{"id":525586,"date":"2024-11-12T06:00:00","date_gmt":"2024-11-12T05:00:00","guid":{"rendered":"https:\/\/leregardlibre.com\/?p=525586"},"modified":"2025-12-13T22:50:12","modified_gmt":"2025-12-13T21:50:12","slug":"sex-ethnicity-trump-2024","status":"publish","type":"post","link":"https:\/\/leregardlibre.com\/en\/policy\/world\/sex-ethnicity-trump-2024\/","title":{"rendered":"Gender and ethnicity: the danger of simplistic voter analysis"},"content":{"rendered":"

The electoral conclusions of 2024, marked by the election of Donald Trump in the USA, are still largely based on \u00abidentity\u00bb categories. This approach, far from being trivial, influences the perception of society and reduces the diversity of opinions.<\/strong><\/strong><\/p>\n\n\n\n\n\n\n\n

\u00abWhite people voted Trump. Women voted against him.\u00bb In the wake of Donald Trump's election last week, after a first term from 2017 to 2021, political commentators were once again plentiful in dividing voters along gender and ethnic lines.<\/p>\n\n\n\n

And yet, when it comes to understanding a voter's motivations, there are many other possible categories: income level, profession, region, interests, consumer habits, marital status, and above all, political convictions. So why do we focus so much on ethnic origin and gender? Dissect the three most advanced hypotheses.<\/p>\n\n\n\n

1. These data would be the easiest to collect<\/h3>\n\n\n\n

Perhaps gender and ethnicity are simply widely available data. Historically, censuses have favored these categories because of their accessibility and statistical clarity. They are easy to measure, which makes them simple benchmarks in social research and analysis. But are they easy to interpret? If this categorization exists out of habit or convenience, it could well give us a biased view of electoral choices.<\/p>\n\n\n\n

Indeed, this vision often boils down to misleading shortcuts: when we say \u00abwhite people voted Trump\u00bb, this assumes that skin color is a determining factor in voting. In reality, these choices are influenced by complex factors that cannot be reduced to ethnicity or gender. We therefore end up projecting onto groups behaviors that could result from anything other than their skin color or gender.<\/p>\n\n\n\n

2. These categories would have high predictive power<\/h3>\n\n\n\n

It would seem that gender and ethnicity are indicators with strong predictive power. In other words, by observing these categories, we could guess electoral trends with a certain degree of accuracy. For pollsters, using these criteria means taking advantage of data which, from one election to the next, has shown statistical regularity.<\/p>\n\n\n\n

But correlation doesn't mean causation. If a majority of whites vote for Trump, it doesn't necessarily mean they do because\u2019<\/em>they are white. Again, this correlation may be the result of many other common factors within the same group. <\/a>For example, in the 2020 presidential election, 65% of rural white voters voted for Trump, while 52% of urban whites supported Biden. Similarly, in 2020, 60% of white women aged 18-29 voted for Biden, compared with 55% of white women aged 50-64 who supported Trump.<\/p>\n\n\n\n