This study aimed to determine whether the treatment effect differs for patients with stroke who perform robot-assisted upper-extremity rehabilitation by themselves compared to those whose rehabilitation is actively assisted by a therapist. Stroke patients with hemiplegia were randomly divided into two groups and received robot-assisted upper-limb rehabilitation for four weeks. In the experimental group, a therapist actively intervened in the treatment, while in the control group, the therapist only observed. After four weeks of rehabilitation, the manual muscle strength, Brunnstrom stage, Fugl-Meyer assessment of the upper-extremity (FMA-UE), box and block test, and functional independence measure (FIM) showed significant improvement in both groups compared to that before treatment; however, no interval change in spasticity was noted. The post-treatment values showed that the FMA-UE and box and block tests were significantly improved in the experimental group compared to those in the control group. Comparing the changes in the pre- and post-treatment values, the FMA-UE, box and block test, and FIM of the experimental group were significantly improved compared to those in the control group. Our results suggest that active intervention by therapists during robot-assisted upper-limb rehabilitation positively impacts upper-extremity function outcomes in patients with stroke.