Design and Evaluation of Mail Surveys to Estimate Deer Harvest Parameters

Parameter estimates derived from periodic harvest surveys are often a critical component of management strategies for game populations. Knowledge of the accuracy and precision of such estimates is therefore prudent. We conducted a survey of the applicants for 1996 antlerless white-tailed deer (Odocoileus virginianus) tags in the Piedmont physiographic region of South Carolina to evaluate whether the harvest characteristics of this subpopulation of hunters were representative of all hunters in the Piedmont region. Comparison of these estimates to estimates derived from a concurrent and independent mail survey of all licensed deer hunters in the state revealed that antlerless-tag applicants harvested more deer and hunted more days than nonapplicants. We also conducted telephone surveys of nonrespondents to estimate nonresponse bias of mail survey estimates. Telephone survey results indicated that nonresponse bias in mail survey estimates was generally minimal, except for a positive bias in doe-harvest estimates. Use of responses from follow-up mailings to correct for nonresponse bias is not recommended. Sample size recommendations are provided for achieving a desired precision of county estimates of sex-specific harvest in states with hunting conditions similar to South Carolina's.