Sensory affective test methods have been widely used to compare the consumer appeal of different products in sensory studies. In a?ective testing, it is common to include a number of repeated measures on the consumer appeal. The conventional data analysis approach is to conduct hypothesis testing on individual measures separately, which may result in contradicting conclusions and confusion in decision-making. In this paper, we propose a Bayesian hierarchical model for integrated analysis of repeated measures on consumer appeal in a?ective testing. The Bayesian model implements data transformation to transform the measures on the same scale, introduces the random e?ect for consumer panelists to allow variation among panelists, and most importantly, combines the repeated measures to borrow information to produce more consistent results in a?ective testing. We have used a number of real sensory datasets to demonstrate the performance of the Bayesian model.