..\.f. 5.31.23. +. f: .2! Venn-5...: :5 ma... ...d if 5",. n; ..w. I‘.lb'v Li. 1:? 7.). fiftn.» II. 0‘ 41.... .cal .. :1: ‘13. :37 FM“? WWW '. ll mm. “ I 711;“:- ..O|..I kc ‘ y f ‘tlll ‘ z o .0 I 0“ 1 . 1?.ch- ‘I'Aqnlo. . .I .T.I .011! {I .. 7!; : “.ulnlv.{-«v.. 1..- ... IA k» ...... 1 :I I'O'H ."’§"~‘ :"'I a‘ u t ‘ 0"“.- . ‘ll‘...\‘£l& [I‘vglt .v ufll .t. I...- .. ucfiv.\)\ V . vfy0o$t Zv ‘gt .noaly. til-f, I .l‘vlolv if} ’ltll’ , ml ‘1 n It}; .tl . 1.17. I . u‘ I. 7 0|. l.u..nl‘ A ¥n\u|‘..2|’p‘l n .V‘l.i I v: ‘Quirt..r\~ioar. .‘vfi’tsr: ‘5 .‘It‘vll -7 all . ‘7‘. . iii-2;: I. {I L L pint»... ...n. o} V" v‘ 3:? ..DII a. ,. . n: -V“ .‘ .9?!- 4ll‘.l.n \Ahk» . , .v V! ‘25.. -.. ‘ . .V‘ ‘0! q| L0... 1016 ‘. A .. ... 4|. ¢ V O: .04! ...1{..n!.-. f3] .21. aka... . -) M12 _ «mm ‘1 3“: 5: Int. I .i “sat-‘3 llllllllll\‘IIIIHH‘UHIIll!IHIIHHIHH|l|l||H|l|llllll 3 1293 00881 0024 This is to certify that the thesis entitled Asian International Students Learning 'I'nrough Television: A Cultivation Study presented by Jeanette Marie James has been accepted towards fulfillment of the requirements for Mastenmdegree in Cormnication Major professor Date 1-22-93 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Mtchtgen State Untverctty PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE “"t f‘ £13} ' 6M1} Lg" ‘ ......., if A 4.9.. "r; ‘1'“ ' «gee (“A __________L________ [NW § (3 Mt “ ‘ L__________J ”—fl—T rfl] \____\ MSU Is An Affirmative ActiorVEqual Opportunity Institution cmWWS-o.‘ __ .7 _____u_ ._ 17.- ASIAN INTERNATIONAL STUDENTS LEARNING THROUGH TELEVISION: A CULTIVATION STUDY By Jeanette Marie James A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Communication 1993 ABSTRACT ASIAN INTERNATIONAL STUDENTS LEARNING THROUGH TELEVISION: A CULTIVATION STUDY BY Jeanette Marie James Cultivation analysis research is one way to determine how television contributes to the construction of social reality. Cultivation analyses have been performed on a full range of topics on Americans. However, there lacks a preponderance of studies on populations that are not American. This study is designed to examine the learning of stereotypes about Americans through television among a population of Asian international students. To measure this learning, the respondents were questioned on how they characterized Americans and the United States before they arrived in this country. This is contrasted with their current characterizations of Americans and the United States. It is predicted that television exposure will lead to some level of learning about Americans. In addition, it is predicted that television will lead to stereotyping of Americans and the United States. ACKNOWLEDGMENTS I would like to thank several people for their support and encouragement. Without their help this project could not have been successfully completed. First, I would like to express my deepest appreciation to my committee members. You were all patient and encouraging; it meant a lot to me. Each of you played an important and distinct role in this study's development, and I would like to note that. Dr. Atkin, you helped me in transforming my ideas for this study into working constructs. I could not have started (or finished) this project without you guidance and support. Dr. Wilson, you were particularly helpful with the perfection of the hypotheses. This really gave my project the solidity that it needed. Dr. Straubhaar, thank you for the valuable feedback that you offered throughout the entire project. Your comments and advice on the literature review were especially helpful. Mom and Dad, thanks for your love, prayers, and support, I could not have done this without you! Finally, thanks to all of the friends who helped me in so many ways: from covering my shifts at work to cheering me up when I needed it. I'll always remember each of you. TABLE OF CONTENTS LIST OF TABLES LIST OF APPENDICES INTRODUCTION Definition of key terms CHAPTER I. PROBLEM DEFINED Rationale Theoretical Concept of Cultivation Cultivation and Reality Perception Cultivation and sex-role stereotyping International Cultivation Analysis The concept of Acculturation Research Hypotheses Method Construction of hypothhses The Population The Instrument Limits of this study Summary II. LITERATURE REVIEW Research Literature Specific to the Topic Research in Cognate Areas Relevant to the Topic Studies related to stereotypes Critique of validity in research literature Summary of what is known and unknown about the topic Contribution this study will make to the literature iv PAGE vi viii CHAPTER PAGE III. DESCRIPTIVE ANALYSIS 37 Demographic characteristics 37 Acculturation Variables 39 Mass Media use among Asian 41 international students IV. ANALYTICAL ANALYSIS 44 Hypothesis testing 44 Context for Hypothesis One 44 Hypothesis one tested 45 Context for Hypothesis Two 50 Hypothesis two tested 51 Selection of acculturation measure 55 Context for Hypothesis Four 60 Hypothesis 5 64 Summary 69 V. DISCUSSION 72 Problems 72 Contribution of this study to the 76 cultivation process Conclusions 77 APPENDICES 78 REFERENCES 95 TABLE OtUIII-‘DNN 10 11 12 13 14 15 LIST OF TABLES Age distribution Home country distribution Education levels Americans known Language Competence Desire to fit in to U.S. society correlations Television exposure (entertainment) pre-U o S 0 During U.S. exposure Accuracy of TV depiction's Before characterizations of Americans Before correlations of characterizations with shows Before shows with characterizations summed Current characterizations of Americans Current correlations of characterizations with shows Current shows with characterizations summed vi PAGE 42 43 45 47 51 53 54 TABLE 16 17 18 19 20 21 22 23 24 Correlations for acculturation measure High vs Low Assimilation for Beverly Hills 90210 High vs Low Assimilation for The Cosby Show Means for prevalent problems in society Correlations TV themes & prevalent problems Coefficients of determination (r2) Correlation TV themes a prevalent problems (light) Correlations TV themes & prevalent problems (heavy) T—Test for difference between heavy & light viewers vii PAGE 57 58 59 61 62 63 65 67 69 APPENDIX A Frequencies of viewing: Pre-U.S. shows Frequencies of viewing: Current shows Survey LIST OF APPENDICES viii PAGE 78 85 89 INTRODUCTION Cultivation studies have been performed for decades and applied to many situations. When George Gerbner began the Cultural Indicators project in 1967 the focus was solely on violence in network television. The project began under the commission of the United States Government at the recommendation of the Surgeon General. At this time there was a fear that there was too much violence in network television. Gerbner and associates began investigating the idea that television may shape societies' perceptions of the real world. The researchers explored this notion through two hypotheses. First, they hypothesized that heavy viewers of television content would be more likely than light viewers to give responses based on television and not real world facts. Second, they hypothesized that heavy television viewers would have a greater fear of victimization than would light viewers. Hirsch (1980), when reinterpreting Gerbner's data, found no validity for the cultivation hypothesis. In one instance both non-viewers and medium viewers gave more TV answers than light viewers, this clearly does not follow a normal pattern of prediction. Hirsch charged that often the Gerbner team 1 2 was often able to boost the results of their data by mixing the responses of non-viewers with heavy viewers. Hirsch further noted that the results of the Gerbner's data were often slight. When conventional methods of data analysis were employed, the correlation's rarely approached .10 and never exceeded fifteen percent (Hirsch 1980). Nevertheless, cultivation studies have continued to be performed and the topics for analysis have expanded. Several researchers have modified the approaches of cultivation studies in order to examine attitudes and stereotypes that may be learned through television. Potter (1986), in a study of reality perception, began to modify the cultivation process. Potter had three basic rationales. First, there may be a belief on the part of the viewer that television programming is accurate. Second, the viewer must perceive a connection between what is seen on television and direct experiences. Finally, the viewer may believe that television situations are similar to life situations. This study is guided more by Potter's modified cultivation study than a classic Gerbner cultivation study in that it explores the relationship between television viewing and possible reality construction. Typical cultivation studies are more stimulus-response oriented and focus on behaviors that television may induce (i.e., heavy television viewers will have a stronger fear of victimization). This approach assumes that there is a strong link between 3 television exposure and social reality construction, a link that has yet to be firmly validated. Definition Of Key Terms 1. Cultivation: Dominant modes of cultural production tend to generate messages and representations that nourish and sustain the dominant ideologies, world views, perspectives, and practices of the institutions and cultural contexts from which they arise (Gerbner, Gross, Morgan, & Signorielli, 1986). 2. Social Reality: The degree that one's definition of an event is influenced by indirect information sources and not by direct sensory exposure (Elliott & Slater, 1980). 3. Mainstreaming: Represents a relative homogenization, an absorption of divergent views, and a convergence of disparate viewers (Morgan & Signorielli, 1990). 4. Resonance: Real-life experience may confirm or disconfirm the TV messages, thus amplifying or diminishing the television's cultivation potential (Tamborini & Choi 1990). 5. Acculturation: Process of adaptive transformation through re socialization (Gudykunst, 1984). 6. Magic Window dimension: The degree to which a viewer believes television content is an unaltered, accurate representation of actual life (Potter, 1986). 7. Identity dimension: Degree of similarity the viewer perceives between television characters and situations and the people and situations experienced in real life (Potter, 1986). I. PROBLEM DEFINED Rationale Varis (1984) in a follow-up study on the exportation of media programming from developed countries noted that: "There is a oneaway traffic from big exporting countries to the rest of the world, and entertainment material dominates the flow" (p. 143). In Asia and the Pacific region, Varis documented the programming as primarily children's and entertainment content. The United States is the chief exporter of entertainment programming to much of the world and many foreigners derive their impressions of the U.S. through television. The danger in this is that the portrayals may often be too "rosy", stereotypical or biased for a foreigner to obtain an accurate picture of the "real” United States. Korzenny and Schiff (1992) comment that examining effects in an international setting can be tedious and present more challenges. "The assumption that media play a role in modifying human behavior has intuitive and empirical roots. But how mediated messages have an impact across cultures is largely undetermined" (p. 1). This study will empirically explore this impact. Specifically, the purpose of this study is to measure the relationship between Asian international students pre-United States media viewing, examining the learning about Americans and living in the U.S. that occurred during this time, and contrast this against the learning that occurs while these students are living in the United States. Cultivation studies (domestic and international), construction of social reality studies, and the ideas of stereotyping and acculturation guides this research. The Concept of Cultivation The goal of Cultivation Analysis is to determine the extent to which people who watch greater amounts of television hold different conceptions of social reality from those who watch less television, all other factors being equal (Morgan & Signorielli, 1990). Cultivation Analysis is a piece of a three part component in the Gerbner Cultural Indicator process. The aim of this process is to investigate "(1) The institutional processes underlying the production of the media content,(2) images in media content, and (3) relationships between exposure to television's messages and audience beliefs about behaviors" (Morgan & Signorielli, 1990, p. 15). Most cultivation studies focus on the medium of television because of its repetitive nature. In addition, television is more pervasive in our society than are other media forms. Typical cultivation studies begin by "assessing the most recurrent and stable media patterns in television content, emphasizing the consistent images, portrayals, and values that cut across most program genres" (Morgan & Signorielli, 1990, p. 16). After the program content has been analyzed then certain predictions can be made about those who watch more television. People who regularly watch a great deal of television differ from light viewers in many ways. Although all social groups include both heavy and light viewers (relative to the group as a whole), there are overall differences between heavy and light viewers according to sex, income, occupation, race, time use, social isolation/integration, and a host of other demographic and social variables. But there are also differences in terms of the extent to which television dominates their sources of consciousness. Cultivation theory assumes that light viewers tend to be exposed to more varied sources while heavy viewers, by definition, tend to rely more on television. (Morgan & Signorielli, 1990 p. 17) It is through the process of both television content analyses and through predictions about receivers of the television messages that cultivation analysis is able to examine how social realities may be constructed through television. One pattern that has been identified by Morgan and Signorielli (1990) as important in cultivation studies is the concept of mainstreaming. "Mainstreaming means that television viewing may absorb or override differences in perspectives and behavior that stem from other social, cultural, and demographic influences. It represents a homogenization of divergent views and a convergence of disparate viewers" (p. 31). Mainstreaming patterns typically identify a convergence of common views in demographic groups. The general pattern reveals that heavy viewers of television in a demographic group will tend to have common views, while the light television viewers in that same demographic group will have divergent views. Simply put "television cultivates common perspectives" (Morgan & Signorielli, 1990, p. 23). Through the past twenty years of research some of the questions that cultivation analysis has examined are: How does cultivation occur? What demographic subgroups are more likely to show evidence of cultivation? How is cultivation mediated by interpersonal and family relations? What is the role of personal experience in cultivation? How do viewers' orientations toward television influence cultivation? What are the roles of specific programs and genres in cultivation? How and what do other media cultivate? HOW‘Will new technologies influence cultivation? Does cultivation occur in other countries? (Morgan & Signoreilli, 1990) Cultivation and Reality Perception Potter (1986), in a survey of perceived reality research methods, concluded that the studies are generally structured in one of four ways. First, researchers examined the differences by having some viewers watch realistic content and others watch fantasy content to determine who would be more affected. Only in studies where subjects were prompted about the realism of what they were seeing did perceived reality become statistically significant (Potter, 1986). In either case, "perceived reality is more likely associated with individual differences rather than the message itself" (p. 160). 8 The second type of perceived reality research is based on the individual difference perspective and attempts to link perception differences to demographic variables. This type of research clearly demonstrates that perceived reality is directly related to race and age as well as to gender in some cases. The results have typically shown that: "Minorities have higher perceived reality of television content than do whites" (Potter, 1986 p. 161). In addition, "younger children have higher perceived reality than do older children, and older adults have higher perceived reality than do younger adults" (p. 161). In the third type of research, perceived reality is examined as an intervening variable between stimulus material on TV and effects on viewer attitudes and behaviors (Potter, 1986). Researchers have only been able to find a direct effect between the content and receiver's behavior, however. This is problematic because if "perceived reality is indeed an intervening variable in this process, then it should exhibit an interactive, not a direct effect" (p. 161). The fourth method states that perceived reality is a multidimensional perspective while in previous methods it was viewed one—dimensionally. Potters' study is guided by the notion that perceived reality is multidimensional. Potter tested three perceived reality variables in the hope of discovering the impact that television exposure may have on the perception of our society as violent. 9 The first concept is the magic window dimension. "This is the degree to which a viewer believes television is an unaltered, accurate representation of actual life" (Potter, 1986, p. 162). The second, the instruction dimension, "encompasses viewers beliefs about TV as an instructional aid which augments and expands their direct experiences" (p.162). The third is the identity hypothesis. This is the degree of similarity a viewer will feel between characters and situations experienced on television and people and situations in real life. These concepts, while theoretically independent, should correlate in the sense that "People high on the Instructional dimension will probably be high on Magic Window and Identity dimensions" (Potter, 1986, p. 163). Potter summarized that the direct relationship between TV exposure and estimates of real world violence was established in only a few estimates and these relationships were non-existent when demographic controls were introduced. The results, while failing to statistically support the cultivation hypothesis, did yield an enlightening notion. Potter felt that: "The amount of exposure to television seems far less important than the attitudes and perceptions of the individuals being exposed" (Potter, 1986, p. 172). Hawkins and Pingree (1990) examined various psychological processes that may help explain the cultivation process. They divided the processes into three groups: long 10 term inferences, short term inferences, and intermediate processes. Long term inferences usually take place under either mainstreaming or resonance effects. Mainstreaming is a process in which light viewers from two or more demographic groups hold different social reality beliefs while heavy viewers in the subgroup hold more similar beliefs (Hawkins & Pingree, 1990). Mainstreaming leads to greater homogeneity between subgroups who would otherwise show great disparities in their held beliefs because of social, political or cultural differences. In other words, the process brings about a convergence of viewpoint among those who watch more television. Hawkins and Pingree note that this is logical if you assume that certain demographic groups real world experiences ‘will match television portrayals. The author's reason that certain demographic groups, for example, the less educated, would have a smaller pool of real world experiences to draw on and, no matter what their viewing patterns were, television content would more likely match their own experiences. However the authors argue that this explanation, while plausible, makes an assumption that can not necessarily be validated: The real world experiences of the two groups are assumed to be different , and the television world is implicitly assumed to match the real world experiences of the less educated. Beyond that, there is a ‘presumption that social reality beliefs are somehow 11 constructed based on one's total experience pool with little or no attention to the television vs. real-world source of experiences. Thus while conditional effects based on demographic may appear striking they actually provide reasons for speculation, not explanations, about psychological processes.(Hawkins & Pingree, 1990, p.39) The short inference process attempts to directly measure perceived reality as a cognitive trait. Hawkins in a 1987 study identified demographic subgroups, based on school achievement scores and knowledge of current events, that differed in the size of various cultivation relationships (Hawkins & Pingree, 1990, p. 40). Based on this he reasoned that there are some cultivation relationships that lack a direct measurable referent in television content. Instead they are linked to implied content. Hawkins admits that "These studies cannot claim to measure the action of their posited cognitive processes, but what they do measure is at least itself plausibly a result of the posited process as well" (Hawkins & Pingree, 1990, p. 40). The assumption is that the process is much closer and more direct than first order processes. In the short term inference process it is believed that "cognitive orientations are stable over time so that those who say they perceive television to be real will also see it as real when watching or constructing social reality beliefs" (Hawkins & Pingree, 1990, p. 41). Yet, there.are still problems with short term inferences. The main one is that perceived reality is not the exact process of constructing social reality. 12 Intermediate processes are measured by a technique of memory recall. The research of Shapiro, Hintzman, and Pingree led to the following hypothesis: There is a two step process of social reality construction based on television, one step being incidental learning and the other step being the construction of beliefs, which may come from that store of memory (Hawkins & Pingree, 1990). Shapiro reasons that "If cultivation occurs by abstraction from a memory store, it should be possible to assess the relevant stored memories and their individual sources directly, and find them.more closely associated with social reality beliefs than is any measure of television viewing" (Hawkins & Pingree, 1990, p. 41). This approach to testing the cultivation hypothesis is problematic, however, because while it may explain the origins of social reality beliefs, it provides no clear relationship between viewing and beliefs. Without this relationship being established, there can be no clear support for the cultivation hypothesis. Cultivation and sex-role stereotyping Although original work in the area of cultivation began by studying violence in the media, the field has widened to encompass the study of many other television genres. One of these genres is sex-role stereotyping. One study that examined sex-role stereotypes among American adolescents (Morgan, 1982) suggests that television leads to a mainstreaming effect; i.e., television viewing 13 promotes sexism among those who should be least likely to hold traditional sex-role stereotypes. Morgan's data suggest that television viewing statistically correlates with higher levels of sexism for those who are most out of the mainstream. In other words, the heavy television viewers tended to exhibit more sexist attitudes as compared to the light viewers in the same demographic group. The result is a convergence and homogeneity among heavy viewers. It was demonstrated that, particularly among females with high IQ's, there is a tendency towards sexist responses. This is one group that, according to Morgan, should be the least sexist. For women, television viewing predicted scores on an index of sex-role stereotypes 1 year later. Yet, for boys there is no significant link between TV viewing and sexist tendencies (Morgan, 1982). He concluded that: Television seems to have a cumulative effect on sex role stereotypes among girls. The stable assumptions and images about the norms of the adult world, as presented on television, may provide hidden yet vivid and powerful constraints on their notion about the places women should take in the world. (p. 953) Why these images have a stronger effect on women than men is speculative. Perhaps it is because television is viewed by some researchers to be targeted against women. Gerbner and Tuchman in independent studies concluded, through content analysis of various media genres, that there is a "symbolic annihilation" of women in the media through three methods (Morgan, 1982). 14 First, there is an under-representation of women on television. Men outnumber women 3-1 in the television world. Second, there is clear over-victimization of female characters. TV females are more likely than TV males to suffer violence. In addition, a female's risk of victimization decreases if she is married. This subtly implies that the female character is safer if she stays at home. Finally, Gerbner found that there is a trivializing of female characters and that women are portrayed as being unable to mix their careers and home lives (Morgan, 1982). While Morgan offers no implications for Gerbner's observations, it would seem that this is problematic because these images tend to reinforce negative and archaic stereotypes about women particularly unless more counter stereotypical roles are provided to offset this balance. International Cultivation Analysis In the past, cultivation analysis was applied to those living in the United States. While some research has been applied in the international arena, there is still much research to perform in this setting. Indeed, conducting cultivation studies with an international population presents different challenges. There are many scholars who argue about the structure of cultivation theory, and in particular its application in international settings. Tamborini and Choi (1990) discuss several issues surrounding cultural diversity in cultivation research. Two of them are: cultural diversity and the underlying processes ‘1 15 of cultivation and cultural diversity and the television picture. A central issue in cultivation research is that heavy television viewers use the medium nonselectively. Tamborini and Choi explain that in an international context, cultural diversity will play a role in viewer selectivity: The importance of accounting for cultural diversity's influence on selectivity becomes apparent when attempting to study cultivation in cross-cultural populations. For example, the high acculturation motivation among foreign immigrants or sojourners that increase their willingness to learn about host culture, society, and language may lead to heavier consumption of host media. (Tamborini & Choi, 1990, p. 173). Tamborini and Choi argue that "For those foreign sojurners with limited real-life experiences and interpersonal contact, the television world may be considered a magic WindOW‘With high information utility" (p. 173). Another premise in the cultivation hypothesis is the assumption that the images and messages seen on television are relatively coherent. Tamborini and Choi (1990) reason that in instances of foreigners watching U.S. television programming, there may be an incoherent set of images because American television programming might contrast with their native cultures standards and beliefs as well as their own national television programming. Theoretical concept of Acculturation Acculturation is the "process of adaptive transformation through resocialization" (Gudykunst, 1984, p. 209). In the past most acculturation studies have focused on permanent 16 immigrants to the United States, as did Y.Y. Kim's 1977 study. Kim identified several acculturation variables that had strong correlation's with the immigrant's length of stay in the United States. Although the study focused primarily on interpersonal communication channels to determine the immigrants' processes of acculturation into American society, two of the concepts (language competency and interaction potential) are incorporated in this study. Kim.(1977) demonstrated through the testing of three variables, English competency, acculturation motivation and interaction potential, that as the length of time increases for immigrants the become significantly more involved in the host society. English competency was measured through self-report for two reasons: the first was for the actual assessment of competency, and the second was to determine the level of confidence each respondent felt when speaking English. Acculturation motivation was measured in a threefold manner. First, Kim examined the interest that the immigrants had about U.S. culture. Second, Kim examined the levels of intimacy each respondent felt between themselves and Americans. The intimacy measure was subdivided by the amount of casual acquaintances, casual friends, and intimate friends. It was demonstrated that the longer the respondents lived in the United States the more likely they were to be involved in organizations and to have a wider circle of casual friends. 17 Although several of Kim's acculturation questions are incorporated into this study, they were modified slightly because of the population differences. Kim.was able to sample a Korean population that lived in the United States for an average of nine years. For an international student sample this cannot be assumed. Second, Kim's sample consisted of immigrants who had the intention to remain in the United States. For international students, it cannot be assumed that they will remain in the United States beyond degree completion. Research Hyppthesis Hypothesis #1: For Asian international students, before they arrived in the United States, the television programs they watched in their native country will correlate with their past characterizations of Americans. Hypothesis #2: For Asian international students, after they arrived in the United States, the television programs that they currently watch will correlate with their current characterizations of Americans. Hypothesis #3: For Asian international students, there will be a significant difference between (a) how the low acculturation subgroup's characterizations of Americans correlate with the current shows that they watch and (b) how the high acculturation subgroup's characterizations of Americans correlate with the current shows that they watch. Hypothesis #4: For Asian international students, the themes that they see in American television will correlate 18 with their endorsement of prevalent problems in the United States . Hypothesis # 5: For Asian international students, there will be a significant difference between (a) how light television viewers' themes that they see in American television correlate with their endorsement of prevalent problems in the United States and (b) how heavy television viewers' themes that they see in American television correlate with their endorsement of prevalent problems in the United States. Method Congtruction of hvootheppp Hypothesis one and two were constructed based on the idea of Potter's identity dimension. This theory states that there is a degree of similarity between what the viewer perceives between television characters and situations and real life characters and situations. When operationalized for this study, the theory states that watching television may lead to characterizations or stereotyping of Americans. To test hypothesis one, pearson correlations were run against specific shows, that the respondents watched in their home country, with corresponding adjectives to determine if there were significant relationships between the two. Hypothesis two was tested the same way except that the shows are currently available for viewing in the United States. Both hypotheses were tested at 19 two levels: each show to each characteristic and total shows to total characteristics. Hypothesis three examines the relationship between two subgroups: high and low acculturation respondents. The prediction is that there will be a difference in the way these two subgroups characterize Americans based on the shows that are watched. The sample is split at the median between those that are highly acculturated and those who are less acculturated. ’Correlations are computed separately for each group between shows that are currently viewed and characterizations of Americans. A T-Test is performed to determine if there are significant differences between the two groups. Hypothesis four is framed around Potter's identity dimension. To test hypothesis four, pearson correlations were run against each current show and each corresponding theme depicted in that show. In order to test for significance a coefficient of determination is obtained (r2). Hypothesis five is a contrast between those that watch less television (light viewers) and those that watch more television (heavy viewers). The hypothesis states that heavy viewers will be more likely to endorse prevalent problems in American society than light viewers will. The sample is split between light and heavy viewers and correlations run between themes that are seen on television and perceived prevalent problems in society. A T-test is performed to 20 ascertain if there are differences between the light viewers and the heavy viewers. The Population This study was designed to test learning through television among a population of Asian-international students at Michigan State University. The target population was 100 randomly selected students, but due to several data collection errors, the sample size was reduced to 87. The specific nationalities that were tested were Koreans, Chinese, Japanese, and Taiwanese. Inadvertently one Indonesian student was sampled. All of the surveys were conducted on the campus of Michigan State University. The Instrument One area of this survey was borrowed and expanded upon from a previous cultivation study. This section examined the student's exposure to American television programming before they moved to the United States. The show list and corresponding adjective list were expanded upon from a 1985 study performed by Tan, Li, and Simpson entitled American TV and Social Stereotypes of Americans in Taiwan and Mexico. This was done for practical reasons. Tan, Li, and Simpson were able to obtain their show list by conducting their study in the respective countries; for the purposes of this study to do such would have been difficult. Limits Of This Study Typically cultivation studies have failed to confirm the hypothesis' tested. The original cultivation studies focused 21 on exposure to violent TV programming cultivating mistrust and fear in the viewer. These studies typically assume that the process of social reality construction will take place and indeed use it as a synonym for simple learning. Yet social reality is not a simple learning process. As previously discussed, Potter (1986) describes it as a multidimensional process. First, the belief that television portrayals are accurate occurs, then exposure and direct experiences are linked, and finally the viewer may believe television situations are real to life. This study is guided by the notion that reality construction is multidimensional in nature and examines the distinct process of reality construction more closely. While this narrows the scope of the study, it is important to understand the initial processes of reality construction. Clearly, most previous research has failed to establish a clear link between television exposure and social reality construction. Yet these studies are confounded by how television exposure will affect behavior. It is for this reason that the following study, guided by the hypothesis listed above, will not attempt to establish a link between exposure and behavior. Instead it will focus on the learning through television and possible reality construction. Summagy Cultivation theory, using a variety of methods, attempts to measure the degree to which television exposure will affect the social reality of viewers. One belief is that 22 heavy viewers will be more affected than light viewers, holding demographic variables constant. Early studies focused primarily on exposure to violent television programming leading to a "mean world" conception of society among heavy viewers. Typically, the results of these studies showed weak results and in many cases non- existent results, depending on who performed the interpretations. Supporters of the theory continued to argue that since television content exerts a continuous force on the minds of its viewers it would influence the way they saw the world. The problem with this assumption is that these researchers did not thoroughly examine the process of perceived reality formation; failing to establish this link, indeed, further weakens the entire basis of the cultivation hypothesis. The principal variable of this study, like Potter's 1986 study, is perceived reality, or the degree of reality that people see in mediated messages. In this study two of Potter's dimensions are examined: the magic window dimension and the identity dimension. II. LITERATURE REVIEW Research Literature Sppcific to the Topic Choi and Tamborini (1988) tested the cultivation hypothesis in two American communities: Lansing, Michigan and Baltimore, Maryland. The subjects were Korean immigrants and the purpose was to determine if there would be differences among the two samples with regards to the mean world hypothesis. The authors hypothesized that heavy viewers in the urban area of Baltimore would tend to have a stronger fear of victimization than light viewers in the area. Conversely, this sense of trepidation should be lower for those living in the suburban area of Lansing. This specifically measures the resonance effect. The results from this study indicate little support for the view that heavy television viewing leads to fear of victimization. There was even less support for the resonance effect which holds that those surrounded by crime and those viewing heavy amounts of violent programming will have a stronger fear of victimization. Pingree and Hawkins (1981) tested the notion that "media effects are greater without competition from other sources 23 24 and that U.S. television shows are identifiably American" (p. 98). This idea suggests that "the influence of U.S. television content will be greater for non-U.S. viewers beliefs about the U.S. than for their beliefs about their own culture" (p. 98). Their hypothesis was not only rejected, but the researchers found the opposite trend. Television programming clearly had an effect on the populations construction of social reality and it shown that the conceptions extend beyond the culture that creates the programming. The Pingree and Hawkins study took place with school age children in Perth, Australia with the goal of obtaining the students' views about the "meanness" of the world. The subjects were asked questions about perceived violence in both the U.S. and Australia. It was clear from the data obtained that children who watched heavy amounts of television had a view of the world that was based on television. The U.S. was seen as a slightly more violent place to live than was Australia although the difference was not statistically significant. Interestingly enough, the subjects seemed to form their biases about violence in Australian society from heavy viewing of U.S. television programming, while watching non- U.S. programs did not correlate with TV bias responses. With regards to the mean world hypothesis, however, this could be definitively attributed to the fact that most imported U.S. TV programming and little non-U.S. programming 25 is in the genre of crime and adventure stories (Pingree & Hawkins, 1981, p. 101). Research in Cognate AreasyRelevant to the Topic Kang and Morgan's 1988 study examined the impact of television via the source of AFKN television. AFKN broadcasts exclusively American TV programming 75% of which is entertainment. AFKNs' primary audience is U.S. service persons but it is carried throughout the nation due to cable and satellite transmissions. The study was designed to determine if AFKN television viewing led to cultural conflict with the Korean population. The belief was that if AFKN viewing cultivates social realities then those who watch it should be less likely to support Korean positions (Kang & Morgan, 1988). Kang and Morgan derived from their results that female heavy viewers were less likely to endorse traditional views and in general AFKN viewing was strongly related to their perspectives. However, the authors concluded that, females, even those who viewed AFKN in heavy amounts, still had higher moral standards than males. This was shown in such responses as ”unrestricted dating is unethical" and "dating should be discussed with parents" (p. 435). The results from male subjects were often contradictory. For instance, heavy viewing males were more likely to feel that the "Korean family system is better" and simultaneously 26 were less obedient towards their parents. One possible explanation for the incongruity is that "While AFKN may cultivate some acceptance of non-traditional values for males, it may also sharpen their opposition to what they perceive to be inadequacies in the American family system.and their allegiance to the Korean system" (Kang & Morgan, 1988, p. 436)." Morgan and Shanahan (1992) compared the differences in television learning in a Latin American country (Argentina) and an Asian country (Taiwan). Argentina is a country with television programming similar to that found in the United States. Their programming is roughly 80 percent entertainment, about 25 percent of which is imported. By contrast, Taiwan's programming consists of only 25 percent entertainment programming with approximately 35 percent being imported. In this study three dependent variables were measured: gender-role stereotypes, interpersonal mistrust (the mean—world hypothesis) and political attitudes. In Argentina, there was a clear relationship between .television viewing and gender-role stereotyping. The authors found that the heavy television viewers in the Argentina sample were more likely to provide sexist responses. This was more the case for males than for females but was common in both heavy viewer groups regardless of sex. The trend toward strong sex role stereotyping, however, diminishes slightly among the subjects who watched more U.S. television programming. This may be explained by the fact 27 that U.S. programming offers more counter-stereotypical roles for women than does Argentinean programming. With regards to the mean world hypothesis, there was no significant relationship between heavy TV viewing and perceptions of society being violent. This is even true among subjects who watched more imported American television programming. The third dependent variable, authoritarianism, was clearly cultivated. There was also a clear mainstreaming effect in this area: those who should have fallen into the less authoritarian category (females, older students, and those whose fathers attended college) did so. In addition, the strongest association between authoritarianism and TV viewing occurred for light television viewers. U.S. viewing also appears to diminish, not increase, the cultivation effect. As with sexism, exposure to U.S. television seems to moderate the extreme views cultivated by Argentine television. Morgan & Shanahan found that the Taiwan sample, concerning sex role stereotyping, appeared more sexist across the board than the Argentina sample. The authors found that females who watched more television were more progressive than their male counterparts; the exact opposite of the Argentine sample. Regardless of demographic variables, the greater the amount of television viewing, the stronger the level of sex-role stereotyping. In addition, more viewing of U.S. television programming correlated with higher sexism. 28 With regards to the mean—world hypothesis the authors found a slight cultivation effect. In Taiwan there were lower levels of interpersonal mistrust than in Argentina and these levels were directly related to television exposure. Morgan and Shanahan expected to see that as exposure to U.S. programming increased the Taiwanese subjects views' of violence in society would also increase. Yet when U.S. television viewing was tested as an independent variable there was no change in the mean-world scores. They concluded that "overall exposure to television may cultivate mean world beliefs in Taiwan, but watching U.S. television does not" (p. 189). When Authoritarianism was tested as an independent variable in the Taiwan sample it was found that, while they scored higher overall than Argentineans did, there was no relationship between the scores and television exposure. In some cases heavy viewers of U.S. programs scored lower on the authoritarian scale than light viewers but the relationships were insignificant. Overall, Morgan and Shanahan concluded that Taiwan remained uncultivated. The authors argued that, compared to Argentina and the United states, television appears to be relatively less important in terms of adolescents conceptions of social reality. The results of the Argentina sample, however, indicate that there are similar patterns of cultivation as in various U.S. samples. 29 Reeves 1978 study, measured whether accounting for children's perceived reality of television will contribute to the relationship between pro/anti-social television content and pro/anti-social behavior. The theoretical hypothesis was that "the explanation of TV effects based on TV content would be significantly improved by accounting for perceptions of content realism" (p. 683). Eight behaviors were examined in this study. Five were pro-social and included: altruism, affection, expressing feelings of others, expressing feelings of self, and reparation. Three were anti-social: physical aggression, verbal aggression, and deceit. The results, in general, failed to confirm the hypothesis that perceived reality would impact social behavior. Perceived reality made a significant contribution to one relationship which was seen in exposure to affection on TV and affectionate behavior. In one group composed of fourth graders, there was indeed a strong relationship between exposure to pro-social behavior on television and to pro-social actions in real life. "This result suggests that the pro-social reality items may show higher absolute levels of perceived reality and would therefore increase the impact of pro-social behaviors on TV, but there were no significant differences between the means for items in the pro and anti-social sets” (Reeves, 1978, p. 687). 30 Since fourth graders were the youngest subjects, their correlation's could be explained by the fact that they have less real life experience. "It is probable that children have actual experiences with most pro-social behaviors, and when asked to make judgments about similar TV behaviors, those actual experiences provide the basis for the judgments" (Reeves, 1978, p. 687). For older students, perceived reality predicted only verbally aggressive and deceitful behaviors. Reeves comments that "by eight grade, children may have enough experiences with some anti-social behaviors to compare them accurately with similar TV behaviors. It may also be more socially acceptable to report pro—social behaviors in the fourth grade and anti-social behaviors in the eighth grade" (Reeves, 1978, p. 687). Reeves notes that children are often taught by adults that TV behavior should not be emulated, and that television is not real. Interestingly, Reeves study indicates that perceived reality increased the incidence of pro-social behavior and decreased the incidence of anti-social behavior. Elliott and Slater's (1980) study yielded different results. These authors explored three hypotheses related to law enforcement. The first was that law enforcement program exposure will be positively related to perceptions of program realism. Second, direct experience with law enforcement will reduce this realism. Third, individual differences such as sex, age, and intelligence, will influence perceived program 31 realism. Those least likely to have direct experience with law enforcement (high SES, females, and younger students) will perceive law enforcement programs as more realistic. The authors found strong support for the hypothesis that program exposure is positively associated with perceived realism. However, the authors were unable to determine if the programs are watched because they are seen as realistic or whether they are seen as realistic because they are being watched, which would indicate a reciprocal influence. In one area the predictions of the hypothesis did not match the results. When students having direct negative experience with law enforcement agencies (parole, juvenile detention) were questioned about the realism in TV programming they found the programs to be more realistic than any other group. While those with direct positive experience found, as predicted, that the programs were unrealistic. Studies related to stereotypgs Tan, Li, and Simpson (1985) noted that in 1983 the United States was responsible for importing one-third of Asia's television programming, primarily entertainment. The authors hypothesized that since American television is often the main source of information about American culture, ”it is important to understand how American television is perceived by it's foreign audiences" (p. 809). Tan et a1. (1985) found among their respondents that images of Americans depicted in the programs "Dallas" and 32 "Dynasty", shows that portray Americans as dishonest and materialistic, had a modest effect on international viewers perceiving Americans, generally, in this manner. The heavy viewers had negative impressions of U.S. citizens and were more likely than light viewers to list Americans as dishonest and materialistic. Tan, Tan, and Tan (1987) replicated this study in the Philippines. In this study they first wanted to examine the conflicting values between American programming and traditional Filipino values. They reasoned that heavy viewers would place more importance on the themes that were emphasized in the media, such as wealth, and less value on de-emphasized themes, such as wisdom. Second, they tested the link between American TV viewing and aspirations. The assumption was that if American TV emphasizes material comforts then heavy viewers would want to obtain these comforts and would persue wealth. Moreover, they reasoned that it might lead to a desire to travel or migrate to the United States. The results demonstrated that there was a clear conflict between values highlighted in American programming and typical Filipino values, yet there was no demonstrated cultivation effect. For example, "Pleasure and excitement are emphasized by American TV; the typical Filipino student in our sample does not consider these values to be important” (Tan et a1.,1987, p. 69). 33 While it is unclear if American TV viewing is found to raise the aspirations of Filipino students, a strong correlation was found between those who were exposed to heavy amounts of American television programming and a stronger desire to visit the United States. Social stereotypes are learned through various means. Direct experience with the stereotyped group is one such channel. Another is socialization via parents and peers. Television is a common vehicle for learning stereotypes, particularly when direct contact and experience with a culture is low. Messaris and Woo (1991) studied the stereotypes of Korean-American immigrants and found that 100% of the subjects were at some level encouraged to come to the United States because of images seen in movies, magazines and television programming. The most common stereotypes of the US centered on wealth. In 19 interviews (41 subjects total) wealth was mentioned explicitly. One "less-educated man" was quoted as follows: I thought America was a paradise. TV programs and films I saw made me think it was a fancy and dream land: People here eat good food in fancy restaurants, live in a beautiful house-a huge house with wonderful decorations. But that was an illusion. Even average white Americans cannot do that (p. 77). Critigpe of validity in research literature Through a preliminary view of the literature, it is clear that many studies either fail to confirm, or confirm with weak results the hypotheses tested. These studies have merit as a research device more in the instance of generating 34 more questions and not in proving anything conclusively. There are two erroneous assumptions in previous studies. First, some studies assume that social reality construction is already a proven factor and base their studies on that point. Clearly television viewing will lead to some learning, but not necessarily to a construction of social reality. However, most researchers base their studies on the assumption that reality construction has already taken place, instead of first examining the processes to verify if they did indeed take place. The second error occurs when those who attempt to examine the psychological process of reality construction use the information to draw too many conclusions. Typically, these studies will do a decent job of examining the link between exposure and beliefs but will then confound the issue by expecting that a particular belief will lead to a prescribed behavior. A good example of this is Kang and Morgans 1988 study of AFKN television viewing in Korea. As previously discussed, the authors were able, in women, to establish a link between AFKN viewing and a more "progressiv " attitude about their sexual roles. However, these women still were more likely than any other group to consult their parents about dating, and felt that unrestricted dating was unethical. Obviously, what is considered by the authors to be a more "progressive" view will not lead to women behaving in a more "progressive" manner . 35 Summarv of what is knownyand unknown about the topic Many cultivation studies examine the issue of exposure to violence in the media and the construction about the meanness of society. Initially these were carried out in the United States, but there have been international studies as well as studies performed on recent immigrants to the U.S. In general these studies have typically yielded weak or non-existent results, particularly in international studies. WOber (1978) replicated Gerbner's original study in Great Britain found no support for the Cultivation hypothesis. This led the researcher to offer two possible explanations for the failure to confirm. "One is that what may be true in America is not true in Britain, for which difference it will be useful to explore the reasons. The second is that the Gerbner thesis has still not been demonstrated convincingly in America, and the effect exists neither there nor in Britain" (p. 321). Choi and Tamborini (1988) suggests that instead of focusing solely on violence as a research venue, energies should be concentrated on other concepts such as racism, sexism, youth culture and family values. Indeed, these areas merit further exploration. Contribution this study will make to the literature It is the aim of this study to examine the areas that are typically overlooked in cultivation studies. First, while there have been some studies that apply cultivation to recent immigrants or in an international setting, there are 36 few studies that have examined a population of Asian international students. This is a distinct subset of sojourners because of their high education levels and language competency. In addition, the specific reason for residence in the United States is to further their college education, and it cannot be assumed that they will remain in the U.S. beyond degree completion. Second, this study focuses in depth on the process of learning and possible reality construction based on media programming, something that few cultivation studies fully examine. III. DESCRIPTIVE ANALYSIS Demographic Characteristics The age distribution is presented in Table 1. The respondents range in age from 18 to 39. Among these age groups, exactly one third of the respondents (33.2%) fall within the age range of 18-23. The majority, (43.5%), fall in the age group of 24-30. The remaining group of 31-39 year olds comprise approximately one-fifth of the sample (20.5%). Table 1 Age distribution A Percent 18-23 24-30 31-39 The distribution of sex was quite uneven with over two thirds (65.5%) being men and 32.2 percent being women. The remaining 2.3 percent were unanswered responses. The respondents came from one of five countries, which are listed in Table 2 below. Almost half (42.5%) came from Korea. Approximately one third (29.9%) came from Japan. The 37 38 remaining significant portion (18.4%) of the respondents came from Taiwan. Table 2 Home country distribution Percent Korea Taiwan Jaoan Indonesia China no res-onse Over half of the respondents (of those who chose to answer the question regarding their year in school) were enrolled in a graduate program. The majors for students were various, with the most frequently listed being Business, Communication, and Accounting. Table 3 Education levels Grade Level Number Percent Freshman 5 5.5 re 7 7.9 Junior 15 17 Senior 10 11.3 Graduate 33 37.5 No res 17 19.3 TOTAL 87 98.5 39 Acculturation Variables This study examined various issues regarding the experiences of the students since they have been living in the United States. To begin with the students were questioned as to how long they had been studying in the U.S. Most have been residing in the country for a relatively short period of time. Of the respondents surveyed, exactly half (50.6%) have been living in the United States for two years or less, the other substantial majority (38.5%) of the sample have been living in the U.S. for three to six years. In order assess if the students had direct experience with Americans before they moved to the United States the subjects were asked how many Americans they knew in three settings: in their country, from vacations outside of the US and from brief visits to the US. The figures are presented in Table 4. Table 4 Americans Known In your home Vacations outside Brief visits to country of the US the US none=15 (18.1%) none=50 (64.1%) none=46 (57.5%) one=11 (13.3%) one=7 (9.0%) one=3 (3.8%) two=25 (30.1%) two=5 (6.4%) two=5 (6.3%) three=14 (16.9%) three=3 (3.8%) three=6 (7.5%) four=5 (6.0%) four=1 (1.3%) four=1 (1.3%) five+=13 (15.6%) five+=12 (15.4%) five+=19 (23.6%)47 valid cases 83 valid cases 78 valid cases 80 40 The data presented in Table 4 indicates that the students had relatively little direct experience with Americans before they arrived in the US. The respondents were asked to self-evaluate their English proficiency levels as well. Based on the self report data, language competence does not appear to be a problem. Nearly one-fifth (19.5%) indicated that they were "highly comfortable" when conversing with those who spoke only fluent English while relatively few (2.3%) indicated that they were "not at all comfortable" with conversing with those who spoke only fluent English. The specific data is presented in Table 5. Table 5 Language Competence Highly comfortable 19.5% Very comfortable 26.4% Somewhat comfortable 39.1% Slightly comfortable 12.6% Not at all comfortable 2.3% TOTAL 100% valid cases 87 When the respondents were asked specifically how important it was for them to fully assimilate into American society, about one-third (35.6%) felt it was "very important", while nearly (39.1%) expressed that it was only "somewhat important", clearly the students feel it is important to fit into American society. The variable 41 assimilation was then correlated with other acculturation variables in order to see for this sample what can be called a predictor of assimilation. The correlation coefficients are presented in Table 6. Table 6 Desire to fit in to U.S. society correlations assimilation assimilation assimilation participation r= .0274 in organizations desire to r= .2126 stay in the US upon completion of degree Americans r= -.2199 presently known The desire to stay in the United States upon completion of a degree had a slight to moderate relationship with the international students desire to assimilate into U.S. society Mass Media use among Asian international students In general, the respondents did not watch a lot of television before they moved to the US. Table 7 presents the breakdown of television viewing for the students. 41 42 Table 7 Television exposure (entertainment) pre-US national American other programming programming international programming no hours= 3.6% no hours= 18.1% no hours= 70.3% one hour= 32.1% one hour= 57.8% one hour= 18.9% two hours= 32.1% two hours= 13.3% two hours= 5.4% three hours= 20.2% three hours= 6.0% three hours=1.4% four+ hours= 12.0% four+ hours= 4.8% four+ hours= 4.0% valid cases 84 valid cases 83 valid cases 74 This can be compared with the amount of television that takes place now that the students do live in the United States, as is presented in Table 8. Table 8 During U.S. exposure educational programming entertainment programming sports programming no hours=3.5% no hours=16.7% no hours=36.6% one hour=80% one hour=54.8% one hour=57.3% two hours=11.8% two hours=l9.0% two hours=4.9% three hours=2.4% three hours=3.6% three hours=1.2% four+hours=2.4% four+ hours=4.8% four+hours=0.0% valid cases=85 valid cases=84 valid cases=75 Of particular importance to this study is how accurate the respondents believed the imported American television programming they viewed was. One item asked students ”in general how accurately did you think television portrayed what life in the US would be like?" The second item, which was similarly worded, asked students to respond to how 43 accurate television depicted the way Americans behaved. These responses are listed in Table 9. Table 9 Accuracy of television depictions accurate life accurate behavior highly accurately 0.00% 1.1% very accurately 14.1% 19.5% somewhat 52.9% 47.1% accurately slightly 31.8% 29.9% accurately not at all 1.2% 2.3% accurately TOTAL 100% 100% On both accuracy questions the majority of the sample fell in the somewhat accurately category. Based on the respondents falling in a neutral area it can be said that generally, the respondents did not feel that television provided a strong and accurate picture of either what life in the United States would be like or of how Americans behaved. IV. ANALYTICAL ANALYSIS Hyppthesis Testing Hypothesis 1: For Asian International students, before they arrived in the United States, the television programs they watched in their native country will correlate with their past characterizations of Americans. Context for Hyppthesis One In order to provide a context through which hypothesis one can later be examined, it is important to know how each respondent characterized Americans before they moved to the United States. The mean scores for how each respondent characterized Americans before they moved to the United States is provided in Table 10. 44 45 Table 10 Before characterizations of Americans CHARACTERIZATION MEAN essive Ambitious Ar ntative Arr ant Artistic Athletic Conceited Courteous Effecient Individualistic Industrious Intelli nt Materialistic Me Pleasure-lovin Practical Scientific Sensual Strait-forward In general, the respondents had strong opinions of what Americans were like. When examining the means no adjective ranked below five, which is higher than the mid-point for this scale. There were two adjectives that the respondents, based on the means, believed were the most common characteristics among Americans: individualism and pleasure- loving. Hyppthesis one tested The first step in testing hypothesis one involved selecting the most watched shows before the subjects moved to the United States. The subjects were originally presented ‘with a list of 29 shows that were in syndication in their 46 native countries. However, many of these shows were not watched by the respondents and were immediately eliminated ‘ from.further testing. A complete listing of the frequencies of viewing for all the shows is listed in Appendix A. The criterion for the selection of the shows was as follows: when viewing the frequencies, at least half of the respondents could not have selected "never" when asked how often they viewed the show. There were two exceptions to this rule. For the shows The A-Team and Dynasty slightly over half (52.9%) responded that they never viewed the show but there was a high frequency of viewers who always watched the show (Dynasty 13.8%, A-Team 19.5%) so they were included as well. The shows that met the criterion are as follows: MacGyver, StarTrek, The Cosby Show, Six Million Dollar Man, Dynasty, The A—Team.and The Love Boat. The second step to testing hypothesis one was to run pearson correlations on the seven selected shows for all of the adjectives that typically describe Americans, there was no advance prediction on the direction of the relationship. The correlations are listed in Table 11. 47 Table 11 Before correlations of characteristics with shows * (Significant .05) ** (Significant .01) 48 The coefficients, when examined individually, show in some cases, that there are significant relationships between watching certain shows and corresponding characterizations of Americans. In order to determine if there was a totality of learning for all of the shows and the characteristics, the coefficients were summed across all seven shows and converted to z-scores and averaged, these figures are presented in Table 12. Table 12 Before shows with characterizations summed Characteristic Shows Summed pearson r across shows sive .231 .03 ambitious .140 .14 ntative .334 .05 ant .436 .06 artistic -.402 -.06 athletic -.160 -.02 conceited .291 .04 courteous .622 -.09 effecient -.836 -.12 individualistic -.020 .00 industrious -.941 -.13 intelli .427 .06 materialistic .227 .03 -.452 -.06 easure lov 1.005 . 4 actical -.121 -.01 scientific -.242 -.03 sensual 1.511 -.21* strait forward 1.125 .16 49 Hypothesis one can be analyzed at two levels. Potter's identity hypothesis states that there is a degree of similarity between the viewers perceptions of television characters and situations and the people and situations experienced in real life. When analyzing the way that the Asian international students characterized Americans before they moved here (Table 10) against the correlation coefficients in Table 11, there are in some cases, statistically significant and logical relationships. One example of this is The Cosby Show and the characterization of Americans as ambitious and intelligent. This illustration may demonstrate that television plays more of a role in how the respondents saw Americans than direct experience did. Another example is Dynasty and the characterization of Americans as materialistic. This illustration shows a better matching between the respondents actual characterizations of Americans and the television characterizations. When comparing how the respondents characterized Americans against all of the shows (Table 12), there was only one statistically significant relationship, this was for the characterization of Americans as sensual. At one level this can be explained by the fact that the individual coefficients are counteracting each other. However, according to Gerbner, it is the total learning picture that is important not each individual show. 50 Hypothesis one can be partially confirmed. The results are not strong when analyzing them either individually or combined but the television programs that were watched did, in some instances correlate with the characterizations of Americans. This was certainly more the case when examining each show individually as in Table 11 than examining them summed as in Table 12. Hypothesis 2: For Asian international students, after they arrived in the United States, the television programs that they currently watch will correlate with their current characterizations of Americans. Context for Hyppthesis Two In order to provide a context through which hypothesis two can later be examined, it is important to know how each respondent characterized Americans after they moved to the United States. The mean scores for how each respondent characterized Americans after they moved to the United States is provided in Table 13. 51 Table 13 Current characterizations of Americans CHARACTERIZATION MEAN sive Ambitious Ar ntative ant Artistic Athletic Conceited Courteous Effecient Individualistic Industrious Intelli t Materialistic Merce Pleasure-10v Practical Scientific Sensual Strait-forward Generally, the respondents slightly weaker characterizations of Americans now that they have spent time in the United States. In four instances they mean score for the characterizations were slightly higher. This can be seen by comparing the means in Table 10 and Table 13. Hyppthesis two tested The steps for testing hypothesis two were very similar to steps involved in testing hypothesis one. First, there was the process of show selection. The respondents were originally presented with a list of 13 shows. The criterion for the selection of shows was as follows: when viewing the frequencies, at least half of the 52 respondents could not have selected never when asked how often they viewed the show. There was one exception: in the case of the show Beverly Hills 90210, 51.7 per cent of the respondents never watched the show, but one-fourth of the population watched the show either more than once a week or every week so it was included in the selection of shows. The frequencies for the original thirteen shows can be seen in Appendix B. The final selection of shows is as follows: L.A. Law, The Cosby Show, Married with Children, and Beverly Hills 90210. Second, pearson correlations were run on the selected shows with the respondents current characterizations of Americans. There was no advance prediction about the direction of the relationships. The coefficients are shown for each show and each characterization in Table 14. 53 Table 14 Current correlations of characteristics with shows Beverly L . A . Law The Cosby Married Hills Show ‘with 90210 Children aggressive -.18* .01 .04 .02 ambitious -.17 -.05 .18* -.02 argumentative -.15 .04 .13 -.13 arrogant .02 -.04 .10 -.14 artistic -.01 -.11 .03 -.13 athletic .05 -.10 .10 -.20* conceited .00 .21* .23* .01 courteous -.13 -.08 -.00 -.13 effecient -.06 -.14 -.21* -.04 individualistic '-.15 -.14 -.04 ‘-.18* industrious -.03 -.04 -.12 -.12 intelligent -.05 -.17 .06 .05 materialistic .01 -.14 .26** -.03 mercenary .06 .00 -.18* -.17 Pleasure loving -.26** -.18* .16 -.13 practical .06 -.08 .02 -.04 scientific -.05 -.13 -.20* -.01 sensual -.01 .00 .32** -.10 strait forward -.07 —.06 .13 -.13 The final step in testing hypothesis two involved an averaging across the shows, in order to determine if there was a total learning for each of the characteristics. This was done by changing the correlation coefficients to z- scores and obtaining an average for each characterization across all of the shows. coefficients after the transformations. Table 15 lists shows the final 54 Table 15 Current shows with characterizations summed Characteristic Total show Z- pearson r after scores before averaging averaging aggressive -.112 -.03 ambitious -.060 -.02 argumentative -.191 -.05 arrogant -.061 -.02 artistic -.221 -.06 athletic -.153 -.04 conceited .457 .11 courteous -.342 -.09 effecient -.454 -.11 individualistic -.514 -.13 industrious -.312 -.08 intelligent -.112 -.03 materialistic .105 .03 mercenary -.294 -.O7 pleasure loving -.418 -.10 practical -.040 -.01 scientific -.394 -.10 sensual .222 .06 strait forward -.130 -.03 Hypothesis two can be partially confirmed. can be analyzed at two levels. When analyzing the way Asian The results international students characterized Americans after they moved to the United States (Table 13) compared with Table 14, there are in some cases statistically significant relationships. One of the strongest characterizations of Americans was for the adjective pleasure loving. There are two correlations between pleasure loving and the shows Beverly Hills 90210 and L.A. Law. Another example is the characterization of Americans as materialistic and the correlation between materialism and The Cosby Show. 55 When comparing how the respondents characterized Americans across all of the shows (Table 15), there are no significant relationships. This may be explained by some of the weaker coefficients canceling out the stronger coefficients. Interestingly, there seems to be less correlation between the shows and the characterizations of Americans for the respondents after they moved to the U.S. than before. This may be accounted for two reasons: perhaps there were more depiction's of the characterizations of Americans in the before sample; or since the students have more direct experience with Americans, they are less likely to rely on television as their primary source of information about Americans. HypOthesis 3: For Asian international students, there 'will be a significant difference between (a) how the low acculturation subgroup's characterizations of Americans correlate with the current shows that they watch and (b) how the high acculturation subgroup's characterizations of Americans correlate with the current shows that they watch. Selection of acculturation measure The measures for the acculturation variables for JHypothesis three were derived from a 1976 study by Y.Y. Kim in which she identified: intimacy levels with Americans, idesire to stay in America, and acculturation motivation as factors related to immigrants interpersonal communication activities. 56 -The desire to stay in America after degree completion was tested as a measure of acculturation with the characteristics courteous, practical and ambitious. The students were specifically asked "Do you intend to remain in the United States after completion of your degree?" -The degree to which "fitting in" to American society is important was tested as a measure of acculturation with the characteristics courteous, practical and ambitious. The students were specifically asked "How important is it for you to fully assimilate into American society?" -The degree of intimacy that is felt between the international students and Americans was tested as a measure of acculturation with the characteristics courteous, practical and ambitious. The students were specifically asked "Do you consider most of the Americans that you know classmates, casual friends or intimate friends?" The original intention was to create a single measure of acculturation, as is common in many studies. However, there ‘was little inter-correlation between the above listed lneasures as can be seen in Table 6. Therefore, it had to be ‘determined which acculturation variable most accurately :measured acculturation: the desire to stay in U.S., the .importance of assimilation, or intimacy levels. The respondents were asked to rate characteristics that describe Americans twice, how they viewed Americans before 'they moved to the U.S. and after they moved to the U.S. The .after'U.S. scores were used for the correlations with the 57 assimilation variables because it was assumed that these scores would be based on direct experience with Americans versus the before U.S. scores. The acculturation variables were tested against three positive characterization of Americans: ambitious, courteous, and practical. Correlation coefficients for these variables are printed in Table 16. Table 16 Correlation's for acculturation measures ambitious courteous practical Desire to .06 -.14 .07 stay important to -.46 .49 .27 assimilate friendship .04 —.12 -.03 level The only acculturation measure that seemed to have a :modest correlation with the variables was the importance of assimilation. The variable assimilation was split into those 'that ranked high on the measure versus those who scored low on the measure. The assimilation groups were then tested separately against two shows and corresponding characteristics. The selected show were Beverly Hills 90210 and The Cosby Show. These shows were selected because they both have a commonalty (If high socioeconomic status presented in Urban settings vflrile they differ in the ethnicity of the main characters. The results for the high and low assimilation groups are presented in Table 17 for Beverly Hills 90210 and in Table 18 for The Cosby Show. Table 17 High vs Low Assimilation for Beverly Hills 90210 Adjective low mean 10" high high differ-no- correlation mean corrolaticn between two mwmn aggressive 6.66 -.19 7.06 .09 —.49 ambitious 6.44 -.18 5.61 -.11 -.35 argumentative 6.69 -.15 6.82 -.13 -.07 arrogant 6.18 .06 6.28 -.18 .90 artistic 5.92 -.04 4.44 .22 -.94 athletic 7.53 .05 6.88 .01 .15 conceited 5.75 -.02 6.45 .10 -.41 courteous 5.95 -.12 5.72 -.20 .30 effecient 5.42 -.06 4.94 -.07 -.04 individualistic 8.44 -.17 8.00 .04 -.80 industrious 5.97 -.04 5.47 .04 -.30 intelligent 5.18 -.06 5.44 .03 -.34 materialistic 6.98 .05 7.59 -.23 1.34 mercenary 5.76 .00 4.29 .37 -1.23 pleasure 7.86 -.25 7.89 -.42 .67 loving ,practical 5.97 .03 6.78 .22 -.72 scientific 5.74 -.08 5.69 .28 -1.27 sensual 6.18 -.02 5.94 .03 -.18 strait 7.11 -.13 6.00 .29 -1.56 forward G 17. [91" ch the 59 Table 18: High vs Low Assimilation for The Cosby Show Adjective low mean low high high difference correlation mean correlation between two imwun aggressive 6.66 .00 7.06 .31 1.05 ambitious 6.44 .15 5.61 .35 .74 argumentative 6.69 .10 6.82 .40 1.05 arrogant 6.18 .07 6.28 .23 .60 artistic 5.92 .04 4.44 .02 -.06 athletic 7.53 .13 6.88 -.04 -.57 conceited 5.75 .22 6.45 .34 .40 courteous 5.95 -.02 5.72 .09 .38 effecient 5.42 -.26* 4.94 .10 1.23 individualistic 8.44 -.03 8.00 -.04 -.03 industrious 5.97 -.17 5.47 .16 1.48 intelligent 5.18 .14 5.44 -.48* -2.30 materialistic 6.98 .24 7.59 .36 -2.10 mercenary 5.76 -.19 4.29 -.20 -.03 pleasure 7.86 .06 7.89 .72** 2.26 loving practical 5.97 -.05 6.78 .35 1.44 scientific 5.74 -.20 5.69 -.26 -.11 sensual 6.18 .28* 5.94 .51* .89 strait 7.11 .12 6.00 .24 .41 forward Hypothesis three can be partially confirmed. There are no statistically significant differences between high and low acculturation groups for the show Beverly Hills 90210 and the corresponding characteristics. The Cosby Show. However, there are a few for This may be accounted for two reasons: either the respondents found The Cosby Show more closely related to the way Americans live and behave; or the characteristics may be present more in The Cosby Show than they are in Beverly Hills 90210. Nevertheless, there are a few significant differences between the two groups which may 60 indicate that in certain circumstances acculturation variables play a role in the learning process. Hypothesis 4: For Asian international students, the themes that they see in American television will correlate with their endorsement of prevalent problems in the United States. Context for Hyppthesis Four In order to provide a context with which hypothesis four can be analyzed, it is necessary to know what the respondents believe are prevalent problems in American society. The mean scores for each prevalent problem are presented in Table 19. Table 19 Means for prevalent problems in society Prevalent Problem Mean Score crime 8.56 discrimination against Blacks 6.89 discrimination against 4.33 Elderly discrimination against Whites 2.65 environmental negligence 4.87 homelessness 5.70 negative stereotyping of 6.47 Asians negative stereotyping of 6.57 Blacks negative stereotyping of Gays 5.68 and Lesbians negative stereotyping of 5.76 Hispanics negative Stereotyping of 3.37 Whites peer pressure 5.40 Apoverty 5.74 racial conflicts 7.04 sexual discrimination 6.11 sexual promiscuity 6.98 teen pregnancy 6.26 violence 8.32 wealth 4.60 62 The mean score for the prevalent problems break down into three areas. The discrimination and stereotyping items against ethnic minorities all receive moderate endorsement as prevalent problems in society. Discrimination and stereotyping of Whites both received little endorsement. Crime and violence both received high endorsement. The subjects were asked in one item to indicate how often they saw the following themes in American television. In a separate item they were asked to indicate what problems they thought were most prevalent in our society. The themes and the problems were a list of twenty identical adjectives. Using pearson correlations each adjective as a theme is correlated the corresponding adjective as a problem, the coefficients are presented in Table 20. Table 20 Correlations TV themes a prevalent problems crime=.42** negative stereotyping of Whites=.50** discrimination against peer pressure=.44** Blacks=.28* discrimination against poverty=.13 Elderly=.37** discrimination against racial conflicts=.40** Whites=.47** environmental negligence=.27 sexual discrimination=.32** homelessness=.16 sexual promiscuity§.26 negative stereotyping of teen pregnancy=.22 Asians=.47** negative stereotyping of violence=.38** Blacks=.44** negative stereotyping of wealth= -.11 Gays and Lesbians=.53** negative stereotyping of work conflicts=.17 Hispanics=.39** 63 The correlation coefficients for many of the items indicate definite correlations at a moderate level. In contrast to the mean scores for each item as a prevalent problem, the typical pattern shows for the discrimination and stereotyping of whites, when correlated as a prevalent theme in society, the correlations are stronger than any other items. This indicates that as theme as well discrimination and stereotyping of whites is not endorsed as being seen in television. Table 21 list the coefficients of determination for each variable below. Table 21 Coefficients of determination (r2) crime=.17 negative stereotyping of Whites=.25 discrimination against Blacks=.07 peer pressure=.19 discrimination against E1derly=.13 poverty=.016 discrimination against Whites=.22 racial conflicts=.16 environmental negligence=.07 sexual discrimination=.10 homelessness=.02 sexual promiscuity=.06 negative stereotyping of Asians=.22 teen pregnancy=.04 negative stereotyping of Blacks=.19 violence=.14 negative stereotyping of Gays and Lesbians=.28 wealth= .01 negative stereotyping of Hispanics=.15 work conflicts=.02 Hypothesis four is generally confirmed. While the results are not strong, the prediction is confirmed, what is endorsed as a prevalent problem in society generally 64 correlates with themes seen on television. One prediction that was not made but that occurred was the themes and stereotyping of whites. In this instance, this was neither endorsed as a prevalent problem nor endorsed as a theme that is seen on television. Hypothesis 5: For Asian international students, those who are exposed to more television will indicate higher ratings on the prevalent problems in the United States scale than the light viewers will. The steps for testing hypothesis 5 were identical as that for hypothesis 4 with one change: correlations for the heavy television viewers and light television viewers were measured separately. Typically in cultivation studies, a heavy viewer is defined as someone who watches four or more hours of television a day. However, as is shown in table eight: current viewing hours for entertainment, it is shown that there is a strong percent of non-viewers, and that the .majority of the sample (54.8%) watch only one hour per day. Therefore, for this population a heavy viewer is defined as one who watches two or more hours of television a day. In order to test hypothesis five, the heavy viewers were separated from the light viewers based on their hours of :media exposure and separated correlation tests were performed on each group. Table 22 shows the results for the light *viewers and Table 23 shows the results for the heavy viewers. Table 22 Correlations TV themes & prevalent problems (light) 65 Problem Mean Score Correlation with Theme crime 8.34 .39** discrimination 6.57 -.25 against Blacks discrimination 4.57 .30** against Elderly discrimination 2.98 .48** against Whites environmental 4.88 .15 negligence homelessness 5.62 .17 negative 6.72 .47** stereotyping of Asians negative 6.40 .43** stereotyping of Blacks negative 5.73 .51** stereotyping of Gays and Lesbians negative 5.55 .47** stereotyping of ‘ Hispanics negative 3.36 .48** stereotyping of Whites peer pressure 5.28 .48** poverty 5.80 .07 racial conflicts 6.68 .51** sexual 5.86 .33* discrimination sexual promiscuity 6.97 .22 teen pregnancy 6.07 .25 violence 8.00 .44** wealth 4.49 .01 work conflicts 4.78 .10 66 Generally, for the light viewers they're endorsement of prevalent problems break down into two groups. First, they moderately endorsed the stereotyping and discrimination of ethnic minority items as prevalent problems in society while endorsing less those items for whites. Second, they strongly endorsed the items violence and crime as prevalent problems in society. 67 Table 23 Correlations TV themes & prevalent problems (heavy) Problem Mean Score Correlation with Theme crime 9.03 .49** discrimination 7.55 .35 against Blacks discrimination 3.84 .56** against Elderly discrimination 1.91 .31 against Whites environmental 4.88 .56* negligence homelessness 5.88 .22 negative 5.96 .47* stereotyping of Asians negative 6.92 .45* stereotyping of Blacks negative 5.56 .57** stereotyping of Gays and Lesbians negative 6.25 .41* stereotyping of Hispanics negative 3.41 .71** stereotyping of Whites peer pressure 5.65 .34 poverty 5.61 .25 racial conflicts 7.81 .38* sexual 6.70 .65** discrimination sexual promiscuity 7.04 .35 teen pregnancy 6.65 .15 violence 9.08 .17 wealth 4.84 .-24 work conflicts 5.68 .26 68 Typically, the heavy viewers did endorse the prevalent problems in society slightly stronger. There is a general pattern of endorsement for both the light and heavy viewers of stereotyping and discrimination of ethnic minorities receiving moderate endorsement as prevalent problems in society. There is a general pattern of crime and violence receiving strong endorsement as prevalent problems in society. However, for the heavy viewers, there were not necessarily correlations between the problem and the theme endorsement. This is evident in the endorsement of violence as a problem in society but of the weak correlation between violence as a theme in television and as a problem in society. The final step for testing hypothesis five involved a T-test for significant difference between the heavy viewers coefficients and the light viewers, the result are provided in Table 24. 69 Table 24 T—Test for difference between heavy & light viewers crime=-.50 negative stereotyping of whites= -1.63 discrimination against blacks=-2.77 peer pressure=.75 discrimination against elderly=-1.57 poverty= -.69 discrimination against whites=.90 racial conflicts=.67 environmental negligence= -2015 sexual discrimination= 1.93 homelessness=-.23 sexual promiscuity= -.63 negative stereotyping of Asians= .00 teen pregnancy=.44 negative stereotyping of blacks= -.11 violence=1.38 negative stereotyping of wealth= 1.07 gays and lesbians=-.38 negative stereotyping of work conflicts= -.68 hispanics=.33 Hypothesis five can be partially confirmed. When testing for significant difference between the two groups, two items environmental negligence and discrimination against Blacks are significantly different and negative stereotyping of Whites is close. It could be said that the hypothesis is true for certain themes and problems but not for all situations. Summagy Hypothesis one and two were constructed based on the idea of Potters identity dimension which states that there is a degree of similarity between what the viewer perceives between television characters and situations and real life characters and situations. 70 When framed for this study, this theory states that watching television may lead to characterizations or stereotyping of Americans. For Asian international students, this was the situation to a certain degree. Hypothesis one and two were both partially accepted. There were often low correlations between the programs that were viewed and how the students characterized Americans but there were a few significant relationships; this indicates that television exposure may lead to certain amounts of stereotyping. Hypothesis three examined acculturation/assimilation variables role in Asian students characterizations about Americans. There were no significant relationships for one show, Beverly Hills 90210, but there were a few for The Cosby Show. On this basis it could be said that acculturation variables may play a role in how Asian international students characterize Americans in certain circumstances. I Hypothesis four was framed around Potter's identity dimension. Hypothesis four stated that the themes that Asian international students saw in American television programming would correlate with the problems that the respondents saw as prevalent in society. This hypothesis was partially confirmed. There were several significant relationships between the endorsement of prevalent problems in society and themes seen on television. Hypothesis five was a typical light-heavy viewer measure as is seen in most cultivation studies. The hypothesis stated that heavy viewers would be more likely to endorse 71 prevalent problems in American society than light viewers would, based on television exposure. In general, this prediction was confirmed but there were only a few cases where the coefficients were significantly different for the two subgroups. However, sufficient relationships were demonstrated to partially confirm hypothesis five. V. DISCUSSION When working within a typical cultivation framework it is generally assumed that social reality construction will take place. Rarely do these studies stop and examine the learning process, let alone analyze the steps between learning and perceived reality. This study took a different approach. By a careful review and critique of the literature it was determined that the processes of reality construction through television needed careful examination. First, in order to examine the process of learning or reality construction through television, there must be an ample amount of exposure. There are very low levels of entertainment programming being watched among this jpopulation. The levels of exposure, in addition, are much lower now than they were before the students moved to this country . Second, exposure to American television programming and real life experiences should converge to bring about some levels of learning about Americans and life in the United States. This study measure learning through opinions or characterizations about the Americans. 72 73 It can be said with confidence that the Asian international student sample had strong opinions about the United States. This is a learning process, it is just indeterminable exactly how the learning took place. When students were asked how accurate television represented life in the United States 0 percent felt that TV portrayals were accurate. Similarly, when the respondents were asked how accurate the thought television depicted how American behaved and lived 1.1% felt that TV provided an accurate representation. These results indicate that television was probably not the main source of this populations learning about the U.S. before they moved here. If learning through television does occur, it could be measurable through should lead to an estimate of real world situations, as Hawkins and Pingree (1990) and Potter (1986) deduce. This study attempted to measure this by having the population identify themes that are pervasive in current television programming and then translate them back into U.S. social problems (estimate of real world situations). To a certain extent it worked, there were many significant correlations between the themes and the problems, some of them at moderate levels. Pram—ens There were several problems encountered with this study. The biggest problems will be discussed below. To begin with, not enough was known about the populations exposure patterns when the were living in their native country. One of the key 74 premises of cultivation studies is that people are exposed to television. The population did have access to television, actually their overall viewing hours were higher before they moved to the United States than the are now. However, the fact still remain that television viewing was not a big part of this populations daily lives. In a classic cultivation study a heavy viewer is one who watches four or more hours of television per day, but for this sample a heavy viewer was defined as one who watched two or more hours per day. Basically, this study split the viewers into two groups, light and moderate, instead of three. Low exposure levels also contradict another central ideal of cultivation studies--selectivity. Cultivation researchers reason that, in the United States, heavy television viewers tend to be non-selective about the programs that they watch. This research population had to be extremely selective with what television programs they 'watched, because over half of them.watched 0 to 1 hours of television per day. The issue of selectivity may account for two problems with this study. First, many of the shows, both before the population moved to the United States and after, that were selected for the were not included in the final correlation tests because of low viewer numbers. It wasn't assumed that every show ‘would be watched by every respondent, but the pattern of 'viewing tended to show that either significant portion of the 75 sample watched a certain show or no one watched the show. Second, if selectivity is indeed a valid issue then the population for this study did not receive a wide variety of programming with which to form diverse impressions about the United States. This is problematic for this study because four sections of the study (show themes, prevalent problems, and before/after characterizations) all are designed to measure a wide variety of issues. The majority of problems with this study were centered around instrumentation and design. With regards to the instrument itself not enough pre-testing was performed on the prevalent problems and themes section. While it can be argued that these shows display the themes, specifically pre- testing them an Asian population would have been logical. However, the best way to analyze the shows for the themes ‘would have to perform a message/content analysis. Although this is time consuming it provides the most objective measure of program content. There is one final instrument construction problem to note. Often times it is desirable to pose open end questions -to one's population. This study incorporated one open ended iquestion which asked the respondents to discuss which specific impressions of American television turned out to be ‘true and which impressions did not. Out of 87 cases, less 'than 20 percent chose to answer the question. Although it ‘was interesting to look at some of the responses it could not be incorporated into this study. 76 Finally, while this study did attempt to measure how Asian students characterized Americans it was presumed that the learning occurred through television. A better approach to performing this study would have been to find out by verbal interviews what other factors contributed to the populations learning about Americans. Although the nature of cultivation studies depend primarily on learning through television exposure, identifying other factor that contribute to learning could only enhance the understanding of the overall cultivation process. Contributions of this study to the cultivation process Although this study did not overwhelmingly support it's hypotheses, its content can still be valuable to the overall understanding of the cultivation process, particularly with regards to the construction of social realities. The data from this study indicates that there are relationships that merit closer examination. The entire prevalent problems and corresponding themes section, which tested hypothesis four and five, and were constructed based on Potter's identity dimension which states that the degree of similarity the viewer perceives between television characters and situation and the people and situations experienced in real life, yielded interesting results. Hypothesis four and five overall showed the most promise for future research. Future studies should definitely examine more relationships between heavy viewers and light viewers and even subdivide these groups by sex, age or race 77 if possible. Valuable information about the cultivation processes can best be gained by fully examining these subgroup relationships. Conclusions One of the aims of this study was to determine if there was a correlation between exposure and a construction of real world experiences based on that exposure. Examining this idea with the use of the identity dimension provided a framework with which this relationships could be examined in a direct and non-biased manner. Typically, cultivation studies will attempt to match reality construction through television by asking respondents ‘what their chances are of being involved in an act of ‘violence and then provide them with two choices: one being the real world answer and another an answer that is based on television. By providing the respondents with the choices 'the chances of obtaining an accurate idea of what the respondents are learning are significantly lowered. More 'validity can be lended to examining the concept of learning 'through television by letting the respondents choose freely rmhat themes they think are prevalent and what problems they :feel are pervasive. The results may not be as strong, but ‘they are arguably less biased and hopefully, more accurate. APPENDICES AEHHNWDIX A. Frequencies of viewing: Pre U.S.Shows THREE SCO Valid Cum Value Label Value Frequency Percent Percent Percent 1 11 12.6 12.6 12.6 2 6 6.9 6.9 19.5 3 9 10.3 10.3 29.9 4 61 70.1 70.1 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 HAWAI I PO Valid Cum Value Label Value Frequency Percent Percent Percent 1 9 10.3 10.3 10.3 2 7 8.0 8.0 18.4 3 11 12.6 12.6 31.0 4 60 69.0 69.0 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 LOVEBOAT Valid Cum Value Label Value Frequency Percent Percent Percent 1 6 6.9 6.9 6.9 2 21 24.1 24.1 31.0 3 16 18.4 18.4 49.4 4 44 50.6 50.6 100.0 Total 87 100.0 100.0 Valid.cases 87 Missing cases 0 MISSIONI Valid Cum ‘Value Label value Frequency Percent Percent Percent 1 18 20.7 20.7 20.7 2 14 16.1 16.1 36.8 3 6 6.9 6.9 43.7 4 49 56.3 56.3 100.0 78 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 KOJAK Valid Cum Value Label Value Frequency Percent Percent Percent 1 4 4.6 4.7 4.7 2 12 13.8 14.0 18.6 3 9 10.3 10.5 29.1 4 61 70.1 70.9 100.0 9 1 1.1 Missing Total 87 100.0 100.0 Valid cases 86 Missing cases 1 ATEAM Valid Cum Value Label Value Frequency Percent Percent Percent 1 17 19.5 19.5 19.5 2 12 13.8 13.8 33.3 3 12 13.8 13.8 47.1 4 46 52.9 52.9 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 THECOSHO Valid Cum ‘Value Label Value Frequency Percent Percent Percent 1 32 36.8 36.8 36.8 2 24 27.6 27.6 64.4 3 12 13.8 13.8 78.2 4 19 21.8 21.8 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 CHUURLIES Valid Cum Value Label Value Frequency Percent Percent Percent 1 18 20.7 20.7 20.7 2 8 9.2 9.2 29.9 3 13 14.9 14.9 44.8 4 48 55.2 55.2 100.0 Total 87 100.0 100.0 Valid cases DIFFERST Value Label Valid cases STARTREK Value Label Valid cases SIXMIDOL Value Label Valid cases LASSIE Value Label Valid cases DAIJJHB Value Label 87 87 87 87 87 bUNH Total AUNI—l Total :5me Total IwaI-l Total 80 Missing cases Missing cases Missing cases Missing cases Missing cases Valid Value Frequency Percent Percent 6 6.9 6.9 1 1.1 1.1 12 13.8 13.8 68 78.2 78.2 87 100.0 100.0 Valid Value Frequency Percent Percent 21 24.1 24.1 21 24.1 24.1 18 20.7 20.7 27 31.0 31.0 87 100.0 100.0 Valid Value Frequency Percent Percent 23 26.4 26.4 12 13.8 13.8 10 11.5 11.5 42 48.3 48.3 87 100.0 100.0 Valid Value Frequency Percent Percent 7 8.0 8.0 6 6.9 6.9 15 17.2 17.2 59 67.8 67.8 87 100.0 100.0 Valid Percent Value Frequency Percent Cum Percent Cum Percent 24.1 48.3 69.0 100.0 Cum Percent 26.4 40.2 51.7 100.0 Cum' Percent 8.0 14.9 32.2 100.0 Cum Percent 81 1 10 11.5 11.5 11.5 2 13 14.9 14.9 26.4 3 14 16.1 16.1 42.5 4 50 57.5 57.5 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 DYNASTY Valid Cum Value Label Value Frequency Percent Percent Percent 1 12 13.8 13.8 13.8 2 12 13.8 13.8 27.6 3 17 19.5 19.5 47.1 4 46 52.9 52.9 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 MAGNUMPI Valid Cum Value Label Value Frequency Percent Percent Percent l 9 10.3 10.3 10.3 2 9 10.3 10.3 20.7 3 12 13.8 13.8 34.5 4 57 65.5 65.5 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 HOTEL Valid Cum Value Label Value Frequency Percent Percent Percent 1 2 2.3 2.3 2.3 2 2 2.3 2.3 4.6 3 7 8.0 8.0 12.6 4 76 87.4 87.4 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 QUINCY Valid Cum Value Label Value Frequency Percent Percent Percent 1 4 4.6 4.6 4.6 2 4 4.6 4.6 9.2 3 12 13.8 13.8 23.0 4 67 77.0 77.0 100.0 Total 87 100.0 100.0 82 Valid cases 87 Missing cases 0 HARTHART Valid Cum Value Label Value Frequency Percent Percent Percent 1 9 10.3 10.6 10.6 2 11 12.6 12.9 23.5 3 4 4.6 4.7 28.2 4 61 70.1 71.8 100.0 9 2 2.3 Missing Total 87 100.0 100.0 Valid cases 85 Missing cases 2 FANTASYI Valid Cum Value Label Value Frequency Percent Percent Percent 1 4 4.6 4.8 4.8 2 16 18.4 19.0 23.8 3 9 10.3 10.7 34.5 4 55 63.2 65.5 100.0 9 3 3.4 Missing Total 87 100.0 100.0 Valid cases 84 Missing cases 3 MACGYVER Valid Cum Value Label Value Frequency Percent Percent Percent 1 39 44.8 44.8 44.8 2 8 9.2 9.2 54.0 3 7 8.0 8.0 62.1 4 33 37.9 37.9 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 CHEERS Valid Cum Value Label Value Frequency Percent Percent Percent 1 16 18.4 18.4 18.4 2 10 11.5 11.5 29.9 3 10 11.5 11.5 41.4 4 51 58.6 58.6 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 JEFFER 83 Valid Cum Value Label Value Frequency Percent Percent Percent 1 5 5.7 5.7 5.7 2 8 9.2 9.2 14.9 3 12 13.8 13.8- 28.7 4 62 71.3 71.3 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 HAPPYD Valid Cum Value Label Value Frequency Percent Percent Percent 1 5 5.7 5.7 5.7 2 8 9.2 9.2 14.9 3 4 4.6 4.6 19.5 4 70 80.5 80.5 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 GOODTIME Valid Cum Value Label Value Frequency Percent Percent Percent 1 3 3.4 3.4 3.4 2 4 4.6 4.6 8.0 3 8 9.2 9.2 17.2 4 72 82.8 82.8 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 LALAW Valid Cum Value Label Value Frequency Percent Percent Percent 1 6 6.9 6.9 6.9 2 14 16.1 16.1 23.0 3 12 13.8 13.8 36.8 4 55 63.2 63.2 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 ROSEANNE Valid Cum Value Label Value Frequency Percent Percent Percent 1 11 12.6 12.6 12.6 2 5 5.7 5.7 18.4 3 6 6.9 6.9 25.3 84 4 65 74.7 74.7 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 ADIFWRLD Valid Cum Value Label Value Frequency Percent Percent Percent 1 7 8.0 8.0 8.0 2 9 10.3 10.3 18.4 3 2 2.3 2.3 20.7 4 69 79.3 79.3 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 JUMPZ IST Valid Cum Value Label Value Frequency Percent Percent Percent 1 11 12.6 12.6 12.6 2 6 6.9 6.9 19.5 3 12 13.8 13.8 33.3 4 58 66.7 66.7 100.0 Total 87 100.0 100.0 Valid cases 87 Missing cases 0 MARRIEDW Valid Cum Value Label Value Frequency Percent Percent Percent 1 15 17.2 17.4 17.4 2 10 11.5 11.6 29.1 3 8 9.2 9.3 38.4 4 53 60.9 61.6 100.0 9 1 1.1 Missing Total 87 100.0 100.0 ‘Valid cases 86 Missing cases 1 ROSE Value Label Mean Maximum Valid cases STEP Value Label Mean Maximum Valid cases ‘WINGS Value Label Mean Maximum Valid cases BEV Value Label .APPENEUDKIB Frequencies of viewing: Current shows 3.149 4.000 87 3.862 4.000 87 3.632 4.000 87 Value Frequency fiuNI-I Total Std dev Missing cases 2 3 4 Total Std dev Missing cases Valid Value Frequency Percent Percent 2 2.3 2.3 8 9.2 9.2 77 88.5 88.5 87 100.0 100.0 .408 Minimum Valid Value Frequency Percent Percent 2 3 4 Total Std dev Missing cases Value Frequency Percent Percent 85 Valid Percent Percent 13 14.9 14.9 10 11.5 11.5 15 17.2 17.2 49 56.3 56.3 87 100.0 100.0 1.126 Minimum 9 10.3 10.3 14 16.1 16.1 64 73.6 73.6 87 100.0 100.0 .667 Minimum Valid Cum Percent 14.9 26.4 43.7 100.0 1.000 Cum Percent 2.3 11.5 100.0 2.000 Cum Percent 10.3 26.4 100.0 2.000 Cum Percent Ahearu ImaxfiiJnum Valid cases FULL Value Label Mealxnt Maxchrnum Va1.jL<3 cases LA Va lue Label Me‘="=jl1num Va 1 id cases c<:”531312 value Label 3.172 4.000 87 3.184 4.000 87 3.241 4.000 87 2.954 4.000 87 DWNH Total Std dev Value Frequency Percent Percent fiUNI—l Total Std dev waI-i Total Std dev fith-i Total 13 14.9 14.9 10 11.5 11.5 12 13.8 13.8 52 59.8 59.8 87 100.0 100.0 1.136 Minimum 0 Valid Value Frequency Percent Percent 4 4.6 4.6 14 16.1 16.1 26 29.9 29.9 43 49.4 49.4 87 100.0 100.0 . 889 Minimum Valid Value Frequency Percent Percent 9 10.3 10.3 16 18.4 18.4 32 36.8 36.8 30 34.5 34.5 87 100.0 100.0 .975 Minimum Std dev 86 Missing cases Missing cases Missing cases Missing cases 9 10.3 10.3 12 13.8 13.8 21 24.1 24.1 45 51.7 51.7 87 100.0 100.0 1.025 Minimum 0 Valid 10.3 24.1 48.3 100.0 1.000 Cum Percent 14.9 26.4 40.2 100.0 1.000 Cum Percent 4.6 20.7 50.6 100.0 1.000 Cum Percent 10.3 28.7 65.5 100.0 1.000 NORTH Value Label Mean Maximum MEL Value Label Mean Maximum Valid cases MARR Value Label Mean Maximum SIS Value Label Mean Maxdnmmt Valid cases 3.644 4.000 3.609 4.000 87 2.851 4.000 3.736 4.000 87 Value uthN Total Std dev Value thI-l Total Std dev 87 Missing cases Value thH Total Std dev Value #MNH Total Std dev Valid Frequency Percent Percent 8 9.2 9.2 15 17.2 17.2 64 73 6 73 6 87 100.0 100.0 .647 Minimum Valid Frequency Percent Percent 4 4.6 4.6 5 5.7 5.7 12 13.8 13.8 66 75.9 75.9 87 100.0 100.0 .798 Minimum Valid Frequency Percent Percent 16 18.4 18.4 20 23.0 23.0 12 13.8 13.8 39 44.8 44.8 87 100.0 100.0 1.186 Minimum Valid Frequency Percent Percent 2 2.3 2.3 5 5.7 5.7 7 8.0 8.0 73 83.9 83.9 87 100.0 100.0 .673 Minimum Missing cases Cum Percent 9.2 26.4 100.0 2.000 Cum Percent 4.6 10.3 24.1 100.0 1.000 Cum Percent 18.4 41.4 55.2 100.0 1.000 Cum Percent OOtQN OI-‘OU 1 10 1.000 MAD Value Label Mean Maximum Valid cases PAM Value Label Mean Maximum Valid cases 3.690 4.000 87 3.598 4.000 87 88 Valid Value Frequency Percent Percent 2 6 6.9 6.9 3 15 17.2 17.2 4 66 75.9 75.9 Total 87 100.0 100.0 Std dev .597 Minimum Missing cases Valid Value Frequency Percent Percent hUNI-I Total Std dev Missing cases 1 11 10 1.1 1.1 12.6 12.6 11.5 11.5 74.7 74.7 100.0 100.0 Minimum Cum Percent 6.9 24.1 100.0 2.000 Cum Percent 1.1 13.8 25.3 100.0 1.000 APPENDIX C Survey Hi, my name is Jeanette James. I am a graduate student in the Department of Communication at Michigan State University. I am conducting a study on Asian International students learning about the United States through television. The results of this survey are going to be used in my Master's Thesis project. It would be greatly appreciated if you could spare about 15 minutes to complete this questionnaire. The information from this survey is completely confidential. You indicate voluntary agreement to participate by completing and returning this questionnaire. Thanks in advance for your participationll Section I In this first section, please reflect back to the television programming and impressions of this programming that was viewed in your native country: 1. Please mark how often you watched the programs listed below while living in your native country: almost every about once once in rarely/ week a month two months never Three's Company Hawaii Five-O Love Boat Mission Impossible Kojak The A-Team The Cosby Show Charlie's Angles Different Strokes Star Trek Six Million Dollar Man Lassie Dallas Dynasty _____ __ Magnum Fl 89 90 Hotel Quincy Hart to Hart Fantasy Island MacGyver Cheers Jeffersons Happy Days Good Times L.A. Law Roseanne A Different World 21 Jump Street Married With Children 2. Before You came to the United States how many hours did you watch television daily... National/Local programming:_ hours per day American programming :_hours per day Other International programming—hours per day 3. Which American television programs did you think most accurately portrayed American culture and people? a. b. c. 4. In general, how accurately did you think television portrayed what life in the United States would be like? highly very somewhat slightly not at all accurately accurately accurately accurately accurately 5. In general, how accurately did you think television portrayed how Americans behave? highly very somewhat slightly not at all accurately accurately accurately accurately accurately 6. Listed below are twenty adjectives. Before you came to the United States which of these traits did you think were common of the American population in general. Please rate the commonness of the traits on a scale of 0 to 10 ( 0 means not at all common). Write the number in the corresponding blank. Individualistic Courteous Conceited Materialistic Practical Artistic Athletic Argumentative Ambitious Sensual Intelligent Aggressive Straight-forward Mercenery Pleasure-loving Scientific Arrogant Industrious Courteous Effecient___ 7. How many Americans did you know personally: In your country? From vacations outside the U.S.? From brief visits (2 weeks or less) in the U.S.? 8. Which specific impressions of the U.S. from American television turned out to be true? Which did not? PART II In the previous section you were asked to reflect back on your impressions of American culture and people before you moved to the United States. New in this section think of the culture as you know from your own experiences and how television portrays U.S. society. 9. Of the following programs on television this year, including re-runs, list how often you watch them. more than almost every once or rarely/ once a week week twice a month never 92 Roseanne Step By Step Wings Beverly Hills 90210 Full House L.A. Law The Cosby Show Northern Exposure Melrose Place Married with Children Sisters Mad About You Family Matters 10. Which of these shows most accurately depicts the way Americans really behave and live? Please circle your choice. Roseanne Sisters Mad About You Family Matters. Full House Step By Step Wings Beverly Hills 90210 L.A. Law The Cosby Show Northern Exposure Melrose Place Married with Children 11. How many hours of television programming do you watch currently? educational/news: _hours per day entertainment: _hours per day sports: _hours per day other:_hours per day 93 12. How often are the following themes shown in American television programming (entertainment only). Please rate on a scale of 0 to 10 with 10 being the most prevalent. poverty discrimination against blacks— sexual promiscuity_ discrimination against whites— peer pressure__ discrimination against the elderly___ crime__ racial conflicts_ wealth_ environmental negligence teen pregnancy__ negative stereotyping of Whites___ workplace conflicts_ negative stereotyping of Blacks— violence negative stereotyping of Hispanics____ sexual discrimination negative stereotyping of Asians— homelessness negative stereotyping of gays/lesbians— 13. Now that you live in the United States what adjectives do you feel best describe the American population in general? Please rate on a scale of 0 to 10 (ten being the highest). Write the number in the corresponding blank. Individualistic Courteous Conceited Materialistic Practical Artistic Athletic Argumentative Ambitious Sensual Intelligent Aggressive Straight-forward Mercenery Pleasure-loving Scientific Arrogant Industrious Courteous Effecient 14. How many years have you been living in the United States? 15. How comfortable are you when you converse with those who speak only fluent English? highly very somewhat slightly not at all comfortable comfortable comfortable comfortable comfortable 16. How important is it for you to fully assimilate into American society? highly very somewhat slightly not at all 94 important important important important important 17. Do you consider most of the Americans that you know... classmates casual friends intimate friends 18. Do you intend to remain in the United States upon completion of your degree(s)? YES NO 19. What organizations are you involved in (on or off campus)... 20. In your opinion what are the most prevalent problems in the United States today. Please rank order these problems with number 0 being the least prevalent and number 10 being the most prevalent problems in the U.S. poverty__ discrimination against blacks— sexual promiscuity_ discrimination against whites_ peer pressure_ discrimination against the elderly— crime__ racial conflicts_ wealth environmental negligence teen pregnancy_ negative stereotyping of Whites— workplace conflicts_ negative stereotyping of Blacks— violence___ negative stereotyping of Hispanics_ sexual discrimination____ negative stereotyping of Asians— homelessness negative stereotyping of gays/lesbians— 21. What is your: Age Sex Year in school 8. Program of study Native Country REFERENCES LIST OF REFERENCES , Choi, J., & Tamborini, R. (1988) Communication-acculturation and the cultivation hypothesis: A comparative study between two korean communities in the U.S. Howard Journal of Communication, 1, 57-74. Elliott, W.R., & Slater, D. (1980) Exposure, experience and perceived tv reality for adolescents. Journalism Quarterly, §1(3), 409-414. Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1986) Living with television: The dynamics of the cultivation process. In J. Bryant, & D. zillman, (Eds.), Perspectives on Media Effects (pp. 17-40). Hillsdale, NJ: Lawrence Erlbaum. Gudykunst, W.B., & Kim, Y.Y., (1984). Communicating with strangers: An approach to intercultural communication (lst ed.). New York: MacGraw—Hill. Hawkins, R.P., & Pingree, S.(1990). Divergent psychological. processes in constructing social reality from mass media content. In M. Michael, & Signorielli, N (Eds.)., Cultivation Analysis: New Directions in Media Effects Research. (pp. 35-49). Newbury Park, CA: Sage. .«Hirsch, P.M., (1980) The 'scary world' of the nonviewier and other anomalies: A reanalysis of Gerbner et. al's. . findings of cultivation analysis, part I. Communication Research, 1(4), 403-456. Kang, J. G., a Michael, M. (1988). Culture clash: Impact of U.S. television in Korea. Journalism Quarterly, §§(2), 431—438. 95 96 Kim,Young Yun (1977). Inter-ethnic and intra-ethnic communication: A study of Korean immigrants in Chicago. International and Intercultural Communicational Annual, 2, 53-680 Korzenny, F., & Schiff, E. (1992). Media Effects Across Cultures: Challenges and Opportunities. In F. Korzenny, S. Ting-Toomey, a E. Schiff (Eds.). Mass Media Effects Across Cultures. (pp 1-8). Newbury Park, CA: Sage Publications. Messaris, P., & Woo, J. (1991). Image vs. reality in Korean-Americans' responses to mass-mediated depictions of the United States. Critical Studies ip Mass Communication, e, 74-90. ‘,Morgan, M. (1982). Television and adolescents' sex role stereotypes: A longitudinal study. Journal pf Personality and Social Ps chol , g;(5), 947-955. Morgan, M., & Shanahan, J. (1992). Comparative cultivation analysis: Television and adolescents in Argentina and Taiwan. In F. Korzenny, S. Ting-Toomey, & E. Schiff (Eds.). Mass Media Effects Across Cultures. (pp 173- 197). Newbury Park, CA: Sage Publications. Morgan, M., & Signorelli, N. (1990). Cultivation analysis: conceptualization and methodology. In Morgan, M., & Signorielli, N (Eds.).,Cultivation Analysis: Neg: Directions in Media Effects Research. (pp. 13-35). Newbury Park, CA: Sage Publications. Pingree, S., & Hawkins, R. (1981). U.S. programs on Australian television: The cultivation effect. Journal of Communication, ;1(1), 97-105. g Potter, W.J. (1986). Perceived reality and the cultivation hypothesis. Journal of Broadcasting and Electronic Media, ;Q(2), 159-174. Reeves, B. (1978). Perceived reality as a predictor of children's social behavior. Journalism Quarterly, 5;, 682-689,695. 97 Tamborini, R., & Choi, J. (1990). The Role of Cultural Diversity in Cultivation Research. In Morgan, M., & Signorielli, N (Eds.).,Cultivation Analveie: Neg Directione in Media Effecpe Research. (pp. 157-179). Newbury Park, CA: Sage Publications. Tan, A.S., Li, S., & Simpson, C. (1985). American TV and social stereotypes of Americans in Taiwan and Mexico. Journalism Quarterly, §;(4), 809-814. Tan, A.S., Tan, G.K., & Tan, Alma S. (1987). American TV in the Philippines: A test of cultural impact. Journalism Quarterly, §g(3), 65-72,144. Varis, T. (1984). The international flow of television programs. Journal pf Communication, ;g(1) 143-152. WOber,J.M. (1978). Televised violence and paranoid perception: The view from Great Britan. Public Opinion Quarterly, g;(3), 315-321. "Illllllllliiilllllilll“