Detecting response styles by using dual scaling of successive categories
A response style denotes a certain mapping of latent preferences to a rating scale that is common among a certain group of individuals. For example, individuals from the same country may assign high ratings to the majority of objects regardless of the specific preferences for the objects. The existence of response styles causes problems in international and cross-cultural research as it makes it hard to compare findings. Moreover, even within homogeneous samples, response styles make it difficult to expose the underlying preference structure. Detecting the existence and influence of a response style is typically a difficult issue as the underlying preferences are not directly observable. Hence, we can never be sure if the observed ratings are the result of a response style or an adequate representation of the preferences. In this paper, we consider the use of dual scaling as a tool to detect the existence of a response style. By means of a simulation study, we assess the performance of the proposed method.
|Econometric Institute Research Papers|
|Report / Econometric Institute, Erasmus University Rotterdam|
|Organisation||Erasmus School of Economics|
van de Velden, M. (2007). Detecting response styles by using dual scaling of successive categories (No. EI 2007-41). Report / Econometric Institute, Erasmus University Rotterdam. Retrieved from http://hdl.handle.net/1765/10561