Muscular strength and endurance are important for individuals with mental retardation (MR) because they are integral parts of daily living and recreation skills. They also are more like to have employment opportunities that require physical labor. Although a few studies have focused on muscular strength, there is limited information on muscular endurance of individuals with MR. The lack of research may be related to the difficulty of accurately assessing muscular strength and endurance. Considering that large within-individual variability and lack of motivation are often observed in persons with MR, it is important to evaluate the appropriateness of assessing muscular strength and endurance in this population. Therefore, the purpose of this study was to examine the reliability of muscular strength and endurance assessments in adults with MR using Generalizability theory (G-theory). G-theory is a psychometric theory that not only estimates overall reliability but also evaluates possible sources of error. Nineteen individuals with MR participated in the study. Four separate testing sessions were conducted within a four week period. Muscular strength and endurance were measured with a hand-grip dynamometer by two raters. Each rater tested participants on two different days. For muscular strength, the participants were asked to squeeze the dynamometer as hard as they can three times during each testing session. A 3 random facet (raters, trials, and days) in completely crossed design revealed that the largest source of variance was due to differences between the participants (92%). Among the sources of error variance, 3% was associated with “participant-rater” interaction with overall G-coefficient of .98. Muscular endurance was evaluated with the ratio between the maximum muscular strength and muscular strength after a one minute squeeze of the hand-grip dynamometer as a muscular endurance index. A 2 random facet (raters and days) in completely crossed design indicated that the largest error variance was due to “participant-day” interaction (36%) followed by unexplained variance (32%) and “rater-participant” interaction (32%). The results indicate that some participants systematically performed differently on muscular endurance on different days. Also, the rater had different influence on different participants' muscular endurance performance. Based on the results of this study, we conclude that using a hand held dynamometer can consistently estimate maximum muscular strength, but a new innovative measurement technique is needed to estimate the muscular endurance of individuals with MR. Keyword(s): adapted physical activity, measurement/evaluation, research