Identifying Inconsistent Responses in Stated Choice Surveys Using a Dominance-Based Approach

Surveys focusing on choice behaviour, and in particular, Stated Preference (SP) surveys are widely used in studies across many disciplines, including marketing research, energy forecasting, environmental studies, labour economics and transportation to model and predict people’s choices. There are several econometric modelling approaches that can be used to analyse data resulting from these surveys. All of them are based on assumptions on how individual respondents make their choices. One of the most important assumptions is that respondents are consistent, i.e. they don’t change their behaviour across choice tasks. However, this assumption can be violated due to the complexity of choice experiments or respondent fatigue. There is an increasing interest in detecting inconsistent behaviour and the impact such behaviour may have on estimated models. This research aims to extend previous work on identifying inconsistent behaviour to the case of more complex experiments. A systematic test procedure to detect inconsistent behaviour, that derives individual dominance decision rules is developed and applied to three SP datasets. After identifying inconsistent behaviour in respondents, the impacts of the removal of these respondents on model estimation results are examined. The authors' analysis using three different datasets shows that inconsistent choices commonly occur in SP surveys and removal of inconsistent respondents can result in substantially different coefficients in estimated models.

• 
• Supplemental Notes:
  • This paper was sponsored by TRB committee ABJ40 Travel Survey Methods.
• Corporate Authors:

  Transportation Research Board

  500 Fifth Street, NW
  Washington, DC  United States  20001
• Authors:
  • Rezaei, Ali
  • Patterson, Zachary
• Conference:
  • Transportation Research Board 94th Annual Meeting
  • Location: Washington DC, United States
  • Date: 2015-1-11 to 2015-1-15
• Date: 2015

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References; Tables;
• Pagination: 15p
• Monograph Title: TRB 94th Annual Meeting Compendium of Papers

Subject/Index Terms

• TRT Terms: Choice models; Data cleaning; Data files; Stated preferences; Travel behavior; Travel surveys
• Uncontrolled Terms: Inconsistencies
• Subject Areas: Highways; Planning and Forecasting; Public Transportation; I72: Traffic and Transport Planning;

Filing Info

• Accession Number: 01558295
• Record Type: Publication
• Report/Paper Numbers: 15-5206
• Files: PRP, TRIS, TRB, ATRI
• Created Date: Mar 30 2015 12:22PM