Gay, lesbian, bisexual and transgender identification and attitudes to same-sex relationships in Australia and the United States.
This paper reports on the size of the gay, lesbian, bisexual and transgender population (GLBT) in Australia and offers some comparative assessment of support for legally recognising same-sex relationships in Australia and the United States. Both parts of this paper draw on social attitudes and behaviour data available in the Australian Survey of Social Attitudes 2003 (AuSSA) as well as data sources available for the United States.
The legal and moral question of gay marriage has galvanised considerable political opposition in the United States and, to some extent, in Australia as well. (1) This opposition has become more openly political since gay marriages were first sanctioned in the US states of Massachusetts and California, and all 11 ballot measures in the 2004 American elections to ban gay marriage were successful. (2) In Australia, the debate is slightly more temperate but has led to similar outcomes. In fact, in August 2004, the Federal parliament moved with less controversy to exclude same-sex unions from the definition of marriage in the Marriage Act 1961.
I conclude this article by considering some of the political implications of the gay marriage debate; especially whether support for gay marriage or recognising same-sex relationships is a 'wedge' issue that puts left-of-centre parties, like the Australian Labor Party, at electoral risk.
ESTIMATING THE GLBT POPULATION IN AUSTRALIA
Obtaining reliable estimates of the non-heterosexual population in Australia (and most countries) and estimating the level of non-heterosexual sexual contact are understandably complicated. We would expect that increasingly gay and lesbian visibility and declining antipathy towards homosexuals among the populations of advanced democracies would assist social researchers in better estimating both identification and activity. The Australian Census now routinely records the number of same-sex couples in Australia, which provides one limited estimate of the gay and lesbian population in Australia. (3) The Australian Survey of Social Attitudes (4)--Australia's survey in the International Social Survey Program (ISSP) managed by the Centre for Social Research at the Australian National University--includes a question about respondent sexual identification. The question asks:
Are you: Heterosexual or straight, bisexual, gay, lesbian, transgender or Can't choose.
AuSSA Principal Investigators settled on four categories to encompass non-heterosexual sexual identification, relying on the finding of a Zogby poll conducted in the United States in 2002 that 'GLBT' (gay, lesbian, bisexual, transgender) is the community's preferred way of summarising its diverse sexual identity in that country. (5) Using anecdotal evidence from Australia, we decided that, in asking the question this way, there were few local risks, either of excluding a category of sexual identity or confusing respondents.
As I mentioned above, social researchers face considerable difficulties in measuring the overall GLBT population. The first difficulty arises in asking all respondents to indicate their sexual identification to academic researchers; this carries the risk of decreasing response rates by offending some respondents and adding to apprehension about survey confidentiality. I am able to report here that the sexual identification question did not appear to affect aggregate survey responses. I was one of the Principal Investigators and we fielded two similar AuSSA questionnaires (A and B)--one with the sexual identification question and one without--and there was little difference in response rates between the two. Our partial examination of returned surveys also indicated a very small minority disapproved by writing their objection on the survey. We conclude that there is no risk in including a regular question on sexual identification in either AuSSA or the Australian Election Study, and that the benefits of including this question as a standard demographic item are considerable.
A second and more general problem relates to responses from GLBT--identifying respondents themselves. We can identify two major (and obvious) reasons why any estimate of the GLBT will be conservative: some respondents are not open about their sexuality and sexual activity (either to themselves or to others) or seek to protect their privacy by concealing their real sexual identity, even in confidential research. Mail surveys like AuSSA, however, can and do partly overcome confidentiality problems encountered with survey techniques that cannot guarantee confidentiality or that involve research interviews (in which respondents are identified at least to the interviewer by phone or in person). In this respect, a confidential mail survey may draw out accurate responses from those who are not open about their sexuality. Of course, critics might reasonably argue that estimating the GLBT population does little to overcome the invisibility of this community, especially if estimates of its size and activity are too conservative. We accept the criticism--the reality of same-sex sexual identification or activity cannot be easily measured by any means. We stress instead that our aim is to obtain GBLT population data that will tell us about changes in self-identification over time, which will indicate either real growth in GLBT identification or growing willingness to express a sexual identity outside familiar or trusting environments.
Major opinion pollster, Zogby International, obtained one recent estimate of the GLBT population in the United States, which puts it as high as five per cent of the voting population. (6) We might reasonably expect that open gay and lesbian Americans are more likely to be registered voters. In Australia, a telephone survey conducted over 2001 and 2002 among 10,173 men and 9,134 women aged 16 to 59 years, and reported by Grulich et al. in the Australian and New Zealand Journal of Public Health (ANZJPH), indicates a gay and bisexual identification rate of 2.5 percent for men and 2.2 percent for women. (7) The weighted AuSSA sample (8) puts the gay and bisexual adult male population at 2.6 per cent, and the lesbian and bisexual adult female population at 1.3 per cent (see Table 1). While the male non-heterosexual population is similar across both surveys, AuSSA 2003 finds a smaller number of openly lesbian and bisexual women than the ANZJPH estimate. Although the two estimates for women differ quite a bit, both are lower than the rate found among men, and the relative proportions of bisexual women and lesbians (with more of the former) are similar. The smaller size of the AuSSA sample (n=2104) leads us to conclude that more evidence is required before we can explain the difference between the two estimates of the female gay and bisexual population.
A simple binary logistic regression model that includes respondent age, sex, location, education, employment, and religious identity provides us with a further indication of the factors that predict GLBT identification (that is; identifying as gay, lesbian, bisexual or transgender as opposed to heterosexual/straight or Can't choose) (see Table 2). Age--divided into four groups--is a strong predictor with all age groups under 65 years significantly more likely to adopt GLBT identities than the base category (65 years and over). This finding suggests that the real break in the rate of GBLT identification is found between older age groups, presumably between those still young enough to be positively influenced by the social changes of the 1960s and 1970s and those who were not. As we found earlier, men are more likely to report a non-heterosexual identity.
When we look at the geographical distribution of the GLBT community, we find that living in inner or outer metropolitan areas of cities of 100,000 or more does not predict a higher GLBT identification (compared with the base category--living outside of a city with a population of 100,000 or more). This is an interesting finding given the visibly large gay and lesbian populations of Sydney and Melbourne. It may reflect in part that mail survey results under-represent the GLBT community in inner cities who tend to be a younger, mobile and highly active population, and therefore less likely to respond to surveys. However, the weak predictive significance of geography may also indicate that a less visible GLBT population is located across Australia whose needs and presence are under-recognised. A larger sample, and a more detailed exploration of geographical factors, would help sort out these competing explanations.
Again, education and employment appear to have surprisingly weak explanatory power. None of the educational qualification categories--university degree and above, certificate or diploma, trade qualifications, Year 12 (or equivalent)--predict GLBT identification compared with the base category (not completing Year 12). Respondents with professional employment backgrounds (either currently or in their last job)--where we might expect a higher level of GLBT identification--are no more likely than the base (all other employment backgrounds) to nominate a GLBT identity. Religious identity, however, is a strong predictor with respondents who indicated they did not have a religion more likely to indicate a GLBT identity.
In sum, younger, non-religious and male respondents are more likely to identify as non- heterosexual with some conventional reference points for socially locating gays and lesbians--urban geography, education and employment--not significant. These findings suggest that the GLBT community spreads well beyond the conventional markers of a modern gay community made up of younger, urban professionals and service industry workers. If these results were to be replicated on a larger cross-section, they would have important implications for how GLBT community resources are organised and deployed.
One additional finding from bivariate analysis is a class difference between gay/ lesbian and bisexual respondents, which could be further tested in multivariate scenarios with a larger sample. Using the simple class self-identification item available in AuSSA 2003 (upper, middle, working, and no class), we find that, for both men and women, working class GLBT respondents are more likely to consider themselves bisexual (see Table 3). The same identification divide (not reported in Table 3) is apparent by university/non-university education.
The higher level of bisexual identification among working class respondents may reflect actual experiences and limited opportunities to express a homosexual identity, or suggest that the labels of 'gay' and 'lesbian' carry with them a strong sense of sexual identity that may be unavailable to many Australians.
THE GAY MARRIAGE DEBATE
One might describe progress to greater recognition of gay and lesbian rights in Australia as difficult but steady, with homosexual acts decriminalised across the nation by the mid-1990s, and equal age of consent laws introduced more recently at the state level, such as in NSW in 2003. However, the question of taking the next step--and granting greater recognition to gay and lesbian relationships right up to equal marriage rights--has become a major political conflict in some countries. The gay and lesbian movement, progressive politicians and progressive judges tend to be on one side, and conservative politicians and religious leaders on the other. In the past few years, several jurisdictions in the United States--in San Francisco and in Massachusetts--have moved to effectively legalise gay marriage. This has provoked hostility at the Federal level in the United States. Hostility to greater recognition of gay relationships sharply distinguishes Australia and the United States from English-speaking countries like Canada and New Zealand, and other western European countries where greater recognition and in some cases, equal marriage rights, has been achieved (see Table Four). The Netherlands and Belgium have instituted full marriage equality, and Spain and Sweden are expected to foHow under their social-democratic governments. Civil unions, which give many or all the rights available to married couples, are now the norm in Europe and are likely in New Zealand. Gay marriage equality is now accepted by four Canadian provinces while the Liberal-led minority federal Canadian government contemplates how to deal with equal marriage rights nationally.
While Massachusetts continues to legally sanction same-sex marriages, in August 2004 the Californian Supreme Court annulled over 4,000 marriages that had been licensed in the city of San Francisco. Before the November 2004 presidential and congressional elections, the Senate failed to pass a 'Proposed Constitutional Amendment to Preserve Traditional Marriage' which stated that: 'Marriage in the United States shall consist only of the union of a man and a woman'. But the success of constitutional bans on gay marriage in eleven states (including liberal Oregon) in the 2004 elections suggests that, overall, the majority of American voters are not supportive of gay marriages and the second Bush administration may again seek to amend the US Constitution to make the ban national. (9) It should be added that by defining marriage between a man and a woman, the Clinton-signed Defence of Marriage Act 1996 effectively achieves this federally.
By contrast with the United States where conflict between jurisdictions reflects divergent social interests and legal authority, Australia has easily eliminated any immediate prospect of equal marriage rights. In August 2004, the Federal government garnered Labor's parliamentary support to effectively exclude gay unions from the definition of marriage in the Marriage Act 1961. The legislation now says that: 'marriage means the union of a man and a woman to the exclusion of all others, voluntarily entered into for life' (Sect 5), and further states that:
A union solemnised in a foreign country between: (a) a man and another man; or (b) a woman and another woman; must not be recognised as a marriage in Australia (Sect 88EA).
AUSTRALIAN AND US ATTITUDES ON RECOGNISING SAME-SEX RELATIONSHIPS
The previous section provided reasonable grounds for considering Australia and the United States as somewhat similar cases. Here, I compare AuSSA 2003 findings with two recent opinion polls: a Newspoll survey of 1,200 adults conducted in Australia in February 2004, and a CBS/New York Times poll conducted in December 2003. (10) Taken together, AuSSA 2003 and the Newspoll survey cover Australian attitudes towards both legal recognition for same-sex relationships and gay marriage, and allow us to fully compare Australian responses with the CBS/New York Times responses.
On recognising same-sex couples, AuSSA 2003 asks respondents:
Thinking about relationships and children today, please tell us how much you agree or disagree with each of the following statements: The law should recognise same-sex relationships.
The AUSSA 2003 item does not explicitly identify gay marriage, as the Newspoll survey question does:
Are you personally in favour or against same sex couples being given the same rights to marry as couples consisting of a man and a woman?
The CBS/New York Times poll of 1,914 adults asks two questions, which allow us to contrast support in the United States for gay marriage and civil partnerships:
Would you favour or oppose a law that would allow homosexual couples to legally form civil unions, giving them some of the legal rights of married couples? What about marriage? Would you favour or oppose a law that would allow homosexual couples to marry, giving them the same legal rights as married couples?
First, we examine the two Australian findings (see Table 5). They show support for legal recognition and gay marriage is less than 40 per cent. AuSSA 2003 produces lower support for an arguably more modest (or general) proposition--legally recognising same-sex relationships--than Newspoll, which is probably explained by the higher 'Neither' results (25 per cent versus Newspoll's 18 per cent). (11) However, the findings do not reveal majority opposition to the respective propositions. Clearly, there are still a high number of Australian voters who are undecided about gay marriage or unwilling to offer firm opinions; as Adam Berinsky makes clear, declining to give a definitive response does not necessarily mean that a respondent is without an opinion on the subject. (12) In a separate analysis, Ann Evans and Edith Gray have done multivariate analysis on the legal recognition item in AuSSA 2003, finding that younger respondents, and women, university educated and non-religious respondents are all more likely to support legal recognition. (13)
The CBS/New York Times poll produces a much lower non-opinion score, presumably because face-to-face interviewers prompt respondents to make a definitive choice. We find that in the United States civil unions have more popular support than gay marriage (see Table 5), and that support for civil unions over gay marriage is relatively higher among older Americans. (14) This result suggests that, although older Americans are more opposed to recognising same-sex relationships, some of this opposition is not directed against same-sex recognition per se, but against gay marriage, which older generations cannot reconcile with traditional unions. Comparing Australian and American attitudes towards gay marriage, there is stronger support in Australia: we would need to make unrealistic assumptions about the likely preferences of non-responses in the Newspoll survey to attain disagree responses close to that recorded for the United States. The most likely conclusion from these poll data is that Australians are slightly more tolerant towards same-sex recognition than Americans, if only because there appears to be a higher level of committed disagreement with this kind of recognition in America.
One of the most significant comparative findings is the gender gap in support between the two countries (see Table 6). In Australia, women's support for the propositions is 12 per cent higher in AuSSA 2003 and 18 per cent higher in Newspoll. Women are more inclined to support recognition than men. By contrast, Australian and American men register quite similar levels of disapproval but there is a clear divergence between Australian and American women. In fact, we can conclude that the stronger support for same-sex recognition in Australia can be attributed to higher support among Australian women than among their American counterparts. Younger men are more supportive than older men, but the gap between young and older women is much greater (the difference between men aged 18-34 and 65 and over is around 23 per cent but among women in the same age groups, the difference is a substantial 37 per cent).
Support for same-sex recognition by religion across the two countries reveals interesting findings (see Table 7). Here, the analysis is limited to AuSSA 2003 results and the CBS/The New York Times results for civil unions because, unfortunately, Newspoll does not record religious denomination, which would then allow for a cross-national comparison on the gay marriage item. I compare three major groupings across both countries--Catholics, Protestants and respondents with no religion. (15) In relative terms, we find that American Catholics are more liberal than Australian Catholics, and Protestants are the most conservative broad grouping in both countries, although American Protestants are more likely to be evangelical Christians with very conservative views about homosexuality and gay marriage. This may help explain why Americans appear to register majority disagreement with recognising same-sex relationships. Respondents without a religion in both countries favour same-sex recognition, although non-religious Americans are more in favour than are their Australian counterparts. This can probably be put down to the very large size of the non-religious population in Australia (over 30 per cent in AuSSA 2003); in the United States, it is 10 per cent in CBS/New York Times data, and is closely tied to America's minority liberal cohort.
CONCLUSION: IS GAY MARRIAGE THE LATEST WEDGE?
This paper confirms earlier research estimating the GBLT population at around two per cent of the Australian adult population with higher numbers of gay and bisexual men than of lesbian and bisexual women. The strongest predictors of GLBT identification are being younger, male and non-religious.
Our findings also confirm that there is not majority support for recognising same-sex relationships or gay marriage in Australia, but in all likelihood Australians are more supportive than Americans. Still, considering the dispersion of social attitudes towards legal recognition of same-sex relationships across advanced democracies, Australia remains closer to the conservative end of the public opinion spectrum. The gender divide in Australia is particularly important--women are considerably more open to the prospect of greater recognition of same-sex relationships. Opposition among men remains a challenge for reformers. It is worth keeping in mind that, although, public opinion appears to have moderated, conservative views about homosexuality remain strong in both countries.
Just 10 years ago, the National Social Science Survey found that a large majority of Australians (around 75 per cent) thought that people with AIDS who had contracted the disease through homosexual sexual contact were to blame for their condition. (16) In the CBS/New York Times poll, 42 per cent think that homosexual consensual relations should be illegal. This suggests that many among the majority opposed to recognising gay relationships in the United States are motivated by a rejection of homosexuality.
There are strong feelings on both sides of the debate. Given this, was it politically expedient for the Australian Labor Party to support the Howard government's amendment of the Marriage Act to exclude the possibility of equal marriage rights between heterosexual and gay and lesbian couples? Some may argue that Labor simply avoided the Government's 'wedge' on this issue; in other words, it acted to limit any electoral backlash from its own socially conservative constituency. (17) But AuSSA 2003 finds more support than opposition for recognising same-sex relationships among Labor identifiers (38 versus 35 per cent, with 24 per cent uncommitted). A promising sign for Australian reformers is that there is not majority opposition to recognition, which we find for the United States. Many Australians are still to decide about same-sex relationships. This suggests that a reformist platform need not involve any serious electoral risk for its political advocates.
Table 1: Estimating GLBT identification for Australia, per cent AuSSA 2003 Grulich et al. 2003 (18 and over) (16-59 years) Men Women Men Women Gay 1.6 -- 1.6 -- Lesbian -- 0.5 -- 0.8 Bisexual 1.0 0.9 0.9 1.4 Transgender 0 0 -- -- Total GLBT 2.6 1.3 2.5 2.2 All respondents (N) 1022 1082 9718 9570 Source: Australian Survey of Social Attitudes 2003; Grulich et al. 2003 Table 2: Logistic regression model predicting GLBT identification, AuSSA, 2003 Age (ref: 65 and over) Odd ratios Wald stat 18-34 9.669 4.519* 35-49 8.249 4.087* 50-64 8.079 4.009* Male 2.158 4.892* Location (ref: non metro) Lives inner met 1.417 0.801 Lives outer met 0.54 1.815 Education (ref: Less than Yr 12) Degree and above 0.984 0.001 Certificate/diploma 1.358 0.389 Trade qualifications 0.668 0.336 Year 12 (or equivalent) 0.87 0.047 Professional employment 1.621 1.275 No religion 2.24 5.808* Constant 0.001 39.209* N=2051; Nagelkerke R square = 0.089. * p<0.05. Note: The table reports odds ratios for each independent variable, which measure the chances of each factor increasing or decreasing the rate of GLBT identification. For example, respondents aged under 65 are between eight and ten times more likely than respondents 65 and over to identify as GLBT; and men are twice as likely as women to identify as GLBT. Table 3: GLBT identification by gender and self-identified class, 2003, per cent Males Females Middle Working Middle Working Gay 2.1 0.6 -- -- Lesbian -- -- 0.8 0 Bisexual 0.7 0.8 0.4 1.3 Transgender 0 0 -- -- Total GLBT 2.8 1.4 1.2 1.3 All respondents (N) 422 483 472 457 Source: Australian Survey of Social Attitudes 2003 Table 4: Legal recognition of same-sex couples: state of play in selected countries, December 2004 Countries with equal marriage Recognising civil unions Netherlands (since 2001) Denmark (since 1989) Belgium (since 2003) Norway (since 1993) Canada (four provinces: Sweden (since 1995)--equal Ontario, British marriage likely Columbia, Quebec and Yukon) Luxembourg (since 1996) [US state of Massachusetts] Iceland (since 1996) Hungary (since 1996) Spain (Catalonia, 1998; Navarra, 2000)-- equal marriage likely France (since 1999) Germany (since 2001) Portugal (since 2001) Switzerland (since 2001) Finland (since 2002) Croatia (since 2003) Poland (since 2003) Scotland (since 2004) [US state of Vermont] Considering civil Countries not considering either gay marriage or unions civil unions Czech Republic Australia (Marriage Act 1961 [2004 amendment United Kingdom enshrines marriage between a man and a woman]) New Zealand United States (Defense of Marriage Act 1996 (legislation passed enshrines marriage between a man and a woman) first parliamentary US state constitutional bans (on gay marriage): vote) Hawaii, Nebraska, Alaska, Nevada, Montana, Louisiana, Oregon, Arkansas, Georgia, Kentucky, Michigan, Mississippi, Montana, North Dakota, Ohio, Oklahoma, and Utah Sources: USA Today, 14 July 2004, (updated and amended by author) Table 5: Support for recognition of same-sex relationships in Australia and the United States, 2003, 2004, per cent Australia United States Legal recognition of same- sex relationships* Gay marriage** Civil unions*** Agree 35 38 41 Neither 25 18 7 Disagree 40 44 52 Total 100 100 100 Total N 4,267 1,200 1,914 United States Gay marriage*** Agree 33 Neither 5 Disagree 62 Total 100 Total N 1,914 Sources: *Australian Survey of Social Attitudes 2003; **Newspoll 2004; ***CBS/The New York Times poll, December 2003 Note: 'Neither' category includes 'Can't choose' for AuSSA 2003, and 'Non-response/Don't know' for CBS/The New York Times poll. Table 6: Support for recognition of same-sex relationships in Australia and the United States, 2003-2004, per cent Australia United States Legal recognition of same-sex relationships* Gay marriage** Civil unions*** Males Females Males Females Males Females Agree 28 40 29 47 36 45 Neither 21 23 15 20 8 6 Disagree 51 37 56 33 56 49 Total 100 100 100 100 100 100 Total N 1,953 2,210 na na 472 587 United States Gay marriage*** Males Females Agree 27 37 Neither 5 6 Disagree 68 57 Total 100 100 Total N 472 587 Sources: *Australian Survey of Social Attitudes 2003; **Newspoll 2004; ***CBS/New York Times poll, December 2003. Note: 'Neither' category includes 'Can't choose' for AuSSA 2003 and 'Non-response/Don't know' for CBS/New York Times poll. Table 7: Support for recognition of same-sex relationships by religion in Australia and the United States, 2003-2004, per cent Catholic Protestant No religion Australia US Australia US Australia US Agree 30 45 26 33 51 67 Neither 25 5 24 7 20 10 Disagree 45 50 50 61 29 24 Total 100 100 100 100 100 100 Total N 990 225 1,551 615 1,081 123 Sources: Australian Survey of Social Attitudes 2003; CBS/New York Times poll, December 2003. Note: 'Neither' category includes 'Can't choose' for AuSSA 2003 and 'Non-response/Don't know' for CBS/New York Times poll.
I am grateful to my colleagues in preparing this article: Associate Professor Katharine Betts (Swinburne University of Technology) for both encouragement and suggested references; Mr Peter McCarthy (Macquarie University) for his advice and comments on the draft; and Ms Sophie Holloway (Australian Social Science Data Archive at the Australian National University) for obtaining data and advice on recoding it.
(1) For a discussion of the pros and cons of gay marriage, see A. Sullivan (Ed and intro), Same-sex Marriage: Pro and Con: a Reader, Vintage Books, New York, 1997.
(2) New York Times, 4 November 2004.
(3) See, for instance, B. Birrell and V. Rapson, 'How Gay is Australia?', People and Place, vol. 10, no. 4, 2002, pp. 59-67.
(4) The Australian Survey of Social Attitudes 2003 (AuSSA) is the first in a biennial social survey series conducted by the Centre for Social Research at the Australian National University (ANU). Using a mail survey method, AuSSA provides cross-sectional data on the social attitudes and behaviour of Australians, repeating a core questionnaire for each cross-section and fielding specific modules relevant to the changing needs of the social research community. AuSSA is also Australia's contributing survey to the International Social Survey Program and regularly includes ISSP modules. The sample in 2003 was 4,270 and includes demographic and behavioural variables that survey: sex, year born, income, education, employment, home ownership, union membership, languages spoken, birthplace, ancestry, household composition and religion. Also included are questions about the partner of the respondent, employment, highest-level of education and income. For more information, see http:/aussa.anu.edu.au or http://assda224-100.anu.edu.au/nesstarlight/index.jsp.
(5) See Zogby International/GL Census Partners Poll, 'Attitudes Towards Gay Rights Organizations', 2002, Online available at http://www.glcensus.org/polls/gayrights.htm.
(6) Zogby International Interactive Poll, 13-18 November 2002
(7) A. Grulich, et al., 'Sex in Australia: homosexual experience and recent homosexual encounters', Australian and New Zealand Journal of Public Health, vol. 27, no. 2, 2003, pp. 158-159
(8) AuSSA 2003 is weighted according to a matrix of age (20 and over), educational attainment and sex using data from the ABS Census 2001. The weight is published in the online AuSSA 2003 dataset; Online available: http://assda224-100.anu.edu.au/nesstarlight/index.jsp.
(9) New York Times, 4 November 2004
(10) CBS/New York Times Monthly Poll #2, (December 2003), ICPSR no 3984 (May 2004), [data file prepared by Ms. Sophie Holloway, ASSDA/ACSR]
(11) The different rate of 'non-opinion' responses is largely a product of the different methodologies: Newspoll uses telephone interviews, which prompt a definitive response and the AuSSA 2003 mail questionnaire includes both a 'Neither agree nor disagree' and 'Can't choose' option.
(12) On the larger significance of non-responses to complex or contentious social survey questions, see, for instance, A. Berinsky, Silent Voices: Public Opinion and Political Participation in America, Princeton University Press, Princeton NJ, 2004
(13) A. Evans and E. Gray, 'What makes an Australian family?' in S. Wilson et al., Australian Social Attitudes: the First Report, University of New South Wales (UNSW) Press, Sydney, 2005, forthcoming
(14) For the CBS/New York Times results, we find that, among 18-29 year old respondents, support for civil unions and gay marriage is the same (59 versus 58 per cent); however, among respondents over 64, support for civil unions is relatively higher (32 versus 15 per cent).
(15) For Australia, the aggregate category of 'Protestant' includes Anglican/Church of England, United Church/Methodist, Presbyterian and Reformed, Baptist, Lutheran and Pentecostal. Protestant is one category in the CBS/New York Times poll.
(16) See National Social Science Survey 1993 dataset; Online available: http://assda224-100.anu.edu.au/nesstarlight/index.jsp.
(17) For a definition of the political tactics that might count as wedge politics, see S. Wilson and N. Turnbull, 'Wedge politics and welfare reform in Australia', Australian Journal of Politics and History, vol. 47, no. 3, September 2001, 384-402.
|Printer friendly Cite/link Email Feedback|
|Publication:||People and Place|
|Date:||Dec 1, 2004|
|Previous Article:||Declining marriage rates and gender inequity in social institutions: towards an adequately complex explanation for childlessness.|
|Next Article:||Attitudes to abortion in Australia: 1972 to 2003.|