selected users of the different recreation areas. It is 

 important that reliable total visit statistics be ob- 

 tained for each existing area being investigated. 

 This can usually be done satisfactorily with judi- 

 cious use of traffic counters at most water-based 

 recreation areas. If totals are collected throughout 

 the season, samples for questionnaires or inter- 

 views need be drawn only on a few days — on both 

 weekends and weekdays, as patterns are likely to 

 vary greatly between them. 



(3) The number of questions asked may also be 

 limited. The major concerns are the origin and pur- 

 pose of the trip and limited information about the 

 users. A representative range of areas, facilities, 

 and locational proximities should be covered in 

 such surveys. Fully adequate methods that are rela- 

 tively inexpensive, entail a minimum of difficulty at 

 the site and to the user, and yield meaningful re- 

 sults are available. 



Appendix 2 to Section VIII— Contingent 

 Valuation (Survey) Methods 



(a) Overview. (1) Contingent valuation methods 

 (CVMs) obtain estimates of changes in NED bene- 

 fits by directly asking individuals about their willing- 

 ness to pay (WTP) for changes in quantity of recre- 

 ation at a particular site. Individual values may be 

 aggregated by summing the WTPs for all users in 

 the area. 



(2) Contingent valuation methods consist of de- 

 signing and using simulated markets to identify the 

 value of recreation just as actual markets would, if 

 they existed. Three basic steps are involved: (i) The 

 analyst establishes a market to the respondent; (ii) 

 he permits the respondent to "use" the market to 

 make "trades" and to establish prices or values 

 that reflect the respondent's individual valuation of 

 the recreation opportunities "bought" or "sold"; 

 and (iii) the analyst treats the values reported by 

 the respondent as individual values for the recrea- 

 tion, contingent upon the existence of the de- 

 scribed market. The respondent's bids are used 

 with the data contained in the market description 

 (step i) to estimate the aggregate value of the rec- 

 reation being studied. 



(3) Contingent valuation methods are particularly 

 appropriate for evaluating projects likely to be one 

 of several destinations on a single tnp and projects 

 that will result in a relatively small change in the 

 quality of recreation at a site. Contingent value re- 

 sults may be adversely affected unless questions 

 are carefully designed and pretested to avoid sev- 

 eral possible kinds of response bias. Several tech- 

 niques are available for obtaining the individual 

 bids, which are the basic data for CVM. 



(b) Iterative bidding formats. (1) Iterative bidding 

 surveys ask the respondent to react to a series of 

 values posed by the enumerator. Following estab- 

 lishment of the market and a complete description 

 of the recreational good, service, or amenity to be 

 valued, the respondent is asked to answer "yes" or 

 "no" to whether he is willing to pay the stated 

 amount of money to obtain the stated increment in 

 recreation. The enumerator iteratively varies the 

 value posed until he identifies the highest amount 

 the respondent is willing to pay. This amount is the 

 respondent's "bid" for the specified increment in 

 recreation. 



(2) Iterative bidding techniques are most effective 

 in personal interviews. Mail survey formats have 

 also been used in research studies. These typically 

 ask the respondent to answer "yes" or "no" to a 

 small number of specified values in iterative ques- 

 tions and, finally, ask an open-ended question: 

 "Now, write down the maximum amount you will be 



willing to pay. $ ." At present, mail survey 



applications of the Iterative bidding technique have 

 not been adequately tested and cannot be recom- 

 mended. 



(3) The recreation facilities to be evaluated will 

 be described in quantity, quality, time, and location 

 dimensions. These descriptions should be hypo- 

 thetical in the sense that they do not precisely de- 

 scribe features of actual sites or proposed projects, 

 but they should be precise enough to give the re- 

 spondent adequate information on which to base a 

 valuation. To permit estimation of regional models, 

 quantity, quality, and location dimensions should be 

 varied and the iterative bidding exercise repeated. 

 Verbal descriptions should be precise, and, when 

 practicable, pertinent aspects of the facilities should 

 be displayed or depicted nonverbally (e.g., with 

 photographs, drawings, motion pictures, scale 

 models). 



(4) In most cases, the good to be valued is "the 

 right to use (the recreation facility) for one year." 

 The responses obtained are thus annual measures 

 of the individual's willingness to pay for a given in- 

 crement or decrement in recreation opportunities. 

 Bidding formats that define the good in some other 

 terms (e.g., day of use, trip) can also be used in 

 some applications as long as appropriate estimates 

 of numbers of days of use and trips are available to 

 permit calculation of annual values. 



(5) The institutional rules pertaining to the hypo- 

 thetical market will be described in sufficient detail 

 so that the respondent knows his rights and the 

 rights of all others in the market. These rules 

 should be realistic and credible, they should place 

 the respondent in a role and encourage market be- 

 havior with which he is familiar, and they should be 

 of a kind generally viewed as just, fair, and ethically 



79 



