Descriptive statistics and tests for normality (Shapiro-Wilk test) were performed for all outcome variables. A multifactorial repeated measures analysis of variance(ANOVA) was used to determine whether statistically significant differences existed between baseline and posttraining measuresfor the control and treatment groups for the three selected outcome measures: gait speed, AAS, and NHP. Paired Student’s t tests for the 13 subjects were carried out to assessthe efficacy of the overall training program for the following measures: functional perfor- mance (gait speed,rate of stairclimbing, andAAS); quality of life (NHP score); and muscle performance (muscle strength and spasticity). To reduce type I error, the Bonferroni correction32 was applied; thus, the significance level was set atp < .007.