Whether the news media have a liberal bias has interested politicians, journalists, scholars, and the public. Many seem to believe that a political bias exists. According to a recent survey by the Pew Research Center for the People and the Press (2002), 47% of those who answered a question on media bias believed news organizations in general are politically biased in their reporting. In comparison, 35% of respondents disagreed.
Conservative critics believe that most journalists are liberal and Democrats and that news coverage reflects reporters' political leanings (Corry, 1996; Goldberg, 2001 ; R. Lichter, Rothman, & Lichter, 1986; Limbaugh, 1993; Maddoux, 1990; Maitre, 1994; Olasky, 1988; Rusher, 1988; Sowell, 1992). According to these observers, the news media and reporters are pro abortion, racial quotas, and gay rights, and they are anti business, capitalism, the military, Christianity, and the Republican party.
A different group of critics argues the opposite. In their eyes, conservative voices dominate the mainstream media, and news organizations--most of which they see as controlled by the government and large corporations are "agents of power" that promote and maintain the conservative status quo (Alterman, 2003; Altschull, 1995, 1996; Bagdikian, 1997; Cohen, 1990; Cohen & Solomon, 1993; Croteau & Hoynes, 1994; Croteau, Hoynes, & Carragee, 1996; Gitlin, 1980; Hanson, 1992; Herman & Chomsky, 1988; M. A. Lee & Solomon, 1990; McChesney, 1997; Murdock, 1982; Parenti, 1988, 1993, 1995, 1996).
To investigate whether the media have an ideological (liberal-conservative or left-right) or partisan (Democratic Republican) bias, media scholars have tried several approaches. First, surveys show that journalists tend to vote for Democrats and to take liberal stands on political issues (Dennis, 1996, 1997; T. Lee, 2001; R. Lichter et al., 1986; S. R. Lichter & Rothman, 1981; Weaver & Wilhoit, 1996). Probably due to such factors as journalistic objectivity and management pressure, however, a link between reporters' political beliefs and news coverage has never been convincingly established (Black, Steele, & Barney, 1999; Dennis, 1996, 1997; Dreier, 1983; Epstein, 1973; Fishman, 1980; Gans, 1979, 1985, 2003; Goodwin & Smith, 1994; Knowlton & Parsons, 1994; T. Lee, 2001; Merrill, 1997; "Public Television Study Disproves Liberal Bias Theory," 1993; Schudson, 1978, 1997; Tuchman, 1978). Second, researchers have examined news content and found no significant or consistent partisan or issue favoritism (Dalton, Beck, & Huckfeldt, 1998; Dennis, 1996, 1997; Domke, Watts, Shah, & Fan, 1999; Fedler, Meeske, & Hall, 1979; Fico, Ku, & Soffin, 1994; Graber, 1971 ; Hofstetter, 1976, 1978; Merrill, 1965; Niven, 2002; Patterson, 1994; Severin & Tankard, 1992; Stempel, 1961, 1969; Stempel & Windhauser, 1984).
If social-scientific evidence does not support the claim that the U.S. media have a liberal bias, why are such accusations made? Media critic Michael Parenti (1996) offers several explanations for conservatives' consistent accusations. First, most U.S. media are owned and controlled by large corporations, and consequently conservative voices are dominant and can repeat their complaints with greater frequency than liberal critics. Second, conservative politicians and commentators habitually attack the media to put them on the defensive. As a result, liberal opinions often are self-censored by journalists because "anything short of unanimous support for a rightist agenda is treated as evidence of liberal bias" (Parenti, 1996, p. 103). Third, social realities reported in the news, such as wrongdoings in the government and large corporations, or poverty and pollution, appear liberal or even radical to conservatives.
Parenti's (1996) reasons are echoed by other media observers (e.g., Alterman, 2003) and complemented by the findings of media researchers. These scholars argue that conservative and Republican elites' strategic and frequent complaints have convinced both the media and the public about the existence of such a bias, although news content does not support this claim (Domke et al., 1999; Watts, Domke, Shall, & Fan, 1999). These researchers suggest that journalists, political elites, and the public all tend to believe the media have a liberal and pro-Democratic bias.
The present study was not designed to analyze whether there is an ideological or partisan bias in terms of journalists' stance or news content. Instead, the perception of bias among media audiences was the focus of this investigation. In an interesting twist to other research on media biases, "hostile media" studies suggest that supporters of an issue or a group tend to believe the media favor their opponents (Gunther, 1992; Gunther & Chia, 2001; Gunther, Christen, Liebhart, & Chia, 2001; Pedoff, 1989; Vallone, Ross, & Lepper, 1985). The present research uses hostile media tenets to test the assumption that observers' characteristics, including ideological and partisan positions, contribute to their perception that the news media do not report news fairly and therefore cannot be trusted. Specifically, the study analyzed two large national surveys to examine the connection between partisan and ideological positions and media distrust. The fundamental research question is whether strong liberals and conservatives, and strong Republicans and Democrats, are more likely to distrust the media than moderates and independents. Such differences are expected to emerge because an observer's standpoint could determine what one sees. The literature backing these expectations is reviewed in the next section of this article.
Bias Perception and Hostile Media Research
The American Heritage Dictionary defines bias as "a preference or an inclination" that "inhibits impartial judgment," or "an act or policy stemming from prejudice" (Pickett et al., 2002, p. 138). In the context of news reporting, a bias is the opposite of accuracy, balance, and fairness (Fico et al., 1994; Fico & Soffin, 1995, Lacy, Fico, & Simon, 1991; Simon, Fico, & Lacy, 1989; Streckfuss, 1990). According to these scholars, accuracy means not going beyond the facts of the matter, and balance plays out through giving roughly equal amount of coverage to all involved parties. Fairness is achieved when all involved parties' perspectives are represented, and no more favorable treatment of any side--either qualitatively or quantitatively--exists. Therefore, fairness reveals itself as a form of balance. For the purpose of this study, a bias in the news media is defined as any form of preferential and unbalanced treatment, or favoritism, toward a political or social issue (e.g., pro-choice or pro-life) or political party (Democratic or Republican). Bias, theoretically, can be prevented by remaining impartial and unprejudiced, which is the norm in the journalism profession in the United States (Black et al., 1999; Goodwin & Smith, 1994; Knowlton & Parsons, 1994; Merrill, 1997; Mindich, 1998).
A number of scholars have taken an innovative approach--hostile media perception research--to investigate why audiences perceive media bias. These researchers have discovered that supporters of political groups or issues perceive the media as being unfair or even hostile to their own cause while favoring their opponents (Beck, 1991; Dalton et al., 1998; Gunther, 1992; Gunther & Chia, 2001; Mason & Nass, 1989; Perloff, 1989). Most studies in this line of research have one thing in common: two groups surrounding a single issue--such as Middle East conflicts, primate research, or a UPS strike--tend to believe the media unfairly favor the other side (Christen, Kannaovakun, & Gunther, 1998; Giner-Sorolla & Chaiken, 1994; Gunther et al., 2001; Hastorf & Cantril, 1954; Perloff, 1989; Vallone et al., 1985). Not surprisingly, participants in these studies think the media unfairly favor their opponents. A related study testing the third-person effect also reported that respondents believed that biased media coverage may sway neutral observers toward the other side (Perloff, 1989). Another characteristic of hostile media research, as pointed out by Gunther (1988, 1992), is that the level of observers' self-involvement with an issue or group is likely to determine their views on whether media coverage is credible or biased. In other words, attitude extremity affects trust in media.
The horizon was expanded again in a study by Gunther (1992). His large-scale national survey determined that members of various groups (e.g., Republicans, Democrats, Catholics, born-again Christians) thought that media coverage of their own groups was significantly more negative. A 1998 study conducted by Dalton et al. was another milestone. These researchers first analyzed media content, then polled public opinion on media coverage to compare whether there was a connection between news content and perceived media bias. Their content analysis reported no partisan bias. A large portion of the respondents did not perceive the newspapers or television news programs they saw as biased toward a political candidate. Certain citizens' own partisan views, however, colored their perceptions of media fairness. Even an objectively fair and balanced news story could look unfair and unbalanced if an observer has a bias. For example, people with strong attachments to the GOP were more likely than Democrats to see their daily newspapers as leaning toward Clinton in the 1992 presidential election. Strong supporters of the Democrats were more likely than Republicans to believe that newspapers favored Bush. Their perception of media bias had little to do with political news coverage (Dalton et al., 1998).
Drawing on previous research, especially Gunther's (1988) study in which political ideologies were an independent variable, this study applies the approach of hostile media perception in a different way. Previous research investigated whether supporters or members of a...