PURPOSE: This study is designed as a retrospective study, and identified the clinical usability of Sit to Stand (STS) test for predicting of fall incidence in stroke patients who experienced a fall within 1 year. METHODS: Between July 2011 and November 2012, 69 inpatients with stroke in K rehabilitation hospital were participated under voluntarily signing the informed consent form. STS test and 10m walk test (10MWT) were used to assess the muscle strength of lower-extremity and walking velocity, respectively. Also, we tested dynamic balance and motor function of lower-extremity in affected-side using with the Berg balance scale (BBS) and the Fugl-Meyer assessment of lower extremity (FM-L/E). METHODS: There were significant differences between subjects with fall-experienced group and without subjects without fall-experienced group in STS test, 10MWT, BBS scores and FM-L/E. STS test significantly showed a negative correlation between 10MWT (r=-.657), BBS (r=-.512), and FM-L/E (r=-.563). And, 10MWT have a influence on the performance of STS test (the capacity of explanation = 20%). The cut-off value of STS performance predicting falls experience is ${geq}14.36$ seconds (sensitivity=76%; specificity=79%, area under curve=.785). According to logistic regression analysis of falls experience, subjects ${geq}14.36$ s showed that 4.164 times (odd ratio) increased in falls than subjects < 14.36 s in STS test. CONCLUSION: This study demonstrated that STS test may be a useful tool predicting and measuring falls in patients with stroke. Further study will be needed to elucidate the kinematic analysis of STS test and the relationship between physical activity level and falls in stroke patients.