Valid approaches to conveniently measure stress reactivity are needed due to the growing evidence of its health-impairing effects. This study examined whether the Perceived Stress Reactivity Scale (PSRS) predicts cardiovascular and psychological responses to psychosocial stressors during daily life and during a virtual reality (VR) Trier Social Stress Test (TSST). Medical students answered a standardized baseline questionnaire to assess perceived stress reactivity by the PSRS. The PSRS asks participants to rate the intensity of their typical affective responses to common stressors during daily life. They were further asked to participate in a VR-TSST and in an ecological momentary assessment (EMA) over a period of three consecutive workdays during daily life. Blood pressure and self-reported stress were repeatedly, heart rate variability (HRV) continuously measured during the VR-TSST and EMA. Furthermore, participants repeatedly assessed task demands, task control and social conflict during the EMA. Data was analysed using multilevel analysis and multiple linear regression. Results indicate that the PSRS moderates associations between blood pressure (but not HRV) and demands and control during daily life. Furthermore, the PSRS directly predicted self-reported stress, but did not moderate associations between self-reported stress and demands, control and social conflict. The PSRS did not predict physiological and self-reported stress responses to the VR-TSST. This study partly confirmed convergent validity of the PSRS to stress reactivity in daily life. Furthermore, the lack of association between the PSRS and stress responses to the VR-TSST calls for future studies to search for reliable and valid ways to assess stress reactivity.