Should Reference Alternatives in Pivot Design SC Surveys be Treated Differently?

Abstract

Analysts are increasingly making use of pivot style Stated Choice (SC) data in the estimation of choice models. These datasets often contain a reference alternative whose attributes remain invariant across replications for the same respondent. This paper presents evidence to suggest that respondents react differently to the attributes of these reference alternatives and those of purely hypothetical alternatives. While some such evidence exists in the existing literature, this paper goes further and details a number of different departures from a common treatment of the two types of alternatives, relating both to the observed part of utility and the error term.

Do Respondents’ Perceptions of the Status Quo Matter in Non-Market Valuation with Choice Experiments? An Application to New Zealand Freshwater Streams

“…To overcome such limitations, Scarpa et al [12,39] proposed the use of error components (MXL-EC) in which the additional variance of utility of alternatives different from the SQ can be identified. Since their application, numerous other studies have found the MXL-EC to be better suited in capturing the SQ effects than the conditional logit and nested logit frameworks, and even MXL models without error components [13,15,[39][40][41][42][43]. Within the MXL-EC framework, the SQ effect on the systematic component of utility can be measured by the ASC, while the effect on the stochastic component of utility can be captured by introducing a common error component shared by the utilities associated with alternatives different from the SQ, which takes account of the correlation patterns and increased error variance due to the conjectural nature of the experimentally designed alternatives.…”

“…Hess and Rose [13,14] categorized the SQ alternatives into three formats as follows: "Firstly, … the presence of a status quo alternative which is represented as a null alternative with the attributes and attribute levels of the alternative not shown as part of the experiment. A second form of these experiments involves respondents being shown alternatives with attribute levels based on their own experiences but not the exact levels as described.…”

The Demand for Index-Based Flood Insurance in a High-Income Country

“…Thereby, we allow the two insurance types to be correlated in unobserved factors. This relaxes the well-known IIA assumption of standard logit, and thus might represent a more realistic substitution pattern, in particular in the presence of the no insurance alternative (e.g., Scarpa et al 2005;Hess and Rose 2009). In addition, we specify the insurance premium coefficient to be lognormally distributed.…”

What comes first, origin or production method? An investigation into the relative importance of different attributes in the demand for eggs

“…Due to this, the utilities of the designed alternatives are likely to be more correlated between them than with the nonbuy option and to have a higher variance than the utilities of the non-buy alternative. To take this into account we consider that the experimental designed alternatives share an extra error component, which is missing in the utility of the experienced alternative (Scarpa et al, 2007b) thus estimating an additional error component in the mixed random parameter logit model (Error Component Random Parameter Logit ECRPL) (Scarpa et al, 2005) that has been used in several empirical applications (Campbell, 2007;Scarpa et al, 2007bScarpa et al, , 2008Hess & Rose, 2009;Jacobsen & Thorsen, 2010). This model has been very successful due to its parsimoniousness (it only requires one extra parameter) and it has empirically been found to substantially improve model fit.…”