台灣人怎麼看非洲? 台灣大學生的刻板印象認可程度之探索式研究 - 政大學術集成

全文

(1)International Master’s Program in International Communication Studies College of Communication National Chengchi University. Master’s 治 Thesis. 立. 政. 大. ‧. ‧ 國. 學. io. sit. y. Nat. er. Taiwan’s Eye on Africa: An Exploratory Study of Stereotype. al. n. iv n C U Endorsement AmonghTaiwanese Students e n g c h i University. Student: Barbara Moi Advisor: Professor Kuo Cheng. 103 August 2014. 08.

(2) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v.

(3) Acknowledgements Finally, the journey of a thousand steps comes to an end-it has been a long, long road full of unexpected difficulties and challenges but I have reached the peak of my own Mt. Everest. I have grown so much, learned patience and built character. This research is made possible through the support of so many people. First, I would like to thank my thesis advisor, Professor Kuo Cheng. Her patient guidance has been truly helpful marvellous throughout this process. I would also like to thank my thesis committee members Professor Shih and Professor Weng, whose helpful suggestions helped bring this thesis to successful completion. Big thanks to my IMICS. 政治大. classmates for translating and Ian Wang for guiding me through SPSS like a champion.. 立. My heartfelt gratitude goes to my dear Tanga Clan: Christine Fan and Sebastian. ‧ 國. 學. Cabrera Leon. You encouraged me so much to achieve all my dreams, and were with me. ‧. every step of the way. The two of you are the yams in my soup, the llamas in my Inca, the. sit. y. Nat. Okonkwos in my village. This is not the end!. io. er. I would like to dedicate this thesis to my mother, Sylvia Jerotich Moi for being an. al. epitome of grace and support throughout this journey. I would have cracked several times. n. iv n C under the pressure but she always patched h emen up i Ume strength. I love you, Mom! g candhgave Finally I must mention the people whose loving care both in Taiwan and Kenya have been felt and appreciated throughout my long journey: the whole of TPB (especially my roomie, Stephanie Kasaon), Auntie Sarah and Uncle Joseph Kasaon, Fiona Odhiambo, Santhi Leveque, Melanie Ferrand and Christine Orchard. Next stop: world domination. Asanteni sana!. i.

(4) Abstract University students are some of the more cosmopolitan groups in society, as voracious consumers of media and technology, especially new media, and exposed to a plethora of information about the global community. University courses, interaction with the international student communities, events on-campus and opportunities to travel abroad are some of the avenues available to them. Using quantitative analysis, this study examined the relationships between demographic and contact variables, Western media exposure, cosmopolitanism and perceived bias of media portrayals and its effect on Taiwanese university students’ endorsement of stereotypes of Africa. The research framework is. 治政 informed by the theory of social construction of reality and大 the cultivation theory. Using an 立 online survey, data was collected and analyzed from a sample of 215 Taiwanese university. ‧ 國. 學. students. The analysis reveals that contact with Africa and perceived bias of media portrayals. Nat. sit. y. ‧. have the most significant influence on stereotype endorsement.. io. al. er. Keywords: stereotype endorsement, Africa, Taiwan, university students, cultivation theory,. n. theory of social construction of reality, Western media exposure, cosmopolitanism, contact, perceived bias. Ch. engchi. ii. i Un. v.

(5) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. iii. i Un. v.

(6) III. Table Of Contents I.. Acknowledgement…………………………………………………………...... i. II.. Abstract………………………………………………………………………... ii. III.. Chinese Abstract. iii. IV.. Table of contents………………………………………………………………. iv. V.. List of Tables…………………………………………………………………... vi. VI.. List of Figures…………………………………………………………………. vii. VII.. List of Appendices…………………………………………………………….. viii. 1. INTRODUCTION…………………………………………………………………. 1. 1.1 Purpose of Study………………………………………………………………. 3. 1.2 Significance of Study………………………………………………………….. 3. 政治大 2.1 Stereotypes……………………………………………………………………. 立. 6. Contact………………………………………………………………... 8. Stereotypes and the Media……………………………………………. 8. Stereotypes of Africa in Global Media……………………………….. 9. y. 5. Social Cultural Approach to Studying Stereotypes…………………... 2.2 Media Literacy……………………………………………………………….. 17 18. 2.1.4. ‧. 2.1.3. ‧ 國. 2.1.2. 學. 2.1.1. 5. sit. 2. LITERATURE REVIEW………………………………………………………….. Nat. 2.3 Cultural Protectionism in Taiwan…………………………………………….. io. er. 2.4 Traditional Media Vs. Internet………………………………………………. n. al iv 2.5 Cosmopolitanism……………………………………………………………. n Ch engchi U 2.6 Theory……………………………………………………………………….. 18 20 22. 2.6.1. Theory of Social Construction of Reality……………………………. 22. 2.6.2. Cultivation Theory……………………………………………………. 23. 3. METHODOLOGY………………………………………………………………... 25. 3.1 Method Selection Rationale……………………………………………………. 25. 3.2 Research Design……………………………………………………………….. 26. 3.3 Data Collection………………………………………………………………... 27. iv.

(7) 3.4 Pilot Study…………………………………………………………………….. 28. 3.5 Measurement of Variables…………………………………………………….. 28. 3.5.1. Demographics…………………………………………………………. 28. 3.5.2. Contact………………………………………………………………... 28. 3.5.3. Cosmopolitanism……………………………………………………... 29. 3.5.4. Perceived Bias of portrayals…………………………………………... 29. 3.5.5. Western Media Exposure……………………………………………... 30. 3.5.6. Stereotype Endorsement………………………………………………. 33. 3.6 Research Framework & Hypotheses………………………………………….. 35. 4. RESULTS………………………………………………………………………... 38 38. 4.1.2. 政治大 Demographics…………………………………………………………. 立 Contact………………………………………………………………... 4.1.3. Cosmopolitanism………………………………………….................... 39. Perceived Bias of portrayals………………………………………….... 40. Western Media Exposure…………………………………………….... 40. Stereotype Endorsement……………………………………………….. 42. 4.1 Descriptive Analysis…………………………………………………………... 4.1.6. ‧. 4.1.5. 學. 4.1.4. ‧ 國. 4.1.1. Nat. sit. y. 4.2 Hypothesis Testing……………………………………………………………. 4.2.2. Hypothesis 2………………………………………………………….... er. Hypothesis 1………………………………………………………….... io. 4.2.1. al. 38 39. 43 44. 4.2.5. Hypothesis 5………………………………………………………….... 46. 4.2.6. Hypothesis 6………………………………………………………….... 46. 4.2.7. Hypothesis 7………………………………………………………….... 46. 4.2.8. Research Question 1………………………………………………….... 51. 5. DISCUSSION……………………………………………………………………... 52. 4.2.3. n. 44. 4.2.4. iv n Hypothesis 3…………………………………………………………... Ch U i e h n c g Hypothesis 4………………………………………………………….... 45 45. 5.1. Limitations…………………………………………………………….... 55. 5.2. Implications……………………………………………………………... 55. 6. REFERENCES……………………………………………………………….......... 56. 7. APPENDICES………………………………………………………………........... 64. v.

(8) IV. List Of Tables Table 1: Stereotypes of Africa, Descriptions and Sources……………………………. 15. Table 2: Western Media Exposure Factor Loadings and Reliability Results……….... 32. Table 3: Stereotype Endorsement Factor Loadings and Reliability Results………….. 34. Table 4: Description Statistics for Demographic Variables…………………………... 38. Table 5: Mean values of Demographic, Contact, Cosmopolitanism, Perceived Bias of Portrayals and Western Media Exposure Variables……………………………….. 41. Table 6: Mean Values and Endorsement Level for Stereotype Endorsement.……….. 43. 政治大. Table 7: ANOVA Table for Monthly Expenditures and Western Media Exposure….. 立. Table 8: Hierarchical Regression Analysis of Primitive Africa………………………. ‧ 國. 學. Table 9: Hierarchical Regression Analysis of Socioeconomic Status………………... ‧. Table 10: Hierarchical Regression Analysis of Environment……………………….... y. Nat. 47 48 49 50. er. io. sit. Table 11: Pearson’s Correlation Matrix with all the variables……………………….. 44. Table 12: Multiple Regression Analysis of Monthly Expenditure, Contact,. n. al. Ch. i Un. v. Cosmopolitanism, Western Media Exposure and Perceived Bias on Stereotype. engchi. Endorsement………………………………………………………………………....... 51. Table 13: Results from Addressing the Research Question and Hypotheses……….... 52. vi.

(9) V. List of Figures Figure 1: News report of ‘Dark’ Africa (Duodu, 2000)…………………………………. 12. Figure 2: Research Framework……………………………………………………………. 35. Figure 3: Frequency of contact with Africa……………………………………………….. 39. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. vii. i Un. v.

(10) VI. List of Appendices Appendix A: Variables, Conceptual Definitions and Operational Measures…………….... 64. Appendix B: Chinese Version of Survey Questionnaire……………………………........... 66. Appendix C: English Version of Survey Questionnaire…………………………………... 70. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. viii. i Un. v.

(11) 1. INTRODUCTION Due to globalization, the multiplication of media technology and services means that there is a vast increase in the exchanges between all global communities (Regis, 2009). Cultural products brought through mass media form a part of “daily discourse…creating and sustaining areas of shared experience, symbolism, knowledge and understanding” (BoydBarrett &Braham, 1987, p.3). Contact with the ‘cultural other’ and willingness to learn and engage with diverse cultures are some of the fundamental points of cosmopolitanism as a personal orientation (Roudometof, 2005; Norris & Inglehart, 2009; Jeffres et al., 2002).. 政治大 been colonized several times in立 the past by European and Asian governments. According to. Taiwan is a small island nation with various languages and cultures within it that has. ‧ 國. 學. Harding (1993), Taiwan is a part of Greater China, which refers to the economic and cultural connection as well as the possible political reunification between Taiwan, Hong Kong and. ‧. China (Sinclair Jacka & Cunningham, 1996, p.126). Although Taiwan’s (R.O.C) legitimacy. Nat. sit. y. as a nation has been disabled severely by the People’s Republic of China (P.R.C), Taiwanese. n. al. er. io. society continues to construct a political, cultural and ideological identity of itself separate. i Un. v. from that of the latter ever since its separation in 1949 (Kang & Yang, 2011). After 1960,. Ch. engchi. Taiwan underwent very fast industrialization and economic growth to become a welldeveloped capitalist economy with a thriving consumer society (Lewis, Martin & Sun, 2012). In this way, Taiwan can be seen as a cosmopolitan country, because of its high media access and economic development. The present research is interested to find out how this high media access but low contact exposure will affect stereotype endorsement in the case of Africa. Dixon (2002) reports that stereotypes may lead to inaccurate representations as it affects the encoding and processing of information..

(12) The news and entertainment media is the primary source of images of people of other cultures that depicts their behavior and characteristics. These representations can create perceptions and belief in stereotypes (Sanders & Ramasubramanian, 2012). Using the theory of the social construction of reality, the definition of ‘reality’ or ‘knowledge’ is brought into question because of the social relativity of its nature. As Berger and Luckmann (1966, p15) emphasize, “what is ‘real’ to a Tibetan monk may not be ‘real’ to an American business man.” In the same vein, the present study seeks to discover what relationships stereotype endorsement have with contact with the ‘cultural other’, in this case Africa, cosmopolitanism as a personal orientation, perceived bias of media portrayaks and Western media exposure.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 2. i Un. v.

(13) 1.1 Purpose Of Study Following the steps of stereotype research tradition, the present study will explore stereotype endorsement of a cluster of media-activated stereotypes of Africa as a result of the interplay between Western media exposure, contact and cosmopolitanism, while also examining the influence of perceived bias of media portrayals of Africa and demographic variables (Yunying &Tan, 2011). In the process of determining these relationships, the study will also provide a description of these variables. The main research question that this study seeks to answer is what is the best set of predictor variables among the independent variables. 政治大. that determine stereotype endorsement.. 立. ‧ 國. 學. 1.2 Significance of Study. By offering a snapshot of the relationships that take part in accepting these. ‧. stereotypes from a Taiwanese viewpoint, this study will contribute to the existing body of. y. Nat. sit. knowledge; understanding the dynamic process that shapes perceptions of other groups is an. er. io. inquiry worthy of merit (Lee et al., 2009).. al. n. iv n C To the author’s knowledge, little hresearch i Udone on this topic with Africa as e n g chashbeen its focal point. The geographical, lingual and cultural differences between Taiwan and Africa are massive, therefore making this study’s contribution as unique as it is challenging. The scientific community therefore recognizes the importance of applying these theories and concepts to other global regions that have different cultures (Rybina et al., 2010). In addition, this study has implications for intercultural as well as international relationships between Asia and Africa whereby stereotypes enter into political discourse and policy-making decisions, (Ibroscheva & Ramaprasad, 2008, Cribbs & Austin, 2011).. 3.

(14) The Africa Taiwan Economic Forum (ATEF), which comprises foreign embassies and economic offices credited to Taiwan, is allied with the Ministry of Foreign Affairs (MOFA) and Ministry of Education (MOE) in Taiwan in search of common goals. In addition to advancing trade and investments, one of its main priorities is cultivating a more favorable image of Taiwan through cultural and academic exchanges. The present research is of particular significance to these efforts (About ATEF, n.d).. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 4. i Un. v.

(15) 2. LITERATURE REVIEW 2.1 Stereotypes A stereotype is not necessarily negative; there are also positive stereotypes, a dimension that is usually ignored. Ashmore & Del Boca (1981) put forth that stereotypes are general, dominant beliefs about another group’s attributes that are descriptive and generic by nature and can be either positive or negative. Gorham (1999) states that the media play an important role through frequent exposure of the associations to these attributes (Lee et al., 2009).. 政治大. Tan, Li & Simpson (1986) emphasize that “when there is consensus in the assignment. 立. of traits within a given population of judges, then the stereotype becomes a social norm for. ‧ 國. 學. describing recognized groups.” With no such consensus, an individual's assignment of traits. ‧. to the group is his or her personal stereotype, “regardless of whether there is consensus or not. y. Nat. among other judges.” Therefore, conceptually, social stereotypes are generalized impressions. members of a category (Tan, Li & Simpson, 1986, p810).. n. al. Ch. engchi. er. io. sit. of groups while operationally a social stereotype is the collection of traits assigned to the. i Un. v. Studies have shown that stereotypes form in part because of a limited exposure and experience. Tan et al.’s (1997) study shows that limited contact, leads the one to observe other groups through the media and form opinions thereafter (as seen in Lee et al., 2009). Stereotypes are more than just the social judgments we make about a group; they also influence real world beliefs, intergroup emotions, causal interpretations and support towards out-groups (Ramasubramanian, 2011). To illustrate this concept, let’s consider African Americans, a group who suffer from their fair share of negative stereotypes. Most media portrayals of African Americans have been of a stereotypical nature; either violent, demeaning, and more often criminals rather than victims (Mastro et al., 2009). Tan and 5.

(16) colleagues’"(2000) research found that negative mediated stereotypes of African Americans positively correlated with negative social judgments and decreased support for affirmative action (Ramasubramanian, 2011). Mobile growth in Africa is a good example of the negative side effect of stereotypes accepted as reality. Much of the West refused to invest in the mobile phone industry because they thought that ‘this being Africa’ was bound to fail. It is now a multi-billion dollar industry built by and for Africans (Williams, 2009, p12). Stereotypes serve both cognitive and social functions by simplifying incoming. 政治大. information and solidifying in and out group identities, respectively (Yunying & Tan, 2011).. 立. They justify social inequalities, promote superiority amongst the in-group, and maintain. ‧ 國. 學. predictability in the social environment (Moreno & Bodenhausen, 1999). People naturally. ‧. therefore, seek to protect their stereotypes. It would be a very disorienting prospect to have to constantly change our social schemas in order to accommodate new information.. y. Nat. er. io. al. sit. 2.1.1 Socio-Cultural Approach to Stereotypes. v. n. The socio-cultural approach recognizes that stereotypes result from group-level. Ch. engchi. i Un. interactions as well as from individual level interactions. Consequently, when studying stereotypes, it shifts its focus from the intrapersonal onto intergroup dynamics. It flows naturally that when discussing group-level perceptions, the role that the mass media plays as a disseminator of information is at the forefront of investigation (Ibroscheva & Ramaprasad, 2008). Leyens et al (1994) named this emphasis on the society as the socio-cultural approach. A number of significant studies on stereotypes have investigated the cognitive aspects of stereotype formation, whereas sociocultural studies concentrate more on the symbolic environment within a given cultural system, and how in this context stereotypes 6.

(17) evolve and are transmitted (Ibroscheva & Ramaprasad, 2008). In contrast, the social cognitive approach defines stereotypes as cognitive structures, like social schemas, that affect the encoding and processing of information about out-groups (Ramasubramanian, 2011). Using a sociocultural approach to studying stereotypes, it is crucial to understand that people live in groups where part of the socialization process that provides them with group identity is to form and share beliefs with each other. A society is defined as a collective of people within large, stable social systems who share a common identity, including traditions, cultures, collective memories who occupy a legitimate territory. Society members construct. 政治大. shared beliefs (Tal & Teichman, 2005).. 立. Haslam et al (1998) placed shared stereotypes as a result of group identification that. ‧ 國. 學. leads members to accept group beliefs, norms and attitudes. Research on shared stereotypes. ‧. has demonstrated that these beliefs are usually transmitted down and across generations through group communication channels. If learned at an early age, these stereotypes affect. y. Nat. er. io. sit. behavior and are hard to get rid off (Tal & Teichman, 2005). Adolescence is a critical time when youths draw conceptions of self and others from their peers, parents and society as a. n. al. Ch. i Un. v. whole. Media use is also very high during this period, up to 7 and a half hours a day (Ward, 2004).. engchi. The framework adopted for this study recognizes these communication channels, specifically mass media and education, as social mechanisms influencing stereotypes within a micro-social environment. In addition to media exposure and education, friends and family within this environment are significant in that they pass on information from the above mentioned channels to each other via their relationships (Tal & Teichman, 2005). 7.

(18) 2.1.2 Contact In Hale’s (1998) study on the effects of age, contact and knowledge on stereotype the elderly, her results indicated that stereotyping was dependent on the level of contact. There were no significant age differences in stereotyping scores. Hale’s research further reports that stereotyping was mostly dependent on the contact level. Those with higher contact had higher knowledge and lower stereotype scores. Ademeyemi’s (2011) study sought to find out, among other things, the sources of stereotypes that high school students in Botswana had of outside nationalities. The findings showed that stereotypes were derived from internet, stories, television programmes, instructional materials particularly prescribed textbooks,. 政治大. comments from friends, newspapers, radio programmes, classroom interaction with teachers. 立. and pamphlets and newsletters.. ‧ 國. 學. Fujioka’s (1999) study investigates the effects of television portrayals of. ‧. African-Americans on Caucasians (who have direct contact) and Japanese students (who. sit. y. Nat. have little to no contact). The results proved that mediated messages had more influence. io. er. especially when contact was lacking (Lee et al., 2009). In this study therefore, the. al. information source about Africa through family, friends, university courses and media will be. n. iv n C examined for their part in shaping stereotypes h e n endorsement g c h i Uof Africa for the Taiwanese students 2.1.3 Stereotypes And Media. The U.S has massive influence on global information and entertainment networks. In the global audiovisual exchange of goods and services, the United States is the top exporter (Norris & Inglehart, 2009). This influence spans book publishing, news agencies, international newspapers and magazines, radio and television channels, music, advertising. 8.

(19) and films. It is so much so that the communications flow between other continents is mediated by American news organizations to a large extent (Thussu, 2000, p.163). In early media effects research, Joseph Klapper (1960, p.8) asserted that “mass media does not ordinarily serve as a necessary and sufficient cause of audience effects, but rather functions through a nexus of mediating factors” but rather it operates within a pre-existing sociocultural structure (Boyd-Barrett & Braham, 1987). The media’s influence on stereotypes has been documented in countries outside the United States. For example, Tan et al (1997) found that negative portrayals lead to negative. 政治大. perception of Native Americans who found the portrayals believable; Hewes (2005). 立. demonstrated the effect of parasocial contact which proved that prejudice decrease with. ‧ 國. 學. mediated interaction with media characters of outgroups, such as transvestites. Processing of new information is the key factor to stereotype change (Yunying & Tan, 2011).. ‧. y. Nat. New media has potential for shedding more light on the continent’s media coverage,. er. io. sit. so to speak, by providing more opportunities for representing Africa for the world to see. In 2008, the African Studies Centre Library in Leiden, reported that approximately 2,500 e-. al. n. iv n C journals with articles about Africa, were However, hpublished. i U it was also reported at that time engch that much of online content is produced by non-Africans outside the continent. Melissa Wall’s article “Africa on YouTube” found that most of the content on Kenya and Ghana was either musical or tourist. Furthermore, the sources imposed their own meanings onto the videos and had themselves at the center (Williams, 2009). 2.1.4 Stereotypes of Africa In Global Media In May 2000, Jean-Louis Sarbib, vice president of the Middle East and North Africa international lending agency, said at the international economic conference on Africa that “one fifth of Africans live in a world that is at war with itself or its neighbor.”"That meant 9.

(20) that even though 80% of the continent lives in peace, it’s the unstable 20% that is used to define its image as a whole. Azziz Bahad, former deputy minister of foreign affairs in South Africa asserted that the negative disaster-focused coverage by international media giants like CNN and BBC are bringing back old racial interpretations of Africa (“World Bank official says Western world misinformed about Africa,”"2000). Economic journalist and editor of the London-based Africa Business magazine, Anner Versi, affirms that albeit Africa is low on the priority list, when it does get media coverage in the West it continually consists only of negative news, portraying corruption,. 政治大 Africa is heavily impacted by negative portrayal of Africa by Western media (Duodu, 2000). 立. wars, famine and disease (Versi, 2004). According the World Bank, foreign aid investment in. ‧ 國. 學. ‘In the West’, claims Molefi Kete Asante in his recent History of Africa, ‘the ignorance of Africa is palpable, like a monster that invades our brains with disbelief,. ‧. deception, and disinterest, yet is everywhere around us. We are victims of probably the most. y. Nat. io. sit. uninformed educated people in the world on the subject of Africa"(Williams, 2009, p7). So. n. al. er. when it comes to media coverage of Africa, the most critical issue is that of invisibility: a lot. Ch. i Un. v. of important stories go unreported by Western press. Digestible, predictable stories of war,. engchi. famine and disease are chosen over African history, culture and values. Stories are told only of colonization and westernization point of view, ignoring the historical parts that happen before then (Hawk, 1992). Alan Gelb, World Bank African chief economist, said economic growth and reconstruction programs are often ignored in favor of reports on Aids, conflict and epidemics. Notwithstanding, the World Bank itself also lends to the problem of misreporting of Africa. With a surface area of 11.5 million square miles, Africa is the second largest continent with 54 countries, yet the World Bank treats it as one region. This is reflected in their figures and. 10.

(21) reports, which at best summarize the findings for Sub-Saharan Africa in one part and North Africa in the other. Countries with totally different GNDS per capita levels and sociopolitical traditions are grouped together to summarize figures and make points about social disintegration respectively (Duodu, 2000). Economic journalist and editor of the London-based Africa Business magazine, Anner Versi, affirms that albeit Africa is low on the priority list, when it does get media coverage in the West it continually consists only of negative news, portraying corruption, wars, famine and disease. All of Africa is treated as one “big sorry mass”"such that if conflict breaks out. 政治大 programming of a “Hopeless Continent”"is so in-depth that it causes junior and senior 立. in one country, headlines read something along the lines of “Africa returns to barbarity.”"The. ‧ 國. 學. journalists who are assigned to cover stories in Africa to have confirmation bias. This bias is so deep in the Western identity and subconscious that it is hard for them to see. Murdock. ‧. (1973) asserts that news reports are often in line with what the opinions of those in power. Nat. sit. y. (Siu, 2009). According to Versi, in order to change the Western media’s representation of. n. al. er. io. Africa, we have to change what the journalists want to see (Versi, 2004).. Ch. i Un. v. Sorius Samura, a native of Sierra Leone, is one such television journalist who. engchi. specializes in immersing himself in documentaries, going further than other journalists would to uncover the untold stories unfolding in Africa. His documentary “Living with Aids”, which cites male promiscuity as a major cause of the epidemic, awarded him Broadcaster of the Year in 2006 at the One Media Awards in London, England. He believes the white, liberal elite in the newsrooms are simply too scared and riddled with ‘post-colonial guilt’"to face Africa square on, and in this way they have “failed Africa.”"(Smith, 2006) The themes of failure, famine, disease and coups in ‘Godforsaken countries’ are dominant in literature by ‘old African hands’, so much so that a young Barack Obama. 11.

(22) travelling to Kenya was angered by it, simply because it was without a clear target. The issue does not lie in accuracy, because Africa does have the lion’s share of challenges, but as Richard Dowden, director of the Royal African Society states, rather the generalization of a vast continent, that has more than 2000 cultures and languages as one country full of hapless victims (Williams, 2009). Figure 1 below shows an example of a news clipping reporting a conflict that happens in Sierra Leone.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i Un. v. Figure 1. News report of ‘Dark’ Africa (Duodu, 2000) Chinua Achebe, a Nigerian writer, argues that the massive amounts of derogatory images of Africa in the eyes of the world is built up by Western literature and that is still there today, was originally a means to defend the slave trade and colonialism (Williams, 2009). Ukadike (1990) agrees with Achebe, stating that the exaggeration of these images that are negative and denote savagery serve to justify European rule and Christian missionaries. 12.

(23) ‘involvement’ with Africa (Ramasubramanian, 2005). Today Africa is seen as an exotic place to travel and see animals, Achebe says, and not as a continent full of people (Williams, 2009). A study done by Osunde & Tlou (1996) examined the stereotypes and misconceptions held by Social Studies teachers in American public schools and found the same stereotypes of wild animals, malnutrition, disease, huts, tribes, elephants, jungles, poor, deserts, villages, tigers, natives, superstition. Exactly the same as those found by Beyer and Hicks (1968), which surveyed 3259 students, more than 3 decades ago.. 政治大 two years (2006-8), dispute the theme of fear that has been so connected with Congo in many 立 Fred Robarts, a human security consultant and writer who lived in D.R. Congo for. ‧ 國. 學. books. He says that while Congo may not have so many decent cinemas or cafes, it has less gun and street crime than London (Williams, 2009).. ‧. McCall Smith is the author of a book series based in Botswana, The No.1 Ladies. y. Nat. er. io. sit. Detective Agency that is now a television show starring Jill Scott. It is one of the few media portrayals of Africa that goes beyond the stereotypes into fuller representations. He had this. n. al. to say in an interview:. Ch. engchi. i Un. v. McCall Smith: It’s not just my reaction to the country. People say what a wonderful place. People say that similarly about many sub-Saharan African countries. Without romanticizing the place, people there have great human qualities which really are very very striking. When you see the picture that’s presented of Africa, of sub-Saharan Africa, you don’t see that. I think this is a terrible pity. Interviewer: They have diamonds. They have the Kalahari. They suffer from the AIDS epidemic. That’s pretty much the summary you read in the press. 13.

(24) McCall Smith: There’s that side of it, too, in sub-Saharan Africa. But there are human qualities which are wonderful. Very interesting cultures with a lot to say. Yet all we get is a picture of disasters (Williams, 2009, p11). The literature reveals six major themes that describe major stereotypes about Africa from global media: One Country, Dark Africa, Victim, Conflict, The African and Wild & Savage Africa. These themes are presented with their conceptual descriptions and sources in Table 1:. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 14. i Un. v.

(25) Table 1 Stereotypes of Africa, Descriptions and Sources Stereotypes of Africa. Description. Sources. One Country. The problems reported in any parts of the continent is generalized to all of Africa. Hawk (1992); Williams (2009). Countries with different GNDPs are grouped together to summarize findings. Duodu (2000). If conflict breaks out in one country, it is reported along the lines of “Africa breaks Versi (2004) out in barbarity.”. 立. Hawk (1992). 學. In reference to both the color and the perceived ignorance of the people. Primitive and Uncivilised.. Hut, dark, tribalism, war, famine, disease, animism, primitivism, poverty, third world, developing. (1998). y. Nat. “When Hollywood Goes to Africa” (n.d.). Portrayal as a failure, in need of salvation from the white man.. Hawk (1992); Wyk (2007). Compassion Usury: ceaseless broadcasting of victim images and emotions to pull at the heart strings of countries to donate time, money, goods and services to ‘poor’ Africa.. Susan Moeller (1992) as seen in Wyk (2007). n. al. er. sit. The Ghost and The Darkness (1996): based on a true story in 1898 about maneating lions but depicting Kenyans as being ignorant as their predecessors 100 years ago. io. Victim. Hawk (1992); Chavis. ‧. ‧ 國. ‘Dark’ Africa. 政治大. Ch. engchi. i Un. v. AIDs, epidemics, disease, famine, flies in food and face, stomachs distended. 15. Chavis (1998).

(26) Conflict. The unstable 20% is used to define the image of all Africa. “World Bank official says Western world misinformed about Africa”"(2000). Focus on negative disaster-based reporting. Williams (2009); Hawk (1992);Versi (2004); Chavis (1998). Fear, danger, war, corruption. “When Hollywood Goes to Africa” (n.d.). Blood Diamond (2006): based on a real conflict, highlighting black conflict, greed and violence, with a white lead. The African. Defined as a black person living on the continent, according to former colonial labels. Therefore, a white South African or an Arab-looking North African are not considered ‘African’. 立. (Hawk 1992). 政治大. ‧ 國. 學. Different skin colour, mentality, culture Superstitious, Primitive, Irrational. ‧. A place where tourists go to see animals, not people.. Williams (2009). y. Nat. sit. io. Adventure, savagery, jungle, safari, Hawk (1992); Chavis natural wonders, big game hunting, King (1998) Kong, hut. er. Wildlife & Savage Africa. al. n. iv n C Tarzan (1999): h adventure, exoticUAfrica, with no people e n g c h i. 16. “When Hollywood Goes to Africa” (n.d.).

(27) 2.2 Media Literacy Ms. Chen (pseudonym), a program buyer at PTS (interview, June 2010, Taipei) had this to say about the Taiwanese audience: I don’t know why but Taiwan audiences seem to feel that is something is ‘educational’ then it must be boring…Entertainment programs are made purely for entertainment value. And once you call a program ‘educational’ well, you basically lose half of your audience (as seen in Lewis, Martin & Sun, 2012, p558).. 政治大 use to expose ourselves to the media to interpret the meaning of the messages we encounter’’ 立 Potter (2008, p.19) defined media literacy as ‘‘a set of perspectives that we actively. ‧ 國. 學. (Kean et al, 2012, p. 204). Media consumers who can be more critical, question source and content and compare the messages with their own existing knowledge are considered to have. ‧. some level of media literacy (Kean et al, 2012). When the audience consumes media. sit. y. Nat. information attentively with more involvement, then stereotypes are less likely to be accepted. n. al. er. io. as social reality and thus bias opinions (Schemer &Wirth, 2009).. Ch. i Un. v. However, Potter (2005, p.10) adds that there are a few dynamics influencing media. engchi. literacy. The abundance of media messages and channels means that consumers are more willing to accept media messages as ‘faulty beliefs’ without questioning them. Another way consumers try to control the overload is by sticking to the same channels and programs. This severely limits their sources and consequently their media literacy. A media literate person is one who is able to control the media, not the other way around (Kean et al, 2012). One story can have very many different impressions depending on the choice of words, selective omission and questionable credibility of sources. This slanting of information is known as media bias (Gentzkow & Shapiro, 2006). In their research Gentzkow. 17.

(28) & Shapiro (2006) found that people tend to expect news to in line with their prior beliefs. Therefore, if offered information that is inconsistent with these beliefs, one is more likely to disregard or mistrust the information source. This also known as the ‘hostile media effect’. Banjo (2008) posits that any one with strong commitment or opinions about a certain group or issue are more likely to have perceptions of bias. Two more processes ‘the theory of presumed influence’ and ‘third person perception’ show that people believe others to be more susceptible to media coverage than they are, therefore they are more likely to have a perception of bias. (McKeever, Riffe & Carpentier, 2012; Banjo 2008).. 政治大. 2.3 Cultural Protectionism in Taiwan. 立. Taiwan’s percentage of imported programming reduced from 24% in 1990 to 20% in. ‧ 國. 學. 1994. This 4% drop may imply an increase in domestic production of programming. More. ‧. specifically, in the 1990s Taiwan showed 72% of its own programming on television, 20% of U.S programming and 8% of Japan’s (Junhao, 1998). Foreign programs are less than 20% of. y. Nat. er. io. sit. Taiwan’s television programming due to ‘cultural protectionism’ (Sinclair, Jacka & Cunningham, 1996, p.131) For example, there are controls that limit free exchanges of. al. n. iv n C television programs between Taiwan and world (p. 146). Market forces govern h ethenoutside gchi U. television in Taiwan, but content is highly politicized. There was extreme regulation 1947-87, under the Chiang family (KMT). Now there is no government censorship, after the lifting of the martial law (Lewis, Martin & Sun, 2012). 2.4 Traditional Media VS Internet Kay and Goldberg (1999) categorized the Internet as a ‘meta-medium’ more than a mass media because of its capability for one-to-one, one-to-many and many-to-many communications (Jensen & Helles, 2011). The Internet has taken away much of broadcasters power, which dominated audiovisual media in distribution of cultural forms in the 1980s, as 18.

(29) millions upon millions turn to sites like Google for their media needs (Jensen & Helles, 2011; Malin, 2011). Traditional broadcasting and the Internet have different opportunities to attract audiences, and therefore one cannot say that the older media is out of the game quite yet. The openness of the Internet allows a diversity of perspectives but coincides with a decrease in focus on any one topic. Age, education, gender and income are demographic variables that predict Internet use (Lindstrom, 1997). On the other hand, traditional broadcasting allows for sharp focus that leads to discussion and debate, but greatly decreases and excludes a vast range of options and differing points of view (Malin, 2011).. 政治大 (Nguyen & Western, 2006; Robinson 立 et al, 2000; Stempel et al, 2000) found that there is a. While most studies reflect competition between Internet and traditional media, others. ‧ 國. 學. positive association between use of T.V and Internet for news and information that is complimentary and supplementary. This coincides with Wright’s (1986) ‘all-or nothing’. ‧. model that “those who are heavy users of one media are also likely to use other media fairly. Nat. sit. y. regularly…” (as seen in Tai-Quan & Zhu, 2011, p. 570). Traditional broadcast channels. n. al. er. io. synergize by directing traffic to their website. Chris Paterson (2005) found that 85% of online. i Un. v. news from Yahoo, Lycos, Excite and AOL is taken almost word-for-word from traditional news sources (Malin, 2011).. Ch. engchi. Internet World Stats for the year 2012 records that more than 2.4 billion people have Internet access worldwide, and 44.8% of them are in Asia. Taiwan ranks fourth in the Asian continent, behind South Korea, Brunei Darussalam and Japan in terms of Internet penetration with a 75.4% penetration of population. Although China, which does not include Taiwan, Hong Kong and Macau, has a 40% penetration rate, it still accounts for 50% of all Internet users in Asia (Asia Internet Usage Stats Facebook and Population Statistics, n.d.) Taiwan ranks second in the Asia Pacific region behind South Korea, with an Internet 19.

(30) adoption rate of 79.2%. Those with a bachelor or college degree and higher are the largest group of Internet users accounting for 50.7% of all users in Taiwan. Students, as opposed to non-students, make up 79.4% of all users and a great majority of them trust the Internet. The Internet uses reported were getting information, communication via text and leisure (APIRA, 2013). In Chou’s (2001) study, more than 95% of Taiwanese college students read electronic news online. Some of the places popular among Taiwanese college students include Bulletin Board Systems (BBS) where participants are able to form discussion groups and derive a sense of social support and belonging. Tanet was Taiwan’s first Internet infrastructure that provided campuses with easy Internet access (Chou, 2001).. 立. 2.5 Cosmopolitanism. 政治大. ‧ 國. 學. The world ‘cosmopolitan’ comes from the Greek word kosmopolitès, which means a. ‧. “citizen of the world” (Kleingeld, 2014). The concept of cosmopolitanism has broad applications: it can refer to cities with low internal and external barriers to accessing global. y. Nat. er. io. sit. information (Norris & Inglehart, 2009) as well as to a personality type whereby one possesses the cultural competency to negotiate divergent cultures with openness and mastery (Hannerz,. al. n. iv n C 1990). In the present study, cosmopolitanism be considered as the latter, a personality h e nwill gchi U variable that may influence stereotype endorsement (Roudometof, 2005). Although it has. cultural protectionism, in this sense, Taiwan can be counted a cosmopolitan country because of its high information access and use of media (Chou, 2001). According to Hannerz (1990), there is a distinction between a ‘typical tourist’ and a cosmopolitan, although the two go hand in hand. Although mobility is an important aspect of being a cosmopolitan, beyond that ‘the willingness to engage with the other’ is crucial. Therefore, going on a short beach holiday without engaging the target culture, making or. 20.

(31) receiving an impact from that culture in any way qualifies as a ‘typical tourist’ and not a cosmopolitan (Gunesch, 2004). Cosmopolitanism reflects both values and identities. Cosmopolitans are tolerant of differences, favour cultural diversity and value down territorial ties. They have a tolerance and trust of other nationalities, no xenophobia, a willingness to understand other people and places, increased interest to travel & work abroad, and tolerance of immigrants & strangers (Norris & Inglehart, 2009). Localism can be seen as the other extreme on the other end of cosmopolitanism; where the latter has an intrinsic interest in indulging in diverse cultures, the former is interested largely on his own culture (Gunesch, 2004).. 政治大. 立. One of the aspects of cosmopolitanism is the interest and exposure to information and. ‧ 國. 學. knowledge about diverse cultures. Jeffres et al. (2012) seek to investigate the ‘knowledge gap’, a phenomenon whereby there is a distinct difference in the level of knowledge between. ‧. classes, considering the advent of the Internet as a game-changer. The findings show that. y. Nat. io. sit. media exposure enhances knowledge of international affairs with print media being used in. n. al. er. conjunction with the Internet as an informational source. The Internet, in particular, showed. Ch. i Un. v. linkage to increased international knowledge level. On the other side of the Internet debate,. engchi. however, is the question of whether it is used as an entertainment medium more than a news print medium. If it is the latter, it stands therefore that not much positive impact towards the knowledge gap can be derived from it. A study by Johnson, Braima & Sothirajah (1999) found that although the Internet, television talk shows, MTV, and radio talk shows had little influence on knowledge of the issue stances of Bill Clinton and Bob Dole in the 1996 campaign, they affected candidates images.. 21.

(32) 2.6 Theory 2.6.1 Theory of Social Construction of Reality According to Ibroscheva & Ramaprasad (2008) and other scholars, the mass media are by no means an accurate, empirical reflection of the world but rather a medium of constructing and maintaining particular interpretations of that reality . Within a larger system of social processes, the media presents certain meanings, usually of the dominant social groups over those of the social subordinate groups "through discourse (Ibroscheva & Ramaprasad, 2008). The influence of the media on construction of reality is only in. 治政 conjunction with the relationships between social groups, knowledge, media literacy, 大立 exposure etc. of those consuming it (Mezzana, n.d.).. ‧ 國. 學. Berger and Luckmann (1967) claim that media content presents a symbolic. ‧. representation of objective reality that affects consumers’ thoughts, experiences and actions. sit. y. Nat. in the social world (Siu, 2009). Hall (1999) confirms that there are multiple meanings that. n. al. er. io. arise which vary according to societal conditions, while Giddens (1984) adds that these. v. meanings are transmitted through discourse within the societal systems, aided by institutions (Siu, 2009).. Ch. engchi. i Un. Tuchman (1976, 1978) reports that based on interactions between news reporters and officials, the news is socially constructed rather than reported factually. In this way, the news is a form of social construction, whereby information that is included or excluded is critical. Entmann(1991, p6-25) described news frames as “keywords, metaphors, concepts, symbols, and visual images emphasized in a news narrative”. You can identify these media frames by seeing what is prominent in news (Siu, 2009).. 22.

(33) Galting identifies communication as one of the types of imperialism that determines the exchange of cultural products between the ‘centre’"and ‘periphery’"nations. News and information flow patterns trickle down from these core nations through transnational news agencies. The effect is usually that the periphery nations know little to nothing of countries not “filtered through the lenses of the developed media systems.”"This is a function of agenda-setting. In short, only information that is deemed important by developed nations is distributed to the rest of the world (Thussu, 2000). Eagly (1987) and colleagues conducted a few of the chief studies in stereotype origin. 政治大 group learn about members of another group through observing behavior. Therefore little 立 and theoretical foundations. They found that, in the sociocultural thesis, members of one. ‧ 國. 學. contact with other group members leads to a reliance on media for global information. As several factors, including agenda setting, determine which information is disseminated, there. ‧. is a very real danger of ‘erroneous information’"being spread worldwide (Ibroscheva &. io. sit. y. Nat. Ramaprasad, 2008).. er. 2.6.2. Cultivation Theory. al. n. iv n C U on violence, the cultivation h etontestg media Originated by Gerber in the 1960s c h i effects theory examines how media exposure influences people’s perception. Gerbner (1973) later developed the cultivation hypothesis which states that media has such a central place in our lives that it dictates our ‘symbolic environment’."This theory posits that the more one exposes themselves to media messages, the more one believes those messages are real and valid (Cultivation Theory, n.d.). Media messages then essentially substitute personal experience and become the means by which one knows about the world. However deviant from reality these views may be, persistent exposure leads to their consensual adoption by society.. 23.

(34) Cultivation is the generation and adoption of the ‘television view’"(Boyd-Barrett & Braham, 1987, p99). Tan et al. (1997) show that cultivation theory can instill behavior, attitudes and values through direct or mediated observation of others. Their research also indicates that adolescents are especially susceptible to messages received through media (Lee et al., 2009). Gerbner (1998) further posits that heavy media consumption leads consumers to describe the real world more like the portrayals they see in the media. It follows therefore that these perceptions will thereafter manifest into attitudes and behavior that are in line with the. 政治大. dominant messages. (Kean et al, 2012) Boulding (1956) expressed concern that the images. 立. transmitted through media are insufficient, and will therefore only serve to disseminate. ‧ 國. 學. stereotypes to those who are heavily dependent on the media (Lee et al., 2009).. ‧. Cultivation effects can be broken down into two levels: first and second order beliefs. The former is a generalized world beliefs, while the latter refers to a more specific attitudes. y. Nat. er. io. sit. about those beliefs (Cultivation Theory, n.d.). Zhang (2010) asserts that cultivation theory provides a useful framework for explaining the effects that stereotypical portrayals on the. n. al. Ch. i Un. v. media have on consumers’ perceptions of outgroups. After heavy exposure this leads to an. engchi. acceptance of these mediated stereotypes as reality. This theory is even more relevant to a study such as this one, Zhang (2010) argues, where perceptions are largely shaped by the media because of few opportunities.. 24.

(35) 3. METHODOLOGY The present research uses a cross-sectional survey method to determine the stereotype endorsement of Africa and examine the relationships between Western media exposure, Cosmopolitanism and other background variables to Stereotype Endorsement among university students in Taiwan. Exploratory research, such as this one, aims to clarify the nature of a problem, define its scope, as well as give the researched a deeper understanding of an issue. McGivern (2003) offers cross-sectional research design as one of the appropriate research designs for. 政治大 within-the-sample differences in order to make compare groups. 立. exploratory research; it provides a snapshot from a cross-section of a population and relies on. ‧ 國. 學. This chapter has four sections. The first section explains the rationale for using an online cross-sectional survey. The second section describes the sample selection and. ‧. procedure. The third section explains the instruments and statistical tests to be performed.. sit. y. Nat. io. n. al. er. 3.1 Method Selection Rationale. i Un. v. The use of questionnaires, to test hypotheses and relationships between variables for. Ch. engchi. descriptive or analytic purposes, is a commonly used technique in small-scale social science research projects (Blaxter, Hughes & Tight, 2001; Czaja & Blair, 1996). According to Blaxter, Hughes & Tight (2001), the advantages of survey research include unbiased questions, the option of future replications, as well as being representative and able to be generalized. However, with numerical data displayed in tables, statistics, charts as the main focus, one disadvantage includes losing the link to wider issues. It is a snapshot in time rather than a “focus on underlying processes and changes.” On top of that, the researcher doesn’t know whether respondents truly understand the questions or not which may lead to truthfulness and accuracy issues (Blaxter, Hughes & Tight, 2001). 25.

(36) When choosing the survey mode, pace and social desirability bias, as well as cost of the survey must be taken into account. Online surveys, which are self-administered, allow the subjects to pace themselves as opposed to being paced by an administrator. Schum (2007) found that the time-pressured nature of telephone surveys created a cultivation effect. In addition to cultivation effect, there is also the problem of social desirability bias found when there is interviewer presence. Subjects tend to give responses they think will be considered acceptable or appropriate by the interviewer. Therefore, to increase accurate answers and reduce this bias, especially on a sensitive topic such as this one, an online survey has chosen as the best survey mode for this study (“Use of Surveys”, 2012).. 政治大. 立. 3.2 Research Design. ‧ 國. 學. The research design for this study follows two steps: first, a pilot study was conducted in order to gather data for validating of research variables; after the measurement instruments. ‧. were validated they were used along with instruments of other variables that were already. sit. y. Nat. validated in previous research, as instruments in the survey questionnaire of the present. n. al. er. io. research. The two key variables that were measured in the pilot study were Western media. i Un. v. exposure and stereotype endorsement. Their factor loadings and reliability results are. Ch. reported in Table 2 and 3, respectively.. engchi. Data analysis was conducted using SPSS statistical software. Inferential statistics, including multiple regression analyses correlation, were used as appropriate to answer the hypotheses, while frequencies and means will be used to illustrate the relationships between the variables. Measures used in this research were derived from existing literature (Rybina et al., 2009; Ibroscheva, 2005; Monroe, 2005; Willnat, He & Xiaoming, 1997). The survey questionnaire was first constructed in English. It has ten questions, a majority of which are multi-item, and takes approximately 10 minutes to complete. There. 26.

(37) were no incentives. Three bilingual graduate students at NCCU translated the survey questionnaire into Chinese. 3.3 Data Collection Taiwanese university students were chosen as the survey population for two main reasons: high media exposure (Chou, 2001), particularly to the Internet and the opportunities available for international knowledge, exposure and travel in university. Using convenience sampling, two samples were used for this study: a pilot study (n=122) and the main sample (n=215).. 政治大. 150 questionnaires were distributed to students in National Chengchi University (NCCU). 立. and Kang Ning Jr College (KNJC), between June 16 and 20, 2014. At NCCU, the author and. ‧ 國. 學. a fellow graduate student distributed questionnaires to students in the Main Library, food courts, and outside classroom buildings. At KNJC, permission was obtained from professors. ‧. to distribute questionnaires to their classes before the lessons began. A total of 122. y. Nat. sit. questionnaires were valid and retained for analysis.. n. al. er. io. Using Survey Monkey, the main sample was drawn from universities across Taiwan. i Un. v. using Facebook as the main platform for promotion. Survey Monkey is a website that allows. Ch. engchi. users to design surveys and collect responses from various sources online, including Facebook. Through the help of Taiwanese students, the survey was entered into various student-body Facebook groups in their universities including: National Cheng Chi University, TaTung University, China University of Technology, National Cheng Kung University and Ming Chuan University. From June 24 to July 7, 2014, 236 surveys were collected. Only 215 surveys were valid and retained for further analysis.. 27.

(38) 3.4 Pilot Study The data collected from the pilot study (n=122) was used for the measurement of two newly developed measures: Stereotype Endorsement and Western Media Exposure. The pilot study was conducted in order to achieve validity and reliability for the main study. The two measures were derived from previous research (Monroe, 2005; Willnat, He & Xiaoming, 1997) and subjected to factor analysis using Principal Component Analysis and orthogonal Varimax rotation. After eliminating several items which showed low correlation coefficients (r<+-.30), two more steps were taken to ensure the data was suitable for EFA (Yong & Pearce, 2013): Bartlett’s test of p <.001 confirmed that there were patterned relationships. 政治大. among the remaining items. Several items were removed from the Western Media Exposure. 立. and the Stereotype Endorsement scales as they were several missing values in the data and. ‧ 國. 學. lacked face validity for the study. The items that remained were used for the main study.. ‧ sit. y. Nat. 3.5 Measurement of Variables. io. er. 3.5.1 Demographics. al. Age, gender and monthly expenditures (excluding rent) were measured, as part of. n. iv n C demographic variables. Age was dummy (under 21years old) and h ecoded i U n ginto c hundergraduate graduate (22 years and older). Monthly expenditures asked subjects whether they spent a) less than NTD 5000, b) NTD 5000-10000, c) NTD 10,000-20000 and d) over NTD 20000, on a monthly basis. 3.5.2 Contact To determine the subjects’ contact experiences with Africa, the present research adapted from Ibroscheva (2005) a 5-point Likert Scale (1= not at all, 2=rarely, 3= sometimes, 4= most of the time, 5= all the time) asked the question: “How much contact do you have. 28.

(39) with Africa and Africans through these sources?” Components include: friends, family, university courses, traditional media and Internet. 3.5.3 Cosmopolitanism There are as many ways to measure cosmopolitanism as there are differing perspectives on its definitions. After an indepth review by Türken & Rudmin (2013, p71), it was found common to all scales was a ‘willingness to positively engage with a cultural other in a positive way.’ They also assert that, cosmopolitans are likely to have fewer stereotypes of and less inclined to disparage the ‘cultural other’.. 政治大 to analyze the impact of cosmopolitanism, among other factors, on consumption behavior in 立. The instrument chosen for this study is a four-item scale adapted by Rybina et al. (2009). ‧ 國. 學. Kazakhstan. The Cronbach’s alpha is .878. The items are:. 1. I like immersing myself in different cultural environments. Nat. sit. y. ‧. 2. I like having contact with people of different cultures. io. al. er. 3. I would enjoy travelling to foreign countries for an extended period of time. v. n. 4. Getting information and news from around the world is important to me. Ch. engchi. i Un. Each item was on a 5-point Likert scale (1= strongly agree, 2=somewhat disagree, 3= neutral, 3= somewhat agree, 5= strongly agree). 3.5.4. Perceived Bias of Portrayals. The scale for the variable Perceived Bias of Portrayals was derived from Ibroscheva (2005) research on American opinion makers’ perceptions of Russians and Eastern Europeans. It is a six-item scale about the nature of media portrayals of Africa on a 5-point Likert scale (1= strongly agree, 2=somewhat disagree, 3= neutral, 3= somewhat agree, 5= strongly agree). The Cronbach’s alpha is . 71. The items are as follows:. 29.

(40) 1. Unbalanced 2. Fair 3. Based on fact 4. Unrealistic 5. Distorted 6. Truthful Items 2, 3 and 6 are reverse coded, so that a higher score means a higher perception of bias of media portrayals. 3.5.5. Western Media Exposure. 政治大. Willnat, He & Xiaoming’s (1997) studied the relationships between foreign media. 立. exposure and perceptions of Americans in Hong Kong, Shenzhen and Singapore. They found. ‧ 國. 學. that foreign media consumption lead to negative stereotypical perceptions towards Americans. ‧. and that different media has different effects. In their study, exposure to Western media was. sit. y. Nat. measured by asking students to indicate what percentage of their total time devoted to media. io. al. n. movies (rented and cinema), books, and news magazines.. Ch. engchi. er. consumption was spent with Western TV, newspapers, radio (live and recorded), comics,. i Un. v. This relative measure of foreign media exposure avoids the often implicit assumption in cultural media studies that the time devoted to media use must directly correlate with potential media effects. When dealing with samples from different countries, methods measuring exposure in absolute terms (for example, in minutes per day) may provide misleading estimates because such measures do not take into account the competitive influence of local media, which is likely to interfere with or change the potential effects of the Western media. Similarly, in the present study, various print, visual, Internet and radio sources were listed to measure Western Media Exposure, with an emphasis on African content. 30.

(41) The KMO measure is .768 with Bartlett test of Sphericity p<.01 confirming patterned relationships. The four factors explain a cumulative variance of 59.9%. Table 2 reports the significant factor loadings, subscales and reliability for the 16-item Western Media Exposure scale. Subjects answered each item on a 5-point Likert scale (1= strongly agree, 2=somewhat disagree, 3= neutral, 3= somewhat agree, 5= strongly agree). Cronbach’s alpha is .856.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 31. i Un. v.

(42) Table 2 Western Media Exposure Factor Loadings and Reliability Results Print. Internet. Visual. Western movie Channels. .726. Western news channels. .641. Western informational Channels. .625. Viewing Western movies. .516. Readers Digest. .672. Time Magazine. .679. 立.652 .787. Africa travel. .757. y .561. al. n. Travel websites. .821. sit. io. Search engine. .808. er. Africa fiction. Nat. Africa information. 政治大. ‧. News Week Magazine. Cronbach’s α. .642. 學. ‧ 國. US & World News. Radio. Ch. News websites. .819. e n.772 gchi. i Un. v. .793. .739. British BBC. .563. Taipei ICRT. .443. Eigenvalues. 2.426. 2.426. 2.317. 2.19. % of variance. 16.60. 15.16. 14.47. 13.72. Cronbach’s α= .856. 32. .591.

(43) 3.5.6 Stereotype Endorsement Bastiam & Aslam (2006) and Carels et al. (2013)’s studies assessed stereotype endorsement by showing participants a list of attributes, and asking participants to indicate on a 5-point and a 7-point Likert scale respectively, how ‘true’ they believed each statement to be. In their study of foreign media exposure on perceptions of Americans, Willnat, He & Xiaoming (1997) chose to use a defined list of stereotype traits to measure stereotypes. While this type of close-ended questions may limit the descriptive opportunities for subjects, research has shown that free-response methodology may end up collecting countless stereotype attributes instead.. 政治大 Monroe (2005)’s research 立on the mass media effect on American’s attitudes towards. ‧ 國. 學. Africa provided a list of such statements. The present study adapted Monroe’s instrument into a 10-item stereotype endorsement scale (Cronbach’s α= .776) with three subscales:. ‧. Primitive Africa (5 items), Socioeconomic Status (3 items) and Environment (2 items).. Nat. al. er. io. neutral, 3= somewhat agree, 5= strongly agree).. sit. y. Subjects reported on a 5-point Likert scale (1= strongly agree, 2=somewhat disagree, 3=. n. iv n C The Kaiser-Meyer-Olkin (KMO) individual items in the Stereotype h values e n gforc all hi U. Endorsement scale were above .5 and the KMO measure was .766. With an eigenvalue cutoff of 1.0, there were three factors that explain a cumulative variance of 58.6%. Table 3 reports the significant factor loadings, subscales and reliability measures of Stereotype Endorsement. Items 6 and 7 are reverse coded.. 33.

(44) Table 3 Stereotype Endorsement Factor Loadings and Reliability Results Primitive Africa 1. Africans lacks most modern features. .787. 2. Most Africans are hunters. .772. 3. Africans generally practice witchcraft. .697. 4. African art is primitive. .604. .744. 政治大 .525. 7. Africans are often scientists. .770. 8. Africans earn less than Europeans. .665. .601. sit. n. al. er. io 9. Africa has heavy torrential rainfall. ‧. .593. 學. 6. African children attend schools. Nat. ‧ 國. 立. Cronbach’s α. y. 5. All Africans are black. Socioeconomic Environme Status nt. Ch. 10. Africa is mostly covered in jungle. engchi U. v ni. .845 .684 .796. Eigen values. 2.52. 1.79. 1.54. % of variance. 25.27. 17.94. 15.43. Cronbach’s α= .776. 34.

(45) 3.6 Research Framework & Hypotheses. 立. 政治大. ‧. ‧ 國. 學 sit. y. Nat. Figure 2- Research Framework. n. al. er. io. Research on the effects of social structural position and other determinants of. v. stereotyping (Oliver & Mendelberg, 2000; Kunovich, 2004) research found that individuals. Ch. engchi. i Un. with lower incomes are more likely to use stereotypes. Therefore the hypothesis: H1: There will be a significant difference in stereotype endorsement between those with higher monthly expenditures and those with lower monthly expenditures Hale’s (1998) research further reports that stereotyping was mostly dependent on the contact level. Those with higher contact had higher knowledge and lower stereotype scores. Therefore the hypothesis is:. 35.