Influence of biological, experiential and psychological factors in wine preference segmentation

Abstract

Background and aims: We sought to determine the influence of selected biological, experiential and psychological variables on self-reported liking and consumption of wine in a sample of 329 Ontario wine consumers.

Methods and results: Cluster analysis revealed three distinct groups, representing plausible market segments: wine lovers; dry table wine likers/sweet dislikers; and sweet wine likers/fortified dislikers. These groups differ in level of wine expertise, wine adventurousness, alcohol intake, bitterness from 6-n-propylthiouracil (PROP), and several demographic variables. PROP hypo-tasters (n=113) and PROP hyper-tasters (n=112) differed in liking scores for nine of the 11 wine styles [ANCOVA, P(F)<0.05]. When wines were grouped according to their dominant sensory properties (dry, sweet, carbonation and heat), liking scores for PROP hyper-tasters were higher than those of PROP hypo-tasters for all classes. Scores also varied with age, expertise and gender for some products. Effect sizes (eta-squared) were generally greatest for age, and those for PROP responsiveness were of similar magnitude as those for gender. As expected, wine consumption frequency was higher for men and experts, and increased with age.

Conclusions: Age is the most robust and consistent driver of wine liking and intake of the variables examined. Taste phenotype also contributes significantly to variation in wine liking.

Significance of the study: Ontario wine consumers fall into one of three wine liking clusters, which differ in experiential, biological, psychological and demographic features that can be targeted through branding and marketing strategies.

Keywords: Ontario; PROP; alcohol consumption; expertise; market segmentation; propylthiouracil; taste genetics; wine liking.

Figure 1

Frequency distribution of 6-n-propylthiouracil (PROP) scores. Arrows indicate lower and upper tertiles.

Figure 2

Profile plot from cluster analysis of wine liking scores showing three groups: cluster 1 (■); cluster 2 (■); and cluster 3 (□).[K-means clustering – criterion: Trace (W), after agglomerative heirarchical clustering (dissimilarity proximity type, Ward's method, Euclidean distance)].

Figure 3

6-n-Propylthiouracil (PROP) responsiveness and wine liking for a range of wine types by PROP responsiveness [PROP hyper-tasters (□) and PROP hypo-tasters (■)]. Data shown are mean liking scores (7-point hedonic scale) ± standard error. Means significantly different (ANCOVA – model in text) at alpha of 0.05 (*), 0.01 (**) or <0.001 (***).