Telephone presurveys, self-selection, and non-response bias to mail and Internet surveys in economic research

Abstract

A telephone presurvey is used in conjunction with mail and Internet follow-up surveys to assess self-selection and item nonresponse bias. Findings suggest that self-selection is present, but item nonresponse bias is limited. The Internet version exhibited no item nonresponse bias.

Acknowledgements

The authors thank Stan Spurlock for useful comments on an earlier draft of this paper. This research was supported by funding from the Advanced Spatial Technologies in Agriculture grant program, Mississippi State University.

Notes

¹ It is important to note that self-selection bias in answering the phone survey is not being tested. This test only represents potential self-selection bias in the follow-up survey.

² Agreement to participate in the follow-up survey did not guarantee that the individuals would actually respond.

³ The median WTP value for those agreeing was US$23, while the median WTP for those not agreeing was US$15, indicating that the midpoints of the distribution were also different as well.