Purpose: In real life there are both straight-paths and curved-paths. To evaluate walking ability of both kinds, a figure-8 walking test (F8WT) was developed. The aim of this study was to validate the measure in hemiplegic patients with walking difficulties and to identify correlations of curved walking ability with straight walking ability, motor function, and walking performance ability. Methods: Twenty subjects participated in this study. Curved walking was measured by a F8WT. Straight walking ability was measured by a 10-meter walking test (10MWT). Dynamic balancea bility was measured by timed up and go (TUG) tests. Walking performance ability was measured using a modified motor assessment scale (MMAS). Motor function was measured by the Fugl-Meyer assessment (FMA) scale. Data were analyzed using Pearson correlation analysis. Linear regression analyses were performed to explore other functional tests in mobility ability by F8WT time, 10MWT (dependent variable). Results: There was a significant positive correlation of F8WT time with 10MWT and TUG. There was a significant negative correlation of F8WT time with MMAS and FMA-coordination. There was a significant positive correlation of 10MWT with TUG. There was a significant negative correlation of 10MWT with MMAS and FMA-coordination. The F8WT time for curved walking ability was attributed to 10MWT for straight walking ability as 94% level of contribution. Conclusion: The results suggest that the F8WT is a good instrument for measuring walking ability because there is a robust correlation of F8WT time with 10MWT, TUG, MMAS, and FMA-coordination in hemiplegic patients who, after stroke, have a mobility deficiency.