Learning and Fatigue Effects Revisited: The Impact of Accounting for Unobservable Preference and Scale Heterogeneity on Perceived Ordering Effects in Multiple Choice Task Discrete Choice Experiments

Abstract

Using multiple choice tasks per respondent in discrete choice experiment studies increase the amount of available information. However, treating repeated choice data in the same way as cross-sectional data may lead to biased estimates. In particular, respondents’ learning and fatigue may lead to changes in observed utility function preference (taste) parameters, as well as its error term variance (scale). Substantial body of empirical research offers mixed evidence in terms of whether (and which) of these ordering effects are observed. In this study we point to a significant component in explaining these differences – we show how accounting for unobservable preference and scale heterogeneity can influence the magnitude of observed ordering effects, especially if combined with too few choice tasks used for the analysis. We do this by utilizing the state-of-the-art modeling methods (H-MNL, S-MNL, HRPL, G-MNL) which we modify to accommodate choice task specific scale parameter. In addition, we investigate possible bias resulting from not accounting for ordering effects. Our empirical study was based in the context of environmental protection – management changes in the protection of Polish forests.