Monday, October 20, 2008

Contingent Valuation Data from env-econ

The data set put up on Environmental Economics is a useful resource for students considering doing an MSc dissertation or extended essay with an interest in contingent valuation. Second years will study this next semester.

Clearly Haab and Whitehead have milked this data dry but that means it gives students an excellent opportunity to try to replicate their results and to try and work out why there are differences (if any).

This is a valuable resource for students, hence this blog post.

1995 Albemarle-Pamlico Sounds Economic Survey

Data Description

In General...

The data for this study is from a 1995 telephone survey conducted by the East Carolina University Survey Research Laboratory. The survey used a random digit dialing sampling scheme. The sample was purchased from Survey Sampling, Inc. and interviews were computer assisted. Of the households that were contacted, 1077 respondents provided data for an overall response rate of 75%.

The demographic profile of the sample is similar to that of Eastern North Carolina. The sample is 43% male, 54% married, 65% white, and 47% are employed full-time. The median age of the sample is 42 and the median education level is 13 years. Household income was elicited in categories. Twelve percent of households earned less than $10,000 and between $10,001 and $15,000, 11% of households earned between $15,001 and $20,000, 14% of households earned between $20,001 and $25,000, 10% earned between $25,001 and $30,000, 24% earned between $30,001 and $50,000, 12% earned between $50,001 and $75,000 and 5% earned above $75,001. With income levels coded at the midpoints of the income ranges (the upper range was coded as $85,000) the mean and median household income is $31,550 and $27,500. Except for race, none of the demographic variables are different between the P and A-P versions of the survey. The A-P sample has more white respondents than the P sample ( 2=3.94[1 df]).

The TRAVCOST variable was constructed as follows. For respondents living west of the Pamlico River, the distance was calculated as the distance from the respondent's county population center to Washington, NC on the Pamlico River. If the respondent lived North or South of the Pamlico River, the distance was calculated as distance from the county population center to the nearest boat ramp on the Sound. Distances were calculated using the Automap software package. The travel cost used is $.20 per mile, average miles per hour is 50, and the opportunity cost of travel time is valued at the wage rate (wage=INCOME/2080). We assume that all trips are day trips so the cost of on-site time is zero.

There are two main versions of the telephone survey. Version 1 contained a contingent market for the Pamlico Sound and Version 2 contained a contingent market for the Albemarle and Pamlico Sounds. The main difference in the two versions is the insertion of "Albemarle and" before Pamlico in all questions and the addition of the plural to Sound(s).

Without looking at the link try and work out why this data might be useful. What questions could be asked?


No comments: