An Evaluation of Attribute Anchoring Bias in a Choice Experimental Setting
FCN Working Paper No. 6/2013
Posted: 2 Feb 2014
Date Written: April 2013
We investigate consumer valuations of technical, home-related energy-saving measures (ESMs) and their attributes through a choice experiment. Unlike previous literature we expect heuristic decision-making, as decisions about ESMs are complex. Specifically, in our experimental design we focus on the effects of attribute anchoring through inclusion of an innovative element. We investigate anchoring-and-adjustment (A&A) in a choice-experimental framework, expecting it on attributes. Our question is: does A&A lead to a correct elicitation of the overall valuation or are adjustments insufficient, leading to attribute anchoring bias? To evaluate this, our participants can read attribute information only after clicking on covering buttons. Clicking data reveal what information they use in what order, which provides objective perceived ESM-attribute importance measures to be included as control variables in the estimations. For the experiment we perform a representative online survey among 1084 German residents. Mixed logit analysis cannot show that ESM valuations are affected by attribute anchoring bias, which indicates accuracy of our estimations, and through inference of those in other choice experiments. We also report statistics on ESM-attribute valuations differentiated by e.g. demographics, which is relevant for suppliers and policy-makers alike.
Suggested Citation: Suggested Citation