The way we measure values is foundational to their study. This paper explores the empirical findings and theoretical discussions in the literature concerning an essential quality for any measure of values, its validity. We discuss an important debate on validity within the literature concerning the ability of value measures to predict value-related behaviour. We interrogate a key assumption behind the value-behaviour link, namely that a participant's responses on a valid value measure will predict their behaviour. Ultimately, we make the case that a better understanding of the relationship between values and behaviour may be possible if we also consider social norms.