Impaired strength adversely influences an older person's ability to perform activities of daily living. A cross-sectional study of 117 independently living men and women (age = 73.4 ± 9.4 year; body mass index (BMI) = 27.6 ± 4.8 kg/m²) aimed to assess the association between body composition and: (1) upper body strength (handgrip strength, HGS); (2) lower extremity performance (timed up and go (TUG) and sit to stand test (STS)); and (3) endurance (6-minute walk (SMWT). Body composition (% fat; lean body mass (LBM)) was assessed using bioelectrical impedance. Habitual physical activity was measured using the Minnesota Leisure Time Physical Activity Questionnaire (MLTPA) and dietary macronutrient intake, assessed using 24 h recalls and 3-day food records. Regression analyses included the covariates, protein intake (g/kg), MLTPA, age and sex. For natural logarithm (Ln) of right HGS, LBM (p < 0.001) and % body fat (p < 0.005) were significant (r² = 46.5%; p < 0.000). For left LnHGS, LBM (p < 0.000), age (p = 0.036), protein intake (p = 0.015) and LnMLTPA (p = 0.015) were significant (r² = 0.535; p < 0.000). For SMW, % body fat, age and LnMLTPA were significant (r² = 0.346; p < 0.000). For STS, % body fat and age were significant (r² = 0.251; p < 0.000). LBM is a strong predictor of upper body strength while higher % body fat and lower physical activity are associated with poorer outcomes on tests of lower extremity performance.