Outlier and Protest Responses 



In the analysis of CVM responses there are two groups of 

 respondents who should be excluded from the sample before any 

 analysis occurs. The first is that group who indicate a 

 willingness to pay the stated bid amount but who would not 

 actually be able to pay that amount given their income. The 

 standard economic definition of demand requires both a 

 willingness and an ability to pay. Therefore those respondents 

 who indicate a willingness but lack the ability to pay the bid 

 amount must be excluded as their response does not meet the 

 constraints of economic theory. Ability to pay was determined by 

 first calculating the percentage of their income which 

 respondents were willing to spend on deer hunting. This was done 

 as follows: 



PERCENT = ( (TOTAL + BID) * TRIPS ) / INCOME 



Where: 



TOTAL = The amount they reported spending on their most recent 



trip. 

 BID = The dollar bid level asked. 

 TRIPS = The number of separate deer hunting trips they reported 



taking this season. 

 INCOME= Their reported annual income. 



This percentage statistic was calculated for each of the four CVM 

 questions. As an initial measure all respondents with a 

 percentage figure greater than 1 were excluded since this group 

 most obviously lacks the ability to pay. The percentages for the 

 remaining respondents were then tabulated for each question 

 giving the following results. 



PERCENT (Quest. 1) 



PERCENT (Quest. 2) 



PERCENT (Quest. 3) 



PERCENT (Quest. 4) 



Since the distribution of the calculated variable PERCENT was 

 somewhat skewed rather than distributed normally a three standard 

 deviation confidence interval was placed around the four 

 calculated means in order to determine the cutoff limit for 

 outlier exclusion. In total, 106 observations were eliminated 

 from the following economic analysis due to a reported 

 willingness to pay which exceeded the cutoff limits. 



The second group of respondents who were excluded from the 

 analysis were those whose responses reflected a "protest" to some 

 aspect of the simulated market. The U.S. Water Resources Council 

 has suggested that a followup question be asked to each CVM 



15 



