Modeling Aggregate Choice for Form and Function Through Metaconjoint Analysis

Abstract

In the previous work, metaattributes have been used to model the relationship between two groups of disparate product attributes. There, preference for form, function, and the relationship between the two were modeled for individual consumers. However, this approach is limited as designers are often called on to choose a design that best appeals to a group of consumers, not individuals. This work expands on the concept and makes it more generally applicable by adapting metaconjoint to model aggregate choice for consumer groups. The results from this work show that a metaconjoint approach can be used to model aggregate choice for form and function and can yield better results on holdout sample predictions than form or function alone.