The purpose of this article is to describe the relationship between poststroke upper limb muscle weakness and cocontraction, and clinical measures of upper limb motor impairment and physical disability. Electrormyographic (EMG) activity of the paretic and nonparetic wrist flexors and extensors of 26 chronic stroke survivors were recorded during isometric wrist flexion and extension. The root mean square (RMS) of the EMG signal was used as a measure of strength of contraction. A ratio of RMS of antagonist and agonist muscles was used as a measure of cocontraction. Upper limb motor impairment and physical disability were assessed with the Fugl-Meyer motor assessment (FMA) and the arm motor ability test (AMAT), respectively. The strength of muscle contraction was significantly stronger in the nonparetic limb (P < 0.001). The degree of cocontraction was significantly greater in the paretic limb (P < 0.001). The strength of muscle contraction in the paretic limb correlated significantly with FMA (r = 0.62 to 0.87, P < or = 0.001) and AMAT (r = 0.66 to 0.80, P < or = 0.001) scores. Similarly, the degree of cocontraction correlated significantly with FMA (r = -0. 70 to -0.64, P < or = 0.001) and AMAT (r = -0. 72 to -0.62, P < or = 0.001) scores. Muscle weakness and degree of cocontraction correlate significantly with motor impairment and physical disability in upper limb hemiplegia. This relationship may provide insights toward development of specific interventions. However, additional studies are needed to demonstrate a cause and effect relationship.