Introduction: Upper-limb robot-assisted therapy (RT) has been shown to improve motor recovery in chronic stroke patients. Some review studies demonstrated that compared with severely affected patients, moderately affected patients with stroke were more responsive to RT. Neuromuscular electrical stimulation (ES) can effectively induce muscle contraction for stroke patients with severe motor deficit. Thus, ES might serve as an adjuvant intervention for RT to optimize treatment effects. This study aimed to examine the effects of combined robot-assisted therapy with electrical stimulation (RT+ES) on motor control, motor impairment, and daily function. Methods: Twenty-nine patients (mean age: 65 years, mean onset time: 26 months) were randomized into one of the three groups: RT+ES, robot-assisted therapy with sham electrical stimulation (RT+sham ES), and control therapy (CT) groups. Each patient received 90-100 minutes of therapy, 5 days per week for a total of 4 weeks. Outcome measures, representing motor control, motor impairment, and daily function, were kinematic variables, the Fugl-Meyer Assessment (FMA), and the Motor Activity Log (MAL), respectively. Kinematic variables included total displacement (TD), peak velocity (PV), percentage of time where peak velocity occurs (PPV), and movement units (MU). All assessments were performed before and after treatment. Analysis of covariance was used to evaluate training effects. Results: Results showed significant group differences in two kinematic variables (PPV, F = 1.82, p < 0.01; normalized MU, F = 3.41, p = 0.05), proximal subscore of FMA (F = 3.58, p = 0.05) and MAL (amount of use, F = 4.38, p = 0.03; quality of movement, F = 3.71, p = 0.05) in favor of RT+ES. There were no significant group differences for TD, PV, the total FMA score, and the FMA distal subscore. Conclusions: These findings suggest that RT+ES may improve motor control by increasing movement smoothness, reduce motor impairment of proximal part of upper limb, and enhance the performance on daily activities. Further studies could may involve the outcome measures at the participation level such as performance on extended activities of daily living or community reintegration, and evaluate the follow-up effects for RT+ES training.