Social norms are argued to be important drivers of social behaviour and solutions to societal problems. A fundamental part of working with social norms for either purpose is measuring them. Yet, this is not simple. While there are accepted incentivised tools for measuring social norms in the laboratory, these are costly and restrictive in the populations they reach thereby limiting their research and policy use. Conversely, non-incentivised survey measures have been more widely used because of their simplicity, yet, these have not been validated and therefore it is unclear what they actually measure. Here we aim to address these issues by developing a behaviourally validated survey measure of social norms that is non-incentivised, simple, and portable across societies. We do this by eliciting behaviour in various settings, identifying social norms with the standard incentivised measures, and then pitting non-incentivised measures in a “horse race” against each other. Our validation will vary location (Italy and India), sample (students, non-students, and nationally representative), and elicitation method (laboratory, lab-in-the-field, online). Once the validation process is completed, we will implement the best-performing survey measure in a survey to identify gender restrictive norms (e.g. women’s freedom to work outside the home) in India and their key predictors