Are preferences stated in web vs. personal interviews different? A comparison of willingness to pay results for a large multi-country study of the Baltic Sea eutrophication reduction

We investigate the prevailing view in the stated preference literature that the data collection mode does not significantly affect the value estimates. Based on data from Computer-Assisted Web Interviews and Computer-Assisted Personal Interviews aimed at assessing the social benefits for Poland from meeting the nutrient load reduction targets defined in the HELCOM’s Baltic Sea Action Plan (2007), we find that the value estimates obtained from the two modes differ significantly. This evidences the existence of a “pure” mode effect as we control for socio-demographic differences between the web-interviewed and personally-interviewed samples by weighting the observations. The relative difference in the derived values between the two modes is used to update the estimates of the economic values of reducing nutrient loadings to the Baltic Sea provided by Ahtiainen et al. (2014) for every Baltic Sea country. In addition to controlling for the mode effect (as different, web and personal, modes were used in different countries), we examine 18 alternative model specifications to find the distribution that captures best the payment-card willingness-to-pay responses. Overall, our study illustrates the extent of the impact that the choice of a data collection mode can have on valuation results.


Issue Date:
Jun 21 2017
Publication Type:
Conference Paper/ Presentation
Language:
English
Total Pages:
16




 Record created 2017-07-05, last modified 2017-09-25

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)