BACKGROUND: It is important to quantitatively assess tremor for accurate diagnosis and evaluation of the response to interventions in patients with essential tremor (ET). OBJECTIVE: The purpose of this study was to investigate the relationship between quantitative measures of postural tremor and clinical rating scale in patients with ET. METHODS: 18 ET patients performed a postural tremor task that required them to hold their arms outstretched parallel to the floor while wearing a gyro sensor based measurement system. The time domain variables were derived from the sensor signals. Additionally, the frequency domain variables were derived from the power spectrum of the angular velocity signal. Spearman correlation analysis was employed in the relationship between the variables and clinical score. RESULTS: The RMS angular velocity of roll and yaw directions at the hand joint were strongly correlated with the clinical rating scale (r = 0.7, p < 0.01). Similarly, the peak power of roll and yaw directions at the hand joint were moderately correlated with the clinical rating scale (r = 0.61 and r = 0.67, p < 0.01). In contrast, no significant correlation coefficients were observed in the peak frequency (p > 0.05). CONCLUSION: These results indicate that hand tremor of roll and yaw directions are more associated with assessment of severity of ET compared to other joints. This study suggests that quantitative measurements of postural tremor should be considered as tremor directionality as well as attachment location.