Understanding the Order Effects in Sequential Monadic Product Tests
This document details an analysis of 22 product tests in which two products have been evaluated by consumers in each study. The objective of the study was to understand if the adoption of sequential designs improves data sensitivity and discrimination in comparison to a monadic design. It also examines how order effects can be exploited to improve recommendation in product development. In addition, a “second position effect” is revealed which can be further utilized to improve product testing results.
Introduction
In the market research industry, it is a wide-spread assumption that a sequential monadic design can improve data sensitivity and discrimination in product testing (e.g. Gacula 1987; Komanska, 1989) in comparison to a pure monadic design. The reasoning behind the assumption is that such a design, for instance, enables consumers to make direct comparisons testing the products after each other, allowing a more critical comparison than having separate consumers testing single products, i.e., monadic. However, critique to sequential monadic designs surface often with respect to order effects (Welch and Swift, 1992; Friedman and Schillewaert, 2012). Order effects are caused by context or the environment in which the product is evaluated. When a product is seen and tested in the first position, a respondent’s context may be the last usage experience with other products in the same category. When seen and evaluated in the second position, the context may well be the product seen and tested first. To take account of the order effect, the
order of presentation is commonly rotated such that each product is presented equally often in each ordered position (i.e., seen first or second) (see, e.g., McBurney and White, 2009).
Customer Experience / Kundenerleben