Principal component analysis (PCA) is a data reduction technique that is increasingly popular within biomechanics. However, a majority of the literature that has used PCA has employed a methodology that is less strongly aligned to the philosophical basis of the technique, and that may not yield the most practically interpretable results. In this technical report I exemplify the two approaches to PCA that have been used in the biomechanics literature, and argue that one method may be superior to the other for the majority of biomechanical applications.