Thesis (M.A. in sociology)--Washington State University, May 2009. / Title from PDF title page (viewed on May, 29, 2009). "Department of Sociology." Includes bibliographical references (p. 67-72).

The contingent valuation method has been used with increasing frequency to estimate the value of nonmarket goods such as wildlife. Increasing use of and demand for the method has lead to the ~) increased use of mail surveys to collect value information. Themail format, however may place limitations the method's ability to generate acceptable value estimates. This study examined the effects of various amounts of information in the hypothetical market created in mail contingent valuation surveys. The problem setting used to examine the information problem was the current controversy surrounding the pesticide carbofuran. This pesticide has some negative impacts on wildlife, most noticeably the bald eagle. Within this context, four different mail surveys were designed and mailed to Virginia residents. Each survey contained different amounts of information on the affects of carbofuranon wildlife. The impact of information on response rates, willingness to pay bids, and aggregate benefit estimates was then examined. The fllldings suggested that mail surveys with lengthy and complex contingent scenarios may tend to be less effective for commodities where respondents have had previous experiences analogous to those presented in the hypothetical market. / Master of Science

No

published_or_final_version / Business Administration / Master / Master of Business Administration

This paper provides probit estimates of willingness to pay for Snake River salmon recovery projects using the contingent value survey method. Two types of survey method are tested, the direct question format and the randomized response format. Additionally, both an increase in electric bill payments and volunteer time are offered as payment vehicles. The results suggest that there is not much difference in estimates of willingness to pay between the two survey formats. They also suggest that for the Oregon State University staff, there is substantial willingness to pay through electric bill increases but not in volunteer time. By guaranteeing the respondents anonymity, the randomized response format is hypothesized to encourage more truthful answers, thereby reducing yea-saying bias. The results indicate that yea-saying may be reduced for those who contribute to environmental causes but overall does not affect willingness to pay. Other studies using the randomized response format that show a reduction in yea-saying and willingness to pay were done in classroom settings while this survey was a mail survey. There is an indication that the privacy and anonymity of the mail survey is enough to mitigate social pressure and yea-saying as opposed to a classroom setting. This suggests that the randomized response format may be more appropriate for group surveys or personal interview surveys. The result that there is substantial monetary support in the form of an increase in electric bills but not in the form of volunteer time payments may be due to the sample being fully employed. Also Snake River salmon restoration projects may seem geographically distant to this sample so volunteer activities may seem unrealistic. / Graduation date: 1998

Thesis (M.B.A.)--University of Hong Kong, 1981.

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.