BACKGROUND: The Stroke Impact Scale-16 (SIS-16) is used clinically and in research to measure quality of life after stroke. There are no studies that have estimated how much change on the SIS-16 is an important amount of change. The aim of this study was to estimate the minimal clinically important difference (MCID) of the SIS-16. METHOD: SIS-16 scores of participants receiving services at 2 outpatient physical therapy clinics who met inclusion and exclusion criteria (n = 36) were taken at baseline and discharge. At discharge, participants and their physical therapists rated their perceived amount of recovery on a Global Rating of Change (GROC) scale. Estimated MCID values were calculated for the SIS-16 using receiver operating characteristics curves with the GROC as anchors. RESULTS: There was no difference in baseline characteristics between participants who reported important improvement and those who did not. There was a fair relationship between change in SIS-16 scores and participants’ (0.41) and physical therapists’ (0.38) GROC scores. Change in SIS-16 scores was significantly greater in participants who reported an important amount of change on the GROC scale and those who did not. Estimated MCID of the SIS-16 ranged from 9.4 to 14.1 depending on the anchor. CONCLUSION: The estimated MCID values presented in this study provide a way for clinicians to evaluate meaningful change in individual patients and for researchers to evaluate meaningful change between groups.