Handgrip strength (HGS) is an important indicator of health. Because HGS is strongly associated with body size, most investigators normalize HGS for some measure of body size as a more sensitive indicator of strength within a population. We aimed to 1) identify the optimal body size dimension to remove (normalize) HGS for differences in body size among adults and 2) generate norm-referenced centiles for HGS using the identified body size dimension. Data were from the National Health and Nutrition Examination Survey, a representative sample of the US noninstitutionalized civilian population. Exclusions resulted in a final sample of 8690 adults 20 yr and older. HGS was measured using handheld dynamometry. Body size dimensions included body mass, height, and waist circumference. The most appropriate dimension(s) associated with HGS is identified using allometry. We fitted centile curves for normalized HGS using the generalized additive model for location, scale, and shape. Findings suggest that neither body mass nor body mass index is appropriate to normalize HGS. Incorporating all three body size dimensions of body mass, height, and waist circumference, or the reduced subsets of body mass and height, or height alone, suggests that the most appropriate normalizing (body size) dimension associated with HGS should be a cross-sectional or surface area measure of an individual's body size (i.e., L2, where L is a linear dimension of body size). Given that height was also identified as the signally best body size dimension associated with HGS, we recommend HGS be normalized by height2 (i.e., HGS/HT2). Centile curves for HGS/HT2 by age group and gender were therefore provided. Scaling adult HGS by height2 may help normalize strength for population-based research.