Over the past decade, researchers have used such techniques to pick apart topics that social scientists have chased for more than a century: from the psychological underpinnings of human morality, to the influence of misinformation, to the factors that make some artists more successful than others. One study uncovered widespread racism in algorithms that inform health-care decisions; another used mobile-phone data to map impoverished regions in Rwanda...
Physicist Neil Johnson at George Washington University in Washington DC, and lead author of the hate study, is accustomed to criticism from social scientists. He says he cited the most relevant references. And as for search algorithms, social-media companies have the power to manipulate them, he says, “just as they are doing now to suppress the prominence of anti-vaccine and COVID-19 misinformation pages and groups”. He has studied misinformation, conflict and extremism and says he gets complaints every time he publishes a high-profile paper. But his work has struck a chord with policymakers: he is frequently asked to consult by organizations who like the quantitative nature of his work and the ability to model what impact interventions might yield. “We can really look at concrete questions in a way that I think they haven’t experienced in interactions with other academics,” he says. Johnson, for his part, is concerned that too many social scientists are rushing into computational approaches without proper training.