<p>Pedestrian safety is a pressing concern in India’s urban mobility landscape, particularly at uncontrolled intersections where infrastructure limitations and unpredictable vehicle movements heighten risk. While previous research has explored pedestrian crossing behaviour, the nuanced role of attitudinal factors—central to the Theory of Planned Behaviour (TPB)—remains underexamined in such contexts. This study applies an extended TPB framework to analyze pedestrian crossing behaviour at six unsignalized intersections in Hyderabad, India. Using a self-rated questionnaire among 1500 pedestrians, the paper investigates how attitude constructs made up by positive attitudes, violations, aggression, errors, lapses, and health and psychological aspects come into interplay with socio-demographic variables in determining both behavioural intention and self-reported crossing behaviours. A Structural Equation Model (SEM) was developed to examine direct and indirect effects among these constructs. The results underscore the predictive strength of behavioural intention and highlight the influence of demographic characteristics such as age, predominantly younger male pedestrians, gender, education, and income on crossing behaviour. The study contributes to both theory and practice by integrating psychological constructs with data-driven modelling, offering insights into the mechanisms shaping pedestrian behaviour in developing urban settings. Based on the findings, targeted interventions are proposed, including awareness campaigns for vulnerable demographic groups and infrastructural improvements for high-risk intersections. These results hold significant implications for urban planners, transport authorities, and policymakers aiming to enhance pedestrian safety through evidence-based, context-sensitive strategies.</p>