Fear and belief predict perceived carnivore abundance in Golestan National Park, Iran

Description of the data and file structure This dataset includes predictor variables for three large carnivore species: the Persian leopard (Panthera pardus tulliana), grey wolf (Canis lupus), and brown bear (Ursus arctos). The data were collected as part of the first author’s PhD research on local community attitudes toward large carnivores in Iran’s first biosphere reserve, Golestan National Park. For each species, the dataset includes site-level variables including habitat type and elevation, as well as respondent-level variables including gender, social or cultural group, age, literacy status, occupation, household income, and experiences of livestock loss caused by these carnivores. Additionally, the dataset includes attitudinal measures that capture local perceptions and emotions, such as fear, perceived danger, happiness-and-pride, views on the importance of protecting populations, and family beliefs regarding carnivore conservation and management. We fully anonymised and coded (i.e., no participants names) in accordance with the ethical clearance procedures described in the manuscript. Also, we provided a separate dataset (in .csv format) and R scripts for fitting Bayesian multilevel regression models. The R scripts are included, and each script contains step-by-step guidance on how to use it, as outlined below: 1. Leopard_Covariates_2025.csv (leopard.R) 2. Wolf_Covariates_2025.csv (bear.R) 3. Bear_Covariates_2025.csv (wolf.R) Code/software These scripts, written in the R programming language. Human subjects’ data Ethical statement: We obtained ethical clearance letter (permit number No. 355135–81947/2022) issued by the Iranian Department of Environment. The data is completely anonymised.