Mapping climate change awareness through spatial hierarchical clustering

Climate change is a critical issue that is expected to remain high on the political agenda for decades to come. While policy discussions at the international and national levels are essential, public awareness of climate change plays an equally crucial role in shaping effective responses. Since countries differ markedly in their exposure to climate-related risks, understanding cross-country differences in climate change awareness is particularly important. In this paper, we present a geographically-informed hierarchical clustering analysis aimed at identifying groups of countries with similar levels of climate change awareness. We employ a Ward-type clustering algorithm that integrates information on climate change awareness, socio-economic conditions, climate-related characteristics, and geographical distances between countries. To select appropriate values for the clustering hyperparameters, we propose a data-driven selection procedure that jointly accounts for within-cluster homogeneity, between-cluster separation, and explicit comparisons between geographically informed and non-spatial partitions. Findings reveal that incorporating spatial information leads to more stable clustering solutions and yields interpretable, geographically compact groupings relative to clustering approaches that ignore geography. In particular, we identify a clear contrast between Western countries, characterized by high and spatially cohesive levels of climate change awareness, and countries in Asia, Africa, and the Middle East, which exhibit lower average awareness and greater internal heterogeneity.