Fig. 2From: The impact of unhealthy food sponsorship vs. pro-health sponsorship models on young adults’ food preferences: a randomised controlled trialPredicted proportion of young adults with awareness of sponsor brands by sponsorship condition. Notes: † reference category for logistic regression analyses; * p < 0.05; ** p < 0.01; *** p < 0.001. Logistic regression analyses included product category as a covariateBack to article page