Longitudinal Predictors of Perceived Climate Change Importance and Worry among Italian Youths: A Machine Learning Approach

Abstract

The current study aimed to investigate the longitudinal predictors of perceived importance of climate change and personal worry among Italian youths. Specifically, we used machine learning techniques to examine the predictive importance of a wide range of socio-demographic factors, political perceptions, attitudes on a national and European level (identity, attitudes, tolerance, support for democracy, authoritarianism, nationalism, political trust), efficacy beliefs, social well-being, political interest, and different forms of participation on perceived importance of climate change and personal worry. In this longitudinal study, we collected data using a questionnaire in two waves at a one-year interval—in 2016 and 2017. Participants were 1288 Italian young adults (61.3% were female; 38.7% were male) whose mean age was 19.18 (SD = 3.29) ranging between 15 and 30 years. Breiman’s random forest algorithm performed better than Friedman’s gradient boosting machines algorithm. The random forest algorithm revealed that age, tolerance toward migrants, and tolerance toward refugees were the most important predictors of perceived importance of climate change and personal worry. Other important predictors were national/European identity, political interest, internal political efficacy, nationalism, social well-being, self-efficacy, authoritarianism, anti-democratic attitudes, EU warmth, and online and civic participation.