ABSTRACT

Objective : The purpose of this study was to investigate the reliability and validity of a Mirror Movement Test forpatients with congenital cerebral palsy. Methods : Eleven items of the Mirror Movement Test were identified through a previous study performed by 72patients with congenital cerebral palsy. The data were analyzed through Winstep version 3.80 using the Raschmodel to examine the uni-dimensionality of the fit of each item, the distribution of difficulty of each item, and thereliability and appropriateness of the rating scale. Results : Based on a Rasch analysis, three (affected side) and two (unaffected side) out of the 72 patients wereconsidered to be inappropriate fits, and one item of the Mirror Movement Test for both sides was determined tobe an inappropriate item. The most difficult items were ‘touching the thumb and fingers’, and ‘clenching andunclenching the fist’. The easiest items were ‘flexing the elbow’, and ‘resistance to abducting the arm’. The fourcategories of rating scale were appropriate. The person and item separation reliability was above 0.90, which isa relatively excellent value. Conclusion : Mirror Movement Test is expected to be clinically useful. Follow-up studies on mirror movement areneeded for patients with various neurological diseases.