On September 9th, 2016, Hillary Clinton gave a speech at the LGBT for Hillary Gala in New York City in which she referred to ‘half’ of Donald Trump’s supporters as a ‘basket of deplorables’ who espouse ‘racist, sexist, homophobic, xenophobic [and] Islamophobic’ sentiments. Although Clinton had prefaced her comments by acknowledging that she was about to make an overgeneralization, the backlash to her comments was swift, with Trump charging that her comments showed her true ‘contempt for everyday Americans’. Clinton apologized for her remarks, but doubled down in depicting Trump’s campaign as one based in ‘bigotry and racist rhetoric.’ Was there any truth to Hillary Clinton’s depiction of Trump supporters? To what extent can American voters for various candidates in the 2016 election be distinguished from each other on the basis of their attitudes? Can supporters of the various presidential candidates be differentiated by measures related to some of the key themes in the 2016 election, such as authoritarianism, sexism, Islamophobia, racism, and attitudes towards LGBTQ individuals?
In an attempt to understand more about how voters differ from each other as a function of their attitudes, I went back to a dataset collected between the spring of 2014 and the fall of 2015. As part of a study on another topic, 465 individuals, mostly from Utah, and mostly men, completed an online survey which included measures of attitudes towards the LGBTQ community (modern homonegativity, old-fashioned homophobia, genderism, and transphobia), racism, Islamophobia, ambivalent sexism (the combination of both hostile and benevolent feelings towards women), and indicators of conservative ideology (right-wing authoritarianism, social dominance orientation, protestant work ethic, and religious orientation). On September 28th, 2016, two days after the first presidential election, I re-contacted these participants and asked them whom they were planning to vote for in the presidential election. Over the following week, 249 responded and indicated their intentions to vote for Clinton, Trump, various third party candidates, or that they were as of yet, still undecided.
Clinton voters in this sample reported significantly more positive attitudes towards a number of groups within our society, including gay men, transgender and gender diverse individuals, women, and ethnic minorities. Clinton voters also showed significantly lower levels of Islamophobia than Trump voters, but were not significantly different on this measure compared to voters who were still undecided or planning to vote for a third party candidate. In other words, given Clinton’s definition of ‘a basket of deplorables’ as being people who hold ‘racist, sexist, homophobic, xenophobic [and] Islamophobic’ sentiments, Trump supporters do appear to exemplify these qualities in comparison to Clinton supporters.
In addition to holding different views about specific groups of people, Clinton voters were also distinguishable from Trump and undecided/third party voters in their views about how the world and society should be structured and should function. Specifically, Clinton voters had significantly lower levels of authoritarianism and were less likely to subscribe to a hierarchically ordered sense of society (social dominance orientation). Clinton voters also tended to be less religious than both Trump and undecided/third party voters. In all cases, the effect sizes for these group differences were large.
Although Trump and undecided or third party voters could not be distinguished from each other on the vast majority of measures, they did significantly differ from each other on two variables: Islamophobia and intentions to vote. With respect to Islamophobia, Trump supporters were set apart from both Clinton and third party or undecided voters. Indeed, this may have be the strongest deterrent to voting for Trump for many of the undecided and third party voters. Regression analysis revealed that individuals were more than 3 times more likely to vote for Trump for every step they increased on the Islamophobia scale and 2.6 times more likely to be undecided or voting for a third party candidate for every step that they decreased on the Islamophobia scale.
If higher levels of Islamophobia predicted a vote for Trump, why did lower levels of Islamophobia not universally predict a vote for Clinton? Indeed, Islamophobia was not a significant predictor of voting for Clinton over any other candidate, despite levels of Islamophobia among Clinton voters being equal to or less than those within the undecided or third party group. The variable that played the strongest role in predicting a vote for Clinton was lower levels of ambivalent sexism, such that for every point decrease on the ambivalent sexism scale, individuals were 3.3 times more likely to indicate an intention to vote for Clinton. Higher levels of ambivalent sexism were also predictive of being an undecided or third party voter, such that for every step up on the ambivalent sexism scale, participants became 2.5 times more likely to be undecided or to pick a third party candidate. In other words, it appears that while lower levels of Islamophobia pulled undecided and third party voters away from Trump, the higher levels of ambivalent sexism among these same voters pushed them away from Clinton and directing them towards the undecided or third party candidate category.
If the reluctance to vote for Clinton was truly due to ideological beliefs alone (i.e., ‘her beliefs and policies do not align with my own, nor would they if she were a man’), then one would expect to see measures like social dominance orientation and right-wing authoritarianism showing up as significant and powerful predictors of voting for candidates other than Clinton. Instead, the strongest predictor of not voting for Clinton was ambivalent sexism, calling into question at least some of the claims that voting against Clinton has nothing to do with her gender. Simply put, ambivalent sexism was the strongest predictor of voting for a candidate other than Clinton in the 2016 US Presidential Election, over and above any ideological differences, or, indeed other forms of prejudice.
If you would like to take the Ambivalent Sexism Inventory to see where you fall, click here.
To read the published journal article on this research, please click here.
This blog originally appeared on the LSE USAPP Blog and can be found here: http://bit.ly/2dWA6JT
Blair, K.L. (2017). Did Secretary Clinton lose to a ‘basket of deplorables’? An examination of Islamophobia, homophobia, sexism and conservative ideology in the 2016 US presidential election. Psychology & Sexuality, 8(4), 334-355. PDF.