1
The national samples
Background and methods
Characteristics of the national samples
âI amâ: in love with the Hindi language, respectful to the national flag, respectful to the national anthem, respectful to my elders
(male, high school student, Delhi)
This chapter outlines the main elements of the research design: the survey methods used for securing samples, the questionnaire and its translation, and the gender vocabularies used to classify respondentsâ comments. In the process, some characteristics of the national samples are introduced, thus setting the stage for a comparative examination of national feminisms in Chapters 2 and 3, illustrated with comments from my research samples. I say âillustrateâ because the national samples in all but two countries (Australia and Japan) are very small and my own deficient monolingualism meant relying on local researchers to translate the questionnaire, originally designed for a western study, and the comments made by respondents. My linguistic limitations are, to a small extent, compensated by a period of living in Beijing and Tokyo, where colleagues and students have acted as both friends and informants. With the exception of Indonesia, I have spent at least two research weeks in every other country in my sample, consulting with academic colleagues and presenting seminars on my findings. Furthermore, I know of no other comparative study exploring young peopleâs attitudes to feminism and gender issues in such a range of countries in Asia and North America. So, while the samples are neither representative of middle class youth in each nation nor comparable in terms of numerical results, the qualitative results explore the very different ways, as well as unexpected convergences, in how some urban middle class youth in each nation frame and consider issues of sexuality and feminism.
The ten countries chosen for analysis are a purposive sample: they represent the major nations in the world and the cultural and religious diversity of Asia, while also reflecting my own research interests and knowledge. Data for these ten countries are shown in Table A1.1 in Appendix 1, covering population, income, religious affiliation and some gender equity measures. The worldâs four most populous countries are represented: China is still just ahead of India, with the USA and then Indonesia well behind (Troutner and Smith 2004:13; Lindsley 2004:223). The two largest economies are represented: the USA followed by Japan (with China rapidly gaining ground). Australia, Canada and the United States represent capitalist economies and western democracies, with Japan the major example of a mature capitalist democracy in Asia. India is the largest democracy in the world, while the Republic of Korea (hereafter South Korea) and Indonesia are more recent democracies. China, the largest communist country in the world, is represented along with a neighbouring communist country, Vietnam: both countries are well along the path to capitalist, if not democratic, reform.
Some details of the sample are shown in Table A3.1 in Appendix 3. Because of funding constraints, only two samples are robust, the Australian and Japanese samples. While I collected many Australian samples myself, elsewhere I asked local researchers to locate a co-operative high school and university, surveying about an equal number of respondents in each, aiming for 50 to 60 respondents in total, and ensuring that at least half the respondents were female. I asked the local researchers to replicate the procedure I usedâto survey students in a class specifically set aside for this purpose, with a researcher on hand to answer questions. Few of the local researchers achieved the exact sampling parameters I requested, although the Chinese, Canadian, Thai and South Korea samples are almost impeccable in this regard. By contrast, because of the ethics clearances required to gain access to almost any high school, the US sample is overweighted towards university students (90 per cent of the sample). At the other end of the scale, 84 per cent of the Australian sample are high school students. In two locations, local researchers recommended diversifying the sample beyond my two sources, to add a vocational college group in Thailand and a Hindi language college in Delhi. Given that my personal contacts were usually mandatory in securing entry into an Australian school, I allowed local researchers to choose the school, college or university which they surveyed. As a result, the Australian, US and Korean samples include some womenâs studies students in the university classes. This is reflected in warm and knowledgeable comments concerning feminism from the female students and some highly charged emotional responses, pro and con, in the Korean male sample.
The questionnaire can be found in Appendix 2. The opening section asked respondents to complete the statement âI amâŚâ up to ten times; one such statement from the Indian sample is reproduced above. Although resistance or inability to think often statements did not vary much, ranging from an average of around 7 to 8.5 (see Table 1.1), there were significant national variations in the way young people met this task. The statement âI amâ has different resonances in different languages. For example, in Mandarin Wo shi calls for a noun rather than an adjective, the latter not requiring shi (âamâ). My local translator negotiated this situation by placing shi in brackets so that the statement read âWo (shi)âŚâ Nevertheless, the linguistic difference might have contributed to the high percentage of categorical affiliations, so that the Chinese were the most likely of any sample to identify by nationality, their student status or gender (see Table 1.1): one Shanghai university respondent wrote âI am a boyâ ten times. Respondents in most of the samples were generally more likely to define themselves in terms of personality characteristics (for example âkindâ or âdumbâ) or their skills and interests (âa footballerâ, âpiano playerâ, âfanâ of Hindi film stars, wanting to travel overseas) rather than categorical affiliations (these are not shown in Table 1.1).
Religion
As will be seen in succeeding chapters, religion is a significant basis for respondentsâ conservative attitudes to gender issues, particularly given the late twentieth century expansion of Christian, Muslim and Jewish fundamentalism, Hindu and Buddhist nationalism, and post-communist re-affirmations of religion such as the Falun Gong (Therborn 2004:73). The major world religions are represented in my sample: Islam, Christianity, Buddhism and Hinduism (see Table A1.1 in Appendix 1).
Thailand is âthe world centre of Buddhismâ (Whittaker 2001:432), in fact, Theravada Buddhism (Limanonda 2000:249). Although women can become monks, Buddhism has been criticized for treating the female as an inferior life form: âIt has been commonly believed that because of bad karma, especially if a man committed adultery, a man would be reborn a woman. And, if a monk committed adultery, he would have to relive 500 lives as womenâ (Pongsapich 1997:14). Buddhism slightly outranks Shintoism in popularity in Japan (see Table A1.1 in Appendix 1). Indonesia is the worldâs largest Muslim country, although Islam has never been declared the official state religion. President Suharto wished to deny religious leaders overmuch power (Suzanne Brenner 1996:676) and an âoverwhelming majorityâ of Indonesians desire their state to be secular (Bennett 2005:10). Islamization of the Indonesian region began in the fifteenth century BPE (Before the Present Era and equivalent to BC), advancing from west to east, so that there are more Hindus in Bali and Christians in Flores and West Timor. Localized forms of animism, Hinduism and Buddhism were hybridized with customary law (adat) to produce a âhighly syncreticâ religious expression (Bennett 2005:9).
Out of its three centuries of colonial history, and a bloody dismemberment of the Indian sub-continent, the secular state of India emerged in 1947 as a predominantly Hindu state flanked by Pakistan on the east and west (predominantly Muslim, and which later became Pakistan and Bangladesh). In recent years, religious tension has again ratcheted up, possibly a response to the anxieties of growing income inequality consequent upon Indiaâs entry into the global economy. Like the respondent quoted above, a number of Indians in my sample expressed their national pride in a religious register: âAn Indian, A religious student, A defender of my country, An appreciator of the Hindi languageâ. This was particularly characteristic of the sample at the Hindu College in Delhi. By contrast, respondents in English language medium institutions in Mumbai cited their English language skills almost as frequently as their Hindi language skills.
Under communism in China and Vietnam, religion was regarded as a sign of
Table 1.1 Categories of affiliation in âI amâ statements, by gender and national samplea (responses as percentage of respondentsb)
backwardness (as it also was in India under prime ministers Jawaharlal Nehru and his daughter Indira Gandhi) (Vervoorn 2002:46). A survey of young people hanging out in shopping malls found that religion was important to 76 per cent of the US sample but only 21 per cent of the Chinese sample, although the Chinese youth did not know âwhat they believe inâ (Lindstrom and Seybold 2004:74, 87). China is the most secular of the nations in this study. However, in Vietnam only a third of the population declare no religion, and Buddhism is the major religion. Although Confucianism influences gender issues in China, its impact is more pronounced in Korea, where half the population declare they have no religion, and half each of the remainder are Buddhist and Christian.
Among Christian nations, the USA is often cited as peculiarly religious whereas Australia is described as a largely secular nation: âBy and large⌠Australians regard religious passion as something that other people haveâ (Maddox 2001:70). A survey conducted by the University of Minnesota found that US people distrusted atheists âmore than any other minority group, including homosexuals, recent immigrants or Muslimsâ (MacAskill 2007:8). Surveys conducted in the 1990s confirm US people as more religious in terms of belief in God, belief in the devil and weekly church attendance.1 Helen Hardacre (2005:242) notes that âThe Shinto world appears small from one perspective: there are about 80,000 shrines in Japan, but only one in 10 of them has a full-time priestâ. However, Jinja Honch (Association of Shinto Shrines) has strong connections with the Liberal Democratic Party, the dominant political party in Japan, and is a key player in the contemporary backlash against gender-equal legislative machinery and womenâs changing work and family roles (Hardacre 2005:241). In Australia, similarly, a Christian faction has close alliances with members of the Liberal-National Party coalition, exercising disproportionate influence (in relation to the number of practising Christians in the community) on âfamily valuesâ and policies promoting the ideal âsingle-breadwinner, nuclear family model of domestic lifeâ (Maddox 2001:287). One mechanism by which this is achieved is to disguise âChristian valuesâ as Australian âtraditionâ which is then ârelated to nationalism, civic order and public safetyâ (Maddox 2005:2). This elision of patriotism and religion is not dissimilar to the way Hinduism in India is aligned with nationalism. Reflecting the comparative degree of religiosity, the USA and Indonesian samples are the most religious of my samples (see Table 1.1:14.6 per cent and 13.3 per cent of respondents, respectively, use this category of affiliation in their âI amâ statements, by contrast with 6.4 per cent of the weighted total). Religion is a marker of internal difference in the USA, respondents generally identifying as âChristianâ or âJewishâ (although one was âJewish and Christianâ and another was a âNew Ageâ adherent). In Muslim Indonesia, by contrast, respondents merely asserted that they were âreligiousâ.
National identity and patriotism
Whereas females were more likely to claim religious adherence (7.7 per cent compared with 4.4 per cent of the weighted sample), the males on average were more likely to identify by nationality or express patriotic affiliation (19.9 per cent com-pared with 16.7 per cent of the weighted sample but a pattern of my Asian samples more so than the Anglophone samples), a characteristic also of Melanie Bushâs (2004:112) New York university sample and a cross-national survey of school children (Barrett 2007:102). The conservative religious cast given to patriotism pos...