The connections between an individual's beliefs concerning climate change and their political ideology is well known by both researchers and the public. Understanding the causes and outcomes of those connections – most importantly, risk perception and policy support – are critical to improving climate change communication. Structural equation modeling (SEM) is a statistical method for understanding these kinds of causal relationships between variables. In this study, we used SEM to analyze a survey of 700+ undergraduate students in the southeast US which asked about their climate change knowledge, perceptions of climate change and climate scientists, and social and identity group belongings, with almost all items adapted from previous studies. Comparisons between models and path coefficients reveal complex relationships between political ideology, knowledge, and risk perception. Interestingly, students' perceptions of climate scientists' credibility was an important impactor on risk perception and has implications for theories and practice of climate change communication.