@article{2018:steiner:a_users_g,
title = {A User's Guide to the Galaxy of Conjoint Analysis and Compositional Preference Measurement},
year = {2018},
note = {Conjoint analysis is often seen as one of the most important
methodological developments of marketing research. Various textbooks
give an introduction to conjoint analysis and alternative preference
measurement methods. Applying variants of conjoint analysis has become
less effortful because commercial software is available that makes
setting up a conjoint survey and analysing the data very easy.
Nevertheless, researchers and practitioners have to make numerous
decisions how to design their conjoint studies, such as how to select
the attribute and levels when planning their first conjoint study.
Decisions on these important ‘craft’ factors will substantially
influence the reliability and validity of the preference measurement
results and are therefore of key importance to researchers and
practitioners who lack sufficient experience using conjoint analysis.
Available textbooks rarely discuss these ‘craft’ factors in much
detail. Thus, the main goal of this paper is to bring these
‘craft’ factors into focus and sensitize researchers and
practitioners towards potential pitfalls that could negatively affect
the validity of their preference measurement studies.To achieve this
goal, the paper first introduces conjoint analysis and explains its’
key value for research and practice. We then explicate commonly used
and more recently developed decompositional, hybrid, and compositional
preference measurement approaches and discuss advantages and
disadvantages in different decision contexts. Our objective is to
guide the potential user in the selection process to an appropriate
approach. The second objective of the paper is a discussion of
additional important ‘craft’ factors. We deliberate on how
researchers can generate and select attributes and levels for their
preference measurement studies as well as how they can explain them to
respondents. Our third objective is to provide users with more
practical guidance how to interpret preference measurement results.
Moreover, we aim to enable users to assess the quality of their
conjoint data with respect to reliability, validity and applicability
measures. We finally also show how market simulations can be set up
and how the external validity can be measured. In sum, our paper,
thus, serves as a user’s guide to the ‘galaxy’ of consumer
preference measurement, and in particular conjoint analysis.},
journal = {Marketing ZFP},
pages = {3--25},
author = {Steiner, Michael and Meißner, Martin},
volume = {40},
number = {2}
}