Purpose: The purpose of this study was to investigate how college students’ perceived stress, cognitive stress, and somatization affect their heart rate variability (HRV). Methods: This study is a cross-sectional survey research on 191 university students, registered at the G University. The perceived stress scale (PSS) and cognitive stress response scale, were used to assess level of stress. The somatization symptom scale of the Symptom Check List 90 (SCL-90), was used to assess level of somatization caused by stress. To assess heart rate variability (HRV), we conducted a five-minute test using a pulse wave analyzer, to analyze short-term HRV. Results: The SCL-90 somatization score had relatively high positive correlation (p<.001) with cognitive stress, but low positive correlation (p<.001) with perceived stress. Cognitive stress response had low negative correlation (p<.001) with 1nSDNN and 1nRMSSD among HRV parameters. Perceived stress was not correlated with HRV. Multiple regression analysis showed that variables of perceived stress, cognitive stress, and somatization symptoms, could not explain HRV. By contrast, one of the HRV indicators, 1nSDNN, was affected by age, gender, and aggressive-hostile thought, the latter being a subscale of the cognitive stress response scale. Conclusion: This study suggests that stress evaluation for people in early adulthood will be more effective, if the evaluation examines cognitive stress and heart rate variability.