Explaining landscape preference heterogeneity using machine learning-based survey analysis

We conducted a national survey on high-quality internet panel to study landscape preferences in Norway, using photos as stimuli. examined preference heterogeneity with respect socio-demographic characteristics and latent topics brought up by the respondents, ordinal logistic regression structural topic modelling (STM), machine learning-based analysis. found that pasture landscapes are most favoured (55%), while densely planted spruce forests least (8%). The contrast was particularly strong between eastern western men women, young old. STM revealed choices were mainly driven for openness, especially women. Other important drivers concerns regarding reforestation of former farmlands, aesthetic properties, forest management, biodiversity issues, cultural values. Our results suggest policies may clash socio-cultural preferences, failure account these undermine success policy.