2003) The impact of individual and interpersonal factors on perceived news media bias. Political Psychology 24(1): 101–117.

Fico

Richardson

Edwards

(2004) Influence of story structure on perceived story bias and news organization credibility. Mass Communication and Society 7: 301–318.

Giner-Sorolla

Chaiken

(1994) The causes of hostile media judgments. Journal of Experimental Social Psychology 30: 165–180.

10.

Gunther

(1988) Attitude extremity and trust in media. Journalism Quarterly 65: 279–287.

11.

Gunther

(1992) Biased press or biased public? Attitudes toward media coverage of social groups. Public Opinion Quarterly 56: 147–167.

12.

Gunther

Liebhart

(2006) Broad reach or biased source? Decomposing the hostile media effect. Journal of Communication 56: 449–466.

13.

Gunther

Schmitt

(2004) Mapping boundaries of the hostile media effect. Journal of Communication 54: 55–70.

14.

Hobbs

(2010) Digital and Media Literacy: A Plan of Action. Washington, DC: The Aspen Institute.

15.

Hobbs

Frost

(2003) Measuring the acquisition of media-literacy skills. Reading Research Quarterly 38(3): 330–355.

16.

Hobbs

Jensen

(2009) The past, present, and the future of media literacy education. Journal of Media Literacy Education 1: 1–11.

17.

Hwang

Pan

Sun

(2008) Influence of hostile media perception on willingness to engage in discursive activities: An examination of the mediating role of media indignation. Media Psychology 11: 76–97.

18.

Kellner

(2005) Toward critical media literacy: Core concepts, debates, organizations and policy. Discourse: Studies in the Cultural Politics of Education 26(3): 369–386.

19.

Keppel

(1991) Design and Analysis: A Researcher’s Handbook. Englewood Cliffs, NJ: Prentice Hall.

20.

Knight Commission on the Information Needs of Communities in a Democracy (2009) Informing Communities: Sustaining Democracy in the Digital Age. Washington, DC: The Aspen Institute.

21.

Kunda

(1990) The case for motivated reasoning. Psychological Bulletin 108: 480–498.

22.

Livingstone

(2004) Media literacy and the challenge of new information and communication technologies. The Communication Review 7: 3–14.

23.

Lord

Ross

Lepper

(1979) Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence. Journal of Personality and Social Psychology 37: 2098–2109.

24.

Masterman

(1985) Teaching the Media. London: Comedia Publishing Group.

25.

Meyer

(1988) Defining and measuring credibility of newspapers: Developing an index. Journalism Quarterly 65: 567–588.

26.

Mihailidis

(2008a) Beyond Cynicism: How Media Literacy Can Make Students More Engaged Citizens. PhD Dissertation, University of Maryland.

27.

Mihailidis

(2008b) Are we speaking the same language? Assessing the state of media literacy in US higher education. Studies in Media and Information Literacy Education 8(4): 1–14.

28.

Perloff

(1989) Ego involvement and the third-person effect of televised news coverage. Communication Research 16: 236–262.

29.

Pew (2009) Press Accuracy Rating Hits Two Decade Low Public Evaluations of the News Media: 1985–2009. Washington, DC: Pew Research Center for the People and the Press.

30.

Rouner

Slater

Buddenbaum

(1999) How perceptions of news bias in news sources relate to beliefs about media bias. Newspaper Research Journal 20(2): 41–51.

31.

Schmitt

Gunther

Liebhart

(2004) Why partisans see mass media as biased. Communication Research 31: 632–639.

32.

Sherif

Hovland

(1961) Social Judgment; Assimilation and Contrast Effects in Communication and Attitude Change. New Haven, CT: Yale University Press.

33.

Stamm

Dube

(1994) The relationship of attitudinal components to trust in media. Communication Research 21: 105–123.

34.

Taber

Lodge

(2006) Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science 50: 755–769.

35.

Tobias

(2008) Culturally relevant media studies: A review of approaches and pedagogies. Simile 8(4): 1–17.

36.

Tsfati

Cohen

(2005) Democratic consequences of hostile media perceptions: The case of Gaza settlers. Harvard International Journal of Press/Politics 10: 28–51.

37.

Vallone

Ross

Lepper

(1985) The hostile media phenomena: Biased perception and perceptions of media bias in coverage of the Beirut massacre. Journal of Personality and Social Psychology 49: 577–585.

38.

Vraga

Tully

Rojas

(2009) Media literacy training reduces perceptions of bias. Newspaper Research Journal 30: 68–81.