We present a dataset on participants’ attitudes towards the consumption of meat (N = 30). Participants were presented with a baseline questionnaire entailing 11 statements. After a baseline measurement, we perturbed the participant’s opinion on one of the 11 items, after which the participant completed the same questionnaire. By repeating this procedure for each of the 11 items, we measured to what extent the perturbation changed the participant’s baseline score. In addition, we asked participants to draw the influence of a specific item onto the other items in a network format. The data are suitable for various purposes, like causal inference and the malleability of attitudes.