Considering the target population in the anthropometric design is crucial to avoid the occurrence of musculoskeletal disorders. In Mexico, there is no evidence that multivariate analysis has been conducted in the northwest female population, so the univariate traditional percentile approach is still used. Univariate and multivariate (boundary) approaches were compared when designing a sitting workspace. Eight anthropometric dimensions (AD) were included in the analysis. A total of 602 females participated in the survey. It has been theoretically demonstrated that multivariate approaches obtained higher accommodation percentages than the univariate approaches for the Mexican female population when considering the accommodation of the total of the anthropometric dimensions simultaneously. The models obtained can help in the sitting workstation design.