[Purpose] The aim of this study was to evaluate the intrarater and interrater reliabilities of a toe grip dynamometer. [Subjects] The subjects were 180 community-dwelling people and 20 university students. [Methods] We assigned 180 individuals to three groups based on age to determine the intrarater reliability. The groups consisted of young (age, 20 to 39 y), middle-aged (age, 40 to 59 y) and older adults (age, 60 to 79 y). Interrater reliability was investigated using 20 university students as subjects. We calculated intraclass correlation coefficients to assess the intrarater and the interrater reliability. The intrarater reliability was assessed for each group by gender. We calculated intraclass correlation coefficients for the interrater reliability by comparing the first measurements made by two testers. The Bland-Altman analysis was used to assess fixed and proportional bias. [Results] The intrarater reliability showed a substantial to almost perfect agreement in male and almost perfect agreement in female subjects. For the intrarater reliability, a fixed bias was found in most measurements, but proportional bias was not found at all. The interrater reliability showed almost perfect agreement. Fixed bias and proportional bias were not found for the interrater reliability. [Conclusion] The intrarater and the interrater reliabilities of the toe grip dynamometer were substantial, indicating its suitability for clinical use.