The sample size was calculated using the mean difference in modified Ashworth scale (MAS) for spasticity between the experimental and control groups. Using G-power software (Ver.3.1) founded on pilot study results, we set the effect size at 0.992. When a 2-tailed test with a test power of b=0.8 and significance level of a=.05 was applied [16]. The calculated sample size was 30. Four additional participants were recruited in anticipation of a dropout rate of approximately 15% during the study