BACKGROUND Sensory amplitude electrical stimulation (SES) and repetitive task practice reduce impairments and arm dysfunction when delivered separately following stroke. OBJECTIVE To determine if home-based, task-specific arm exercise was more effective when administered concurrent with SES. METHODS Thirty-eight subjects with chronic stroke and mean Fugl-Meyer Assessment (FMA) score 28/66 (15-45) participated. Subjects were randomly assigned to an SES (n = 20) or sham stimulation (n = 18) group. Subjects engaged in task-based home exercise for 30 minutes, twice daily, for four weeks while wearing a glove electrode on the impaired hand. Experimental subjects received SES while control subjects received sham stimulation during exercise. METHODS FMA and Arm Motor Ability Test (AMAT). RESULTS There were no significant between-group differences for outcome measures. There was a significant difference between the pre- and post-test scores in the SES group AMAT median time (P = 0.003 95% confidence interval (CI): -14.304, -6.365; effect size: 0.84). Practice time was not associated with changes in outcomes. Subjects with more sensorimotor dysfunction had significantly greater improvements on AMAT median time (P = 0.037). There was a significant relationship between baseline FMA score and FMA change score (r = 0.402; P = 0.006). CONCLUSIONS This study describes a unique SES delivery system via glove electrode that enabled delivery of SES during home-based arm task practice in stroke survivors. Task practice with concurrent SES did not demonstrate significantly better effects than task practice with sham stimulation, however there was a trend for greater improvement in one activity measure.