Development and Analysis of Joint Internet–Telephone Hunter Surveys

Abstract

Wildlife management agencies survey hunters to estimate harvest rates and opinions. Telephone surveys have been considered an effective survey tool for minimizing non-response bias. Recent changes in human demographics and telephone technology such as caller identification and cellular telephones have been linked to reduced response rate, increased likelihood of non-response bias, and increased cost of surveys. These changes have resulted in an interest in using the Internet as a survey tool. There are complications associated with the use of Internet surveys. Analysis methods for combining data from telephone and Internet surveys are developed. An example survey of duck hunter season preferences is presented to demonstrate this method. People providing e-mail addresses to the Colorado Division of Wildlife are a non-random sample of hunters. The survey is used to show there is a difference in estimates from the e-mail and telephone respondents and that corrections for the differences can be made.

Keywords:

• harvest survey
• interactive voice response
• e-mail survey
• nonresponse bias
• survey automation
• survey design

The author thanks J. Gammonley, T. Remington, and M. Lloyd for their efforts in implementing the duck season preference survey and Corybant, Inc. for conducting the survey. T. Shenk, M. Alldredge, J. Vaske, and two anonymous referees provided comments on the manuscript.

Notes

SAS Institute Inc. (2006). SAS 9.1. Cary, NC.