An Alternative Method for Relating Consumer and Descriptive Data Used to Identify Critical Flavor Properties of Mango (Mangifera Indica L.)1

An alternative method for measuring response and analyzing data in consumer research was explored in this study which attempted to identify critical flavor properties of mango. Twelve mango samples were evaluated according to 13 flavor descriptors by a trained panel using 150‐mm unstructured lines. Consumers (n = 120), intercepted in supermarkets, each tasted 3 samples and responded ‘yes’ or ‘no’ to an intent‐to‐purchase question. Consumer response, coded by respectively assigning 0 or 1 to each negative or positive answer, was expressed as a probability of purchase and related to descriptive attributes using logistic regression. Critical descriptors were peachy, grassy, and green banana (α= 0.10). The function generated had an index of discrimination of 0.70. Although limitations in flavor variations and possible segmentation among the consumers studied restricted the usefulness of the acceptability function, using a probability of purchase as an index of acceptability and logistic regression to relate consumer response with product attributes was demonstrated.