Recently, Alice Campbell and I revealed the demographic traits associated with people expressing support for equal rights for same-sex couples using the Household, Income and Labour Dynamics in Australia (HILDA) Survey – a large, longitudinal survey that is representative of the Australian population.
My subsequent analyses of the HILDA Survey point to another important factor: cognitive ability. Specifically, there is a strong and statistically significant association between higher cognitive ability and a greater likelihood to support equal rights between same- and different-sex couples.
This may shed some light on why those who stand against equal rights may not be persuaded by evidence-based arguments in the ongoing marriage equality debate.
Measuring cognitive ability and support for equal rights
From time to time the HILDA Survey collects one-off information from participants. During the 2012 face-to-face interviews respondents participated in three hands-on tests aimed at determining their cognitive ability. Such tests evaluated the degree to which participants were able to:
recall and recite backwards progressively longer strings of numbers;
correctly pronounce 50 irregularly spelled words; and
match symbols and numbers based on a printed key against time.
These tests are not perfect. They may contain some measurement error, may be culturally biased, and may not constitute a complete measure of cognitive ability. Yet they are widely recognised instruments routinely employed in psychological and educational research, and have been shown to be highly correlated with overall intelligence.
My analysis involved estimating the degree of support for the rights of same-sex couples at different levels of this measure of cognitive ability.
To do so, respondents’ scores in the three tests were rescaled and averaged into a composite measure of cognitive ability. Scores ranged from zero (lowest ability) to one (highest ability).
Support for equal rights came from a 2015 HILDA Survey question asking respondents to rate their degree of agreement with the statement “Homosexual couples should have the same rights as heterosexual couples do” on a scale from one (strongly disagree) to seven (strongly agree).
Analyses based on a sample of more than 11,600 people revealed that those with lower levels of cognitive ability in 2012 were much less likely than those with high levels of cognitive ability to express support for equal rights in 2015.
The association was substantially and statistically significant.
Some population groups – older people and those from non-English-speaking backgrounds, for example – may be more opposed to equal rights and also perform worse in cognitive ability tests. For the former group, this may be due to cognitive decline, and for the latter it may be due to English not being their first language.
To prevent this and other factors tampering with the results, I adjusted the models for age, gender, sexual identity, highest educational qualification, religiosity, ethno-migrant background, area remoteness, and state/territory of residence.
After these adjustments, as expected, the association between cognitive ability and support for the rights of same-sex couples faded moderately. Yet it remained large and statistically significant.
It is worth emphasising that education is controlled for in the models. Therefore, the results cannot be explained by people with high cognitive ability having higher educational qualifications.
The results were also quite robust: the patterns remained when excluding respondents from a non-English-speaking background, measuring support in 2011, and considering the measures of cognitive ability separately. However, the magnitude of the association differed across tests.
This finding poses the question of whether the pattern extends to people’s views about social equity in other life domains.
To test this, I extended the HILDA Survey analysis to examine the associations between cognitive ability and supportive attitudes toward women’s emancipation, women’s capability as political leaders, and single mothers.
The same pattern emerged across all of the outcomes. Higher levels of cognitive ability were unambiguously associated with greater levels of support for egalitarian worldviews.
The findings do not mean that all who intend to vote “no” in the marriage ballot have a low level of cognitive ability. Nor do they mean that all those who intend to vote “yes” have a high level.
Yet the results suggest that, on average, people who stand against equal rights for same-sex couples are less likely to have cognitive resources that are important to participating in meaningful debate.
These may include the ability to: engage in abstract thinking and process complex chains of ideas; separate arguments based on facts from unfounded ones; not feel threatened by changes in the status quo; and critically engage with new or diverse viewpoints.
These results may thus shed some light over why some on the “no” side may be failing to offer or accept evidence-based arguments, or why they keep relying on philosophically, historically or empirically flawed ones.
This applies, for instance, to the scientifically unsupported claim that children are worse off in same-sex households. In fact, these arguments are being exploited by a “no” advertising campaign that relies almost exclusively on emotional instead of rational arguments.
It is possible many supporters of the “no” case could not be convinced by reason and evidence. If so, the “yes” side’s best way to minimise the possibility of a surprise “no” victory – one that’s driven by a mobilised minority – may be to target the overwhelming majority of Australians who support equal rights to have their say.
Francisco Perales, Senior Research Fellow (Institute for Social Science Research & Life Course Centre) and ARC DECRA Fellow, The University of Queensland