Purpose To investigate: (1) the interrater, and test-retest reliability of the figure-of-eight walk test (F8WT) in people with Parkinson's disease (PwPD); (2) the minimum detectable change in the F8WT times; (3) the concurrent and known-groups validity of the F8WT times; and (4) the cut-off times that best discriminate PwPD from healthy people and fallers from non-fallers with PD. Methods This was a cross-sectional study. Forty-three PwPD and 34 healthy people were recruited. The F8WT was performed along with the timed up and go test, 10 m walk test, Berg Balance Scale, Activities-Specific Balance Confidence Scale, Unified Parkinson's disease Rating Scale, and Hoehn and Yahr Scale. Results The F8WT showed good interrater and test-retest reliability (ICC = 0.964-0.978 and ICC = 0.905-0.920, respectively). The MDC was 2.77 s. The F8WT was correlated with other outcome measures. Significant differences in the F8WT times were found between PwPD and healthy people and between fallers and non-fallers with PD (p < 0.001 and p < 0.001, respectively). The cut-off times of 8.43 s best discriminated PwPD from healthy people, while 11.19 s best discriminated fallers from non-fallers with PD. Conclusions The F8WT is a reliable, valid, and easy-to-administer tool in assessing the walking skill of PwPD.