• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • Tagged with
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Optimal economic design of mail surveys: influences on response rates and the impact of responses to a second mailing

Gregory, Alexandra January 1900 (has links)
Doctor of Philosophy / Department of Agricultural Economics / John A. Fox / Mail surveys are used to gather information in order to make inferences about populations. This study examines cost effective methods to maximize response rates to a mail survey. A consumer mail survey was developed and used to study consumer demand for safer foods. In addition, the study will verify if additional responses from follow-up mailings changes econometric analysis results, thus validating follow-up mailing costs. A test was created to maximize response rates and incentives were used in both mailings. An Ordinary Least Square (OLS) model was developed to analyze response rates and mail survey costs while a willingness to pay model and an ordered probit were used for the consumer demand analysis. Results showed that when a consumer survey was sent to a city within the school's state no incentive should be included while if sent to a city outside the school's state an incentive should be included. Moreover, if the outcome from the first mailing resulted in a low response rate a monetary incentive should be included in the follow-up mailing to increase response rates. Results from the consumer willingness to pay for irradiated salad greens showed that consumers are willing to pay higher prices for irradiated salad greens. Furthermore, results showed that there is no statistical difference between the coefficients, in both the willingness to pay and the ordered probit, from the model using additional observations from follow-up mailing and the model that included only observations from the first mailing. Even though coefficients were not statistically different in the ordered probit, significance of the marginal effects for some variables were different between models.

Page generated in 0.0876 seconds