Content-Type: text/html This paper was presented at the Association for Education in Journalism and Mass Communication in San Antonio, Texas August 2005. If you have questions about this paper, please contact the author directly. If you have questions about the archives, email rakyat [ at ] eparker.org. For an explanation of the subject line, send email to [log in to unmask] with just the four words, "get help info aejmc," in the body (drop the ""). (Feb 2006) Thank you. Elliott Parker ==================================================================== Political Correlates of Daytime Talk Show Viewing Carroll J. Glynn The Ohio State University School of Communication Bruce W. Hardy Cornell University Department of Communication James Shanahan Cornell University Department of Communication Running Head: Daytime Talk Shows Corresponding author Bruce W. Hardy Department of Communication Cornell University 338 Kennedy Hall Ithaca, NY 14850 [log in to unmask] Submission for presentation at the 2005 AEJMC Conference Political Correlates of Daytime Talk Show Viewing Abstract This study examined the influence of daytime talk shows on opinion formation, from a cultivation perspective. Specifically, we examined how exposure to daytime talk shows and the extent that these shows are perceived as real are related to support for government involvement in family issues. Not only did we find that both exposure and perceptions were positively related to levels of support, we found a mainstreaming effect toward a liberal position. That is, conservatives mainstreamed towards liberals the more they watched daytime talk shows and, also, the more they perceived these shows to be real. Overall, this study demonstrated that daytime talk shows play a significant role in public opinion formation. Political Correlates of Daytime Talk Show Viewing One of the characteristic developments of American television in the 1990s and beyond is the rise of the daytime talk show. Within this genre are many different types of shows, ranging from the community-oriented therapy of The Oprah Winfrey Show to the shocking sensationalism of Jerry Springer. However, regardless of the format, political and social issues have been, and continue to be, central topics of discussion and debate featured on these shows. Not only do daytime talk shows hosts occasionally take explicit political positions on current issues, these shows consistently feature social issues such as sexuality, family conflict, drug addiction, abuse, and criminal activity. This study examines the relationship between exposure to these daytime talk shows and attitudes concerning political and social issues. Specifically, this study explores the cultivation phenomenon by examining the links between daytime talk show viewing and support for government involvement in family issues – issues which are frequently emphasized on these types of shows. Theoretical background: Cultivation The central hypothesis of cultivation states that television viewing gradually leads to the adoption of beliefs about society that are consistent with those systematically portrayed on television (Gerbner, et al., 1980; Shanahan & Morgan, 1999). That is, perceptions of the "real" world will be "cultivated" by exposure to the same symbolic messages over and over again. As Gerbner et al., (1982) noted, "Television provides perhaps for the first time since pre-industrial religion, a strong cultural link, a shared daily ritual of highly compelling and informative content, between the elites and all other publics." Gerbner and his colleagues then asked, "What is the role of this common experience in the general socialization and political orientation of Americans?" (p.101). Twenty-some years later, this question has yet to be fully answered. Partly due to the fact that television is constantly morphing, trying to study the social effects of this medium is comparable to shooting at a moving target. However, we are able to partly address this question by examining specific types of television content and their relationships with specific political opinions. In this study we examine daytime talk shows and support for government involvement on family issues. The daytime talk show format The format for daytime talk shows is usually informally-guided conversation among the host, guests and the audience. Beginning in 1970, Phil Donahue pioneered this format in which he first interviewed a guest or guests. He would then guide the audience by posing questions to and eliciting reactions from the interviewees. The success of his program led to the rise of other competitors, including a variety of talk show hosts such as Ricki Lake, Montel Williams, Sally Jessy Raphael, Geraldo Rivera, and Oprah Winfrey. The popularity of this type of format is due to the unique opportunity, in a mass media setting, to mix conversation with a more institutionally directed form of discourse such as the more formal news interview (Ilie, 2001). These talk shows provide a middle ground between private, free-flowing individual conversation and more rigidly structured forms of institutional discourse. Because talk shows allow for "participation," they may be an under-recognized arena for the expression of public opinion about an enormous variety of issues. Although radio talk shows have received some attention concerning their influence on opinion formation, expression and participation (see, Hofstetter, 1998; Pan & Kosicki, 1997), daytime television has received less attention. Currently, it is an open question as to how or whether daytime TV talk shows provide a forum that would encourage attention to political and social issues. Content of daytime talk shows In an analysis of 200 episodes of talk shows, Greenberg and Smith (1995) found that themes of marriage, family, children, relationships and sex were predominant. Unfortunately, to our knowledge, apart from Greenberg and Smith's analysis, there are no other systematically collected data available. Therefore, in order to add strength to our theoretical argument, we examined show topics for The Oprah Winfrey Show for the year 2002. Show titles are available on the program's website for several years. From the show's title we ascertained the basic theme of the program. The most prevalent theme in the program was "family/children." This was followed by celebrity/personality interviews, and relationships/sex.[1] Additionally, these themes can be seen while perusing transcripts of the shows. Oftentimes the host will conclude the show with a message intended to drive home a lesson or conclusion center on these themes, as in the following examples: TOLERANCE (Montel Williams on gay parents): At least we found out something. It's not about political correctness. It's just about respect for another human being. If you don't agree, OK. It's not a lifestyle you have to live. And it also is not something that really affects you. So don't run around giving your children an attitude so it affects them that they gotta go and voice their opinion at somebody else. We could stop this garbage now. Just as Ramona said, tolerance is learned and taught. So is ignorance. (March 18, 2003) CHILDREN COME FIRST (Jerry Springer on dealing with unexpected pregnancies): You know, nobody's suggesting that pregnancy means life for the woman must stop, nor is this show about whether stripping or making porn movies are an appropriate way to make a living. No, this is about the undeniable responsibility one has in bringing life into the world and whether decency, if not good sense, requires a modification of lifestyle so as to ensure that the newborn has the best chance of making it. Stories about how rough your childhood was are no excuse for passing that abuse and irresponsibility on to your child. To be using drugs and alcohol or be whoring around while pregnant suggests that one is nowhere near ready to be a mother, and perhaps it's in the best interest of the newborn to be removed from the home at birth, even if the state has to do it. That may shake up the moms-to-be, but nothing else seems to be working. It's a lot easier to become a mother than to actually be one. (October 24, 1996) SELF-EXPRESSION (Oprah Winfrey): The realization that you are a person who's been given life and a voice and that that voice means something--that voice means something not only to you, not only to your family, but to the rest of the world. You were meant and born to speak your voice. (March 18, 2003) RESPONSIBILITY (Montel Williams on paternal responsibility for children): Well, whether he wants to or not, I'd take him to court and make him do what's right. Even if he's not seeing the child--your--your child--not for you, but your child deserves to have child support to help take care of their needs in case your child becomes a gifted student and someday wants to go to a school or go somewhere. That person should be paying to make sure that child has the right to do that. OK? And so make sure you do that. Daytime talk shows rarely take an explicitly political focus. But this does not mean that the shows lack political content. Occasionally, hosts will take explicit political positions on their shows. For example, in 1999 O'Donnell publicly attacked New York City Mayor Rudolph Giuliani on his new homeless policy (Nordlinger, 2000). O'Donnell, on her show, said "He's out of control, this guy! Sure, just, you know, arrest all the homeless people" (Nordlinger, 2000, p. 2). O'Donnell has been involved in other political debates concerning social issues, including gun control, child welfare, and her own homosexuality. Many of these issues were frequently discussed on her show (Nordlinger, 2000). Davis and Owen (1998) stated that "political content is intermittent on these programs" (p. 148). They highlight that the viewers of daytime talk shows tend to be younger and are more likely to be female. They also are more likely to have less education, to have lower income, and not to be white. These characteristics are concurrent with what we would expect from people more likely to be disaffected from the political system and less likely to pay attention to other more politically oriented media. Daytime talk shows and support for government involvement in family issues Unfortunately, the few existing studies on the influence of daytime talk shows have not explicitly looked at their influence on support for government intervention concerning social well-being. Most studies have looked at how perceptions of social relationships are cultivated by the portrayal of atypical behavior often showcased on these shows. For example, Davis and Mares (1998) found that talk show viewers overestimated the frequency of some social problems. As Woo and Dominick (2003) noted, "Some observers, such as Sen. Joseph Liebermann, have suggested that these programs may have negative effects on viewers because their focus on bizarre behaviors and dysfunctional relationships might prompt audience members to think such behaviors are typical…" (p. 109). However, it has been suggested that this effect is more pronounced for those who are unfamiliar with societal norms and values. For example, Woo and Dominick (2003) found that international students who scored low on acculturation – familiarity with host country's norms and values – and watched daytime talk shows were more likely to have negative perceptions and attitude toward human relationships in the United States. Relying on past cultivation research (for an overview see Shanahan & Morgan, 1999), we can assume that heavy viewers of daytime talk shows depend on these shows to learn about social issues. Not only do these viewers extract information on social issues from these shows, but their beliefs should reflect those consistently portrayed. In other words, the more viewers spend time "living" in the televised world of daytime talk shows they will be more likely to see the "real" world in terms of the values and ideologies portrayed on these shows. We, therefore, hypothesize that daytime talk shows encourage a liberal-oriented philosophy of support for women and family on social issues. The shows emphasize repeatedly that there are problems in child-care, keeping families together, and personal health and fulfillment. The presentation of these problems rarely involves specific demands for government action, but we can hypothesize that regular viewers of daytime talk shows would tend to adopt a social and political agenda in favor of greater government assistance with family issues. We put forth the following hypothesis: H1: Viewership of daytime talk shows will be positively related to support for greater government involvement in family issues. As previously outlined, daytime talk shows have frequently been criticized for giving an unfair picture of social relationships, presenting and emphasizing sensational and dysfunctional aspects of society in order to improve ratings. Some programs, such as Jerry Springer, have even been accused of scripting interactions that are presented as real. Therefore, we are interested in the extent to which people actually believe what they see is real and what effect does this perception of realism have on support for government involvement on social issues such as family and child well-being. Davis and Mares (1998) found that perceived realism of talk shows did not make much difference for viewers' judgments about the world. But Greenberg and Smith argued that "most research models would posit that these perceptions mediate or intervene in the viewer's acceptance of what they see and hear on such shows" (p. 91). Potter (1986; 1988) has argued that perceived realism makes a difference in how viewers' conceptions of reality are constructed. Conversely, Shanahan and Morgan (1999) argued that such perceptions play a less important role in viewers' tendency to absorb depictions of reality from television. Shapiro and Lang (1991) hypothesized that the effects of television on perceptions of reality is a result of people simply forgetting that what they view is not real. Simply put, viewers get so involved in the story of a television show that they forget that they are watching fiction. Moreover, Shrum (1995, 1996) has suggested that heavy viewers use images on television as heuristic cues for mental judgments about social reality. He has empirically supported this hypothesis and found that heavy viewers have more readily accessible opinions to issues in directions consistent with what cultivation predicts. Perceived realism could be more important with talk shows than other television shows because these shows are presented explicitly as non-fiction. That is, daytime talk shows explicitly put forth that the guest are telling true-life stories and that the interactions among the host, guests, and audience are unscripted. If viewers believe that these interactions displayed in daytime talk shows are centered on true-life stories then repeat viewers should perceive these issues as very salient problems facing American society. For example, if a viewer repeatedly views pregnant teenage girls on daytime talk shows, and perceives these teenage girls to be truthful and their situations realistic, then this viewer should believe that teenage pregnancy is a major social problem facing our country. Therefore, perceptions that daytime talk shows are real should influence support for government involvement in family issues. We put forth the following hypothesis: H2: Perceptions of talk shows as realistic will be positively related to support for greater government involvement in family issues. Mainstreaming and daytime talk shows Mainstreaming can be defined as a convergence of opinion of individuals of different socio-political subgroups toward the "mainstream." This "mainstream" can be conceptualized as a "relative commonality of outlooks and values that exposure to features and dynamics of the television world tends to cultivate." (Gerbner, et al., 1982, p.104). Therefore, mainstreaming results in an accordance of opinions and world views shared by heavy views of different socio-political backgrounds, whereas light viewers of these different sub-groups would hold divergent views. Mainstreaming can be viewed as an "interaction in which cultivation is stronger for some subgroups, weaker or absent for other groups, and in which heavy viewers' responses [i.e., opinions] are closer than those of light viewers" (Shanahan & Morgan, 1999, p. 141, original emphasis). In other words, the effect of some socio-demographic variable is moderated by exposure to television. Differences in views, or opinions, of different demographic subgroups that are otherwise associated with socio-political or cultural characteristics diminish, or may even become absent, when examining heavy television viewers. Additionally, political science scholars, such as John Zaller (1992), have discussed the idea of an elite opinion consensus on a particular policy across the political spectrum producing a mainstreaming effect. Zaller (1992) asked, "What would be the expected effect on public opinion if virtually all persuasive messages carried in political media on a particular policy were favorable of that policy, and if there were no cueing messages to alert people that the policy was inconsistent with their values?" (p. 98). The answer, of course, is mainstreaming. Although a small conceptual leap is required, one can view the daytime talk shows as an "elite opinion" source providing consistent messages that family issues are salient a problem that needs government intervention. Therefore, heavy viewers of daytime talk shows are more likely to have similar opinions on government involvement on family issues than non-heavy viewers. For this study we expect political ideology, i.e. conservative versus liberal, to influence support for government intervention in family issues. This effect will be moderated by exposure to daytime talk shows. H3: The effect of political ideology, i.e. conservative versus liberal, on support for government intervention on family issues will be moderated by exposure to daytime talk shows. Additionally, we expect to see a mainstream effect among those who perceive the daytime talk show to be realistic. In other words, perceptions of daytime talk shows as realistic should play a moderating role on the influence of political ideology on support of government intervention in family issues. Intuitively, it seems that those who believe that the content of these shows are realistic are more likely to have their perception of reality cultivated by such shows. We put forth the following hypothesis: H4: The effect of political ideology, i.e. conservative versus liberal, on support for government intervention on family issues will be moderated by perception that daytime talk shows are realistic. A meta-analysis of cultivation studies conducted by Shanahan and Morgan (1999) suggested that most mainstreaming consists of liberal views moving toward conservative views. Analyses by Gerbner et al., (1982) showed that, although, television viewing brings left- and right-leanings closer together, "it is the liberal position that is weakest among heavy views" (p. 122). However, it should be noted that, historically, family and child welfare issues are not polarizing and government spending on such issues is usually generally supported regardless of political ideology (Entman & Paletz, 1980). Numerous surveys have shown that respondents are likely to support government spending on services that benefit them while taking a more negative position on other social issues such as taxes, political equality, and crime (Gerbner, et al., 1982). Therefore, given that our first hypothesis predicts that exposure to daytime talk shows will be positively related to support of government intervention in family issues, we expect that the mainstream effect will consist of conservative views moving toward liberal views. Although, this does not seem to be unreasonable, empirical evidence for mainstreaming in a liberal direction is weak at best. However, Gerbner et al., (1982) demonstrated that heavy viewing conservatives and moderates converged toward a liberal position concerning federal funding on health, education, and welfare programs. Given that we are examining support for government involvement in family issues we predict: H5: Heavy viewers of daytime talk shows will mainstream toward a liberal position. Methods Procedures The data for this study came from a national US telephone survey conducted between May 5th and May 30th, 2000. Professional telephone interviewers surveyed 596 randomly selected adults throughout the 48 contiguous states and Washington, D.C. (the response rate was 60%, based on AAPOR definitions) Multiple callbacks were conducted in order to reduce response bias. The confidence level used for analysis was 95% and the standard error was +/- 4%. Measures Demographic variables. In order to isolate the independent contribution of daytime talk show viewing, we controlled for several socio-demographic variable that may possibly precede and influence the relationships among our variables of interest. Specifically, we controlled for respondents' gender, (62% female); age (M = 45 yrs., SD =16.9); education (M = 13.4 years of education completed, SD = 2.0); household income (median = $50,000); and race (21.3% non-white). Political ideology. Respondents' political ideology (M=4.2, SD=1.4) was tapped by a 7-point scale where 1 meant "extremely liberal" and 7 meant "extremely conservative" and the midpoint, 4, meant "moderate." Talk show exposure. To tap respondents' exposure to daytime talk shows we asked people how often they watched the following programs: Jenny Jones, Jerry Springer, Leeza, Montel Williaims, Oprah Winfrey, Ricki Lake, Rosie O'Donnell, and Sally Jessy Raphael. Responses were rated on a four-point scale from 1 ("never") to 4 ("frequently"). An exploratory factor analysis suggested that responses to the specific program items clustered into two factors. The first was comprised mainly of exposure to the more "controversial" shows such as Jenny Jones and Jerry Springer. The second factor comprised of exposure to shows focused more on presenting the host as a friendly therapeutic companion, such as Oprah and Rosie O'Donnell. However, a confirmatory factor analysis (see Loehlin, 1992) of these two factors failed to produce an adequate model fit. Moreover, the two latent variables, or factors, were correlated producing a f coefficient of .51. Therefore, we decided against operationalizing talk show exposure as a two factor construct and create a single factor additive index of exposure consisting of the 8 items (a =.82). Perception that daytime talk shows are real. To tap respondents' perception that daytime talk shows are real we first asked "How accurately do you think talk shows represent issues that are important?" Then we asked "What about most other people, how do you think most other people think talk shows represent issues?" Response options ranged from 1 ("not at all accurately") to 4 ("very accurately"). We also asked respondents to agree or disagree with two statements about talk shows. The first was "Most people who go on talk shows are usually being honest." The second was "Watching talk shows is a useful way to find out about values and norms in our society." Response options ranged from 1 ("strongly disagree") to 5 ("strongly agree"). We then combined these items into a single scale (a = .61). Criterion measure. For our dependent variable, we asked respondents to express their support concerning five different social issues. These were: government provided daycare for children, government required parental leave for employees, government provided health care, expanding Medicare coverage to include prescriptions and spending more money on education. For each issue, respondents could indicate support on a five-point scale ranging from 1 ("strongly oppose") to 5 ("strongly support"). All the issues, on balance, receive support from our sample. Especially strong support is shown for parental leave, Medicare prescription coverage, and spending more money on education. All issues receive support significantly above the midpoint value of 3. We created a summative index from these issue variables, measuring an overall tendency of the respondent to judge that government should be involved in social issues (a = .73). Results In this study we examined the influence of daytime talk shows on political opinion formation. Specifically, we hypothesized that consumption of this type of media will be positively related to support for government involvement in family issues and that a mainstreaming effect will result from heavy viewing of talk shows. That is, regardless of political ideology, heavy viewers of daytime talk shows will have similar levels of support for government intervention, or, in other words, daytime talk show viewing will moderate the influence of political ideology. To test these direct and interactive effects it is necessary to construct two regression models. The first model includes socio-demographic variables, political ideology, exposure to daytime talk shows, and the interaction term (political ideology × exposure to daytime talk shows). The second model includes the same socio-demographic variables and political ideology; however, instead of the exposure measure we have included a perception measure and the corresponding interaction term (political ideology × perception that talk shows are real). We ran each of these models separately because a single model with both interaction terms would end up over-controlling for political ideology as the variable would be included, directly and in the interaction terms, three times. In constructing the interaction terms, the main effect variables were standardized by translating them into z-scores before the interaction terms were formed in order to avoid multicollinearity problems among the interaction terms and their components (Cohen, Cohen, West, & Aiken, 2003). [Insert Table 1 about here] Table 1 details the results of the two regression models. Age, education, and income were all negatively related to support for government intervention on family issues. Non-whites were more supportive than whites, and females were more supportive of government intervention than males. Political ideology produced the largest beta coefficient in both models (ß = -.29, p = .01) meaning that liberals were more likely to support government intervention on family issues than conservations, as would be expected. Supporting our first hypothesis, exposure to talk shows was significantly and positively related to our criterion measure (ß = .15, p = .01). Supporting hypothesis 2, perception of talk shows as reality was also positively and significantly related to the criterion measure ((ß = .10, p = .05). Finally, supporting hypothesis 3 and hypothesis 4, both the interaction terms were positively and significantly related to support for government intervention on family issues. Specifically, the interaction between exposure to daytime talk shows and political ideology produce a beta coefficient of .09, p = .05, while the interaction between perceptions of daytime talk shows produced a beta coefficient of .12, p = .01. Supporting hypothesis 5, we found that conservatives do, indeed, mainstream toward liberal views. This relationship held for exposure to daytime talk shows as well as perceptions that daytime talk shows are real. That is, conservatives mainstreamed toward liberals the more they watched daytime talk shows and, also, the more they perceived these shows to be real. For illustrative purposes we graphed the means of support for the subgroups. Figure 1 illustrates difference in the means for 1) low exposure/conservative, 2) high exposure/conservative, 3) low exposure/liberal, and 4) high exposure/liberal. Figure 2 illustrates the difference in means for 1) low perception/conservative, 2) high perception/ conservative, 3) low perception/ liberal, and 4) high perception/liberal. [Insert Figure 1 and Figure 2 about here] Discussion The findings from this study add to our limited knowledge on the relationship between exposure to daytime talk shows and attitude concerning political and social issues. Before we discuss the implications of our results it is first necessary to briefly discuss some potential limitations to our study. Like many studies using cross-sectional surveys, the data are dependent on self-report. Therefore, there is a potential for social desirability bias concerning our criterion measure, i.e., respondents reporting more support for government involvement in social issues because they believe that it is the socially acceptable answer. Another limitation to cross sectional data concerns causality, i.e., does media use precede opinion formation or does certain opinions drive media use? Although, there is no statistical test for causality we have based our interpretation of the analyses on a strong theoretical argument building on past cultivation research. Therefore, we are confident in modeling these relationships as unidirectional. Given these limitations, this study makes a number of significant contributions to our understanding of the role of daytime talk shows in public opinion formation. First, we found a direct relationship between exposure to daytime talk shows and support for government intervention in family issues, above and beyond socio-political characteristics. Our analyses suggest that daytime talk shows do seem to offer possibilities for issue discussion that is relevant to viewers' political thoughts and views on family issues. Secondly, we found a direct relationship between the extent that daytime talk shows are perceived as real and our criterion measure. In other words, the more real viewers believe these shows to be the more influence these shows have on their political views. This supports Potter's (1986; 1988) hypothesis that perceived realism is an important factor in how viewers' conceptions of reality are cultivated. More interesting, however, is the mainstreaming effect produced by viewing daytime talk shows. Our analyses suggest that exposure to these shows and that the perception that these shows are real pushes viewers to adopt liberal positions above and beyond their self-identified political leanings. This suggests that these shows highlight a liberal "mainstream" and that those viewers outside this "mainstream" will move toward such position as exposure increases. However, we caution against generalizing this result to all social issues. Recall that we only analyzed support for government involvement in family issues and that support for federal funding of programs concerning such issues is consistently positive regardless of political leanings. Future research should examine the mainstream effect of daytime talk shows on other social issues. It would be interesting to see the relationship between daytime talk shows and attitudes toward other issues such as the death penalty, political tolerance, and abortion. Overall, this study has demonstrated that daytime talk shows play a significant role in public opinion. Recently, communication researchers have begun to examine the civic consequences of entertainment media that also provides public affairs information – often referred to as "info-tainment" (see, Kwak, Wang, & Guggenheim, 2004, Moy, Xenos, & Hess, 2004). Realizing that many citizens are extracting information on political and social issues from non-traditional news outlets, i.e. info-tainment television, communication scholars have just begun a research agenda to explore the role that this type of television show has in informed public opinion and active democratic citizenship. We, therefore, suggest that future research continues to examine the impact that daytime talk shows may have on American democratic society. For instance, future research should examine what effects these types of shows have on willingness to participate in public deliberation. The deliberative format of these shows may have a significant impact on viewers' willingness to express conflicting opinions in a hostile environment or a willingness to listen to opposing views. A recent study by Hardy, Scheufele, and Wang (2005) found that different types of media influence these dimensions of deliberation differently. These researchers found that television news views was directly related to expressing conflicting views while newspaper news use was directly related to listening to conflicting viewpoints. These authors suggest that this is a result of the presentation format of information, i.e., newspapers tend to give both sides of a story while television focuses on only one side. It would be interesting to see if the deliberative aspects of daytime talk shows influence respondents' willingness to participate in a deliberative forum. As the popularity of the daytime talk show continues, their impact on public opinion will, as well, continue. We have found that daytime talk shows are directly related to support for government involvement in family issues. Unfortunately, this is a very small slice of public affairs. Therefore, we are unable to predict the overall significance of exposure to daytime talk shows and future research is required to fully understand how these shows fit into the American political landscape. References: Cohen, J., Cohen. P., West. S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavior sciences (3rd ed.). Hillsdale, NJ: Lawrence Erlbaum. Davis, R. & Owen, D. (1998). New media and American politics. New York: Oxford University Press. Entman, R. E., & Paletz, D. L. (1980). Media and the conservative myth. Journal of Communication, 30, 154-165. Gerbenr, G. & Gross, L. (1976). Living with television: The violence profile. Journal of Communication, 26, 173-199. Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1982). Charting the mainstream: Televisions contribution to political orientation. Journal of Communication, 32, 100-127. Greenberg, B., & Smith, S. (1995, November). The content of television talk shows. Report prepared for the Kaiser Family Foundation. Hardy, B. W., Scheufele, D. A., & Wang, Z. (2005, May). Defining deliberation: Key determinants and distinct dimensions. Paper to be presented to the 2005 annual conference of the International Communication Association. New York, NY. Hofstetter, C. R. (1998). Political talk radio, situational involvement, and political mobilization. Social Science Quarterly, 79(2): 273-286. Ilie, C. (2001). Semi-institutional discourse: The case of talk shows. Journal of Pragmatics, 33(2): 209-254. Kwak, N., Wang, X., & Guggenheim, L. (2004, August). Linking all the way: The relationship between television entertainment talk show viewing and political engagement among young adults. Paper presented to the annual meeting for the Education in Journalism and Mass Communication. Toronto. Loehlin, J. C. (1992). Latent variable models: An introduction to factor, path, and structural analysis (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum. Morgan, M., & Signorielli, N. (1990). Cultivation analysis: New directions in media effects research. Newbury Park: Sage. Moy, P., Xenos, M. A., & Hess, V. K. (2004, May). Priming effects of late-night comedy. Paper presented to the annual conference of the International Communication Association. New Orleans, LA. Nordlinger, J. (2000, June 19). Rosie O'Donnell, political activist – a celebrity and her platform. National Review. Retrieved March 11, 2005 Online: http://www.findarticles.com/p/articles/mi_m1282/is_11_52/ai_62838140. Pan, Z. D., & Kosicki, G. (1997). Talk show exposure as an opinion activity. Political Communication, 14(3): 371-388. Potter, W. J., (1986). Perceived Reality and the Cultivation Hypothesis. Journal of Broadcasting & Electronic Media, 30(2): 159-174. Potter, W. J. (1988). Perceived Reality in Television Effects Research. Journal of Broadcasting & Electronic Media, 32(1): 23-41. Shanahan, J., & Morgan, M. (1999). Television and its viewers: Cultivation theory and research. Cambridge: Cambridge University Press Shapiro, M. A., & Lang, A. (1991). Making television reality: Unconscious process in the construction of social reality. Communication Research, 18, 685-705. Shrum, L. J. (1995). Accessing the social influence of television: A social cognition perspective on cultivation effects. Communication Research, 22, 402-429. Shrum, L. J. (1996). Psychological processes underlying cultivation effects: Further tests of construct availability. Human Communication Research, 22, 482-509. Woo, H., & Dominick, J. R. (2003). Acculturation, cultivation, and daytime TV talk shows. Journalism and Mass Communication Quarterly, 80, 109-127. Zaller, J. (1992). The nature and origin of mass opinion. Cambridge: Cambridge University Press. Table 1: Dependent Variable: Support for Government Intervention on Family Issues Beta Beta Socio-demographics Age -.06 -.09* Race -.15** -.16** Education -.10* -.11* Gender -.11* -.13** Income -.18** -.17** Political Ideology Ideology -.29** -.29** Daytime Talk Shows Exposure .15** - - Perceptions - - .10* Interactions Ideology × Exposure .09* - - Ideology × Perceptions - - .12** R2 (%) 25.6 24.8 Note: ** p= .01 * p= .05 Race coded 1 for white and 0 for non-white. Gender coded 1 for males and 0 for female. Figure 1: Interaction between political ideology and exposure to daytime talk shows Figure 2: Interaction between political ideology and perception that talk shows are real [1] Reliability for this analysis was measured with Scott's pi, which was equal to .73.