Background: The ability to turn while walking is essential for people's activities of daily living. Difficulties in turning while walking are commonly shown in people with multiple sclerosis (PwMS). The figure-of-eight walk test (F8W) is a clinical test assessing walking skill in a curved pathway; however, its reliability and validity have not been systematically examined for PwMS.Purposes: The study is aimed to investigate: (1) the test-retest reliability of the F8W in PwMS; (2) the standard error of measurement and minimum detectable change in the F8W times; (3) the concurrent and known-groups validity of the F8W times; and (4) the cut-off times that best discriminate fallers from non-fallers with MS.Method: This cross-sectional study included 41 PwMS and 33 healthy people. The F8W was performed along with the Timed Up and Go Test (TUG), Berg Balance Scale (BBS), Activities-specific Balance Confidence Scale (ABC), and Expanded Disability Status Scale (EDSS). To determine the test-retest reliability, the F8W was conducted twice, 7-10 days apart. The reliability was assessed using the intraclass correlation coefficient (ICC), BlandAltman plots, standard error of measurement (SEM), and minimal detectable change (MDC). To examine validity, the correlations between the F8W and the TUG, BBS, ABC, and EDSS were assessed using correlation coefficients, and the completion times of the F8W were compared between PwMS and healthy people, and between fallers and non-fallers with MS. The receiver operating characteristic curve was constructed to determine the optimal F8W cut-off time discriminating fallers from non-fallers with MS.Results: The F8W had excellent test-retest reliability with an ICC of 0.916. Bland-Altman plots showed high agreement between sessions. The SEM and MDC were found to be 0.45 and 1.25, respectively. The F8W indicated a moderate to strong correlation with other outcome measures (correlation coefficients ranged from -0.596 to 0.839, p<0.05). On the F8W, PwMS had a longer time than healthy people while fallers had a longer time than non-fallers with MS (p<0.001, and p<0.001, respectively). The cut-off time of 8.52 s best discriminated the fallers from non-fallers with MS.Conclusions: The F8W is a reliable and clinically available measurement tool for walking skill in PwMS.