6.2. Monetary Valuations from Subjects

  Data solicited from respondents about estimations of values placed on non-monetary costs associated with travel have long been analyzed.   Economists have developed tools such as the contingent valuation method (CV) to value environmental amenities or changes (Clarke, 2000) .  These types of monetary valuations can reveal much about how people value different kinds of tasks.   For example, most people value their time driving a car as being less than the value placed on their work time.   People put an even lower cost on a drive to a recreational spot.   Thus, this discount reveals a utility inherent in the purpose of the journey (McFadden, 1988) .  The empirical data collected in this experiment using various forms of monetary valuations are discussed here.   These values were used to analyze the benefits that severely vision-impaired people place on their ability to reduce travel time, stress, apprehension, or fear, and to gain the ability to travel independently.   This chapter offers a brief review of problems that can reduce validity in the types of questions about monetary valuations discussed in Chapter 5.    

6.2.1. Techniques of Monetary Valuation

Contingent Valuation (CV)are survey methods designed to elicit Willingness To Pay (WTP) amounts to ascertain what monetary benefit people place on goods or services.   These tests ask people to consider certain situations or preferences and respond with a monetary amount that they would pay to be able to receive those goods or services.  It is assumed that choices can accurately reflect well-formed and stable preferences, insofar as these techniques are based on classic economic rationality (McFadden, 1988, p. 339) .  If this is true, it should be possible to deduce from direct questions the social desirability of public policy initiatives such as transportation improvements.   However, people do not always act according to assumptions on which these neo-classical economic models are built.   One must also consider behavior and attitudes due to the wide range of beliefs and perceptions of humankind.   People are more complex and fallible than assumed in the “economically rational man” model.  In addition, the psychological model is dynamic—a process model where the emphasis is on how beliefs or preferences are formed and how information is acquired.  
 
Accurate measurement of a value placed on specific goods or services requires that the subject understands the scenario or task, that it can be preformed, and that all sorts of possible bias and misunderstanding be attended to (Bateman et al., 1999) .  Four barriers that might restrict the subject from giving the information the experimenter is truly seeking are summarized from Sudman & Bradbury (1982) : 

  • Memory: the respondent may have forgotten or remembered incorrectly;
     
  • Motivation: the respondent may be afraid to tell the truth, want to present themselves in a positive manner, support the questioner’s implicit position (experimenter effect), or not care enough to respond accurately;
     
  • Communication: they may not understand what they are being asked;
     
  • Knowledge: they may not know the answer.
     
    To accomplish these tasks from a psychological perspective and increase validity, a study from NOAA (1993) suggests attention to the following:

  • Convey meanings exactly to the respondent;
     
  • Avoid incorporating implicit theory
     
  • Begin with the needs and perceptions of the percipients;
     
  • Enable the respondent to learn what his or her preferences are in the course of the experiment.

    • McFadden (1988) states that preference tests should strive to elicit core preferences and avoid strategic behavior or cognitive illusions, as when a person is confronted with unfamiliar tasks and looks instead for “cues from context to shape an appropriate response” (p. 355).   He states that the more realistic the hypothetical setting, the more likely it is that stated choice behavior will look like real choice behavior.   In some research, such as assessing the desirability of transportation modes not yet in service (such as high-speed rail not familiar to most respondents in the United States), descriptions tend to be highly stylized and the results biased.   In reality, “inexperienced consumers confronted with incomplete information on a commodity may make a biased imputation of unobserved attributes, and may make mistakes in weighting these attributes in comparison with observed attributes” (McFadden, 1988, p. 360) .  Green & Tunstall (1999) suggest that good experimental design must direct respondents’ attention to the issue, enable them to form preferences, and practice their willingness-to-pay .   

    6.2.2.   Design Method and Results

    The focus of most of the field experiment was on determining the extra time and effort needed to make successful independent travel with limited or no sight.  Along with measurements of time costs and success, safety concerns, and the degree of difficulty of various tasks were researched.   The CV and WTP literature emphasizes that a valid research question should be important to the subjects and be one about which they have great interest.   The scenario should be simple and explicit, with little chance of misinterpretation.   There can be little doubt that safe and independent travel is an important concern for this population.   Their quality of life is defined daily by the efforts of mobility and navigation without sight.  The questions were as straightforward as possible: what would they pay to independently travel to types of activities, how much more could they earn, and how much less would they spend on assistance.

    6.2.2.1. Pre-test questions

    The pre-test questions were asked with no reference as to how this equal access would be gained.   It was up to the individual to imagine that kind of scenario.   These questions helped to open up cognitive processing about a heretofore-unattainable goal and gave a baseline against which to test their WTP response after using the RIAS.    These were conducted by telephone, before the subject and interviewer had met.   The interviewer did not previously know these subjects, and so there were no known reasons or motivation to give slanted answers.   The pre-test questions had nothing to do with any technology being studied; they simply asked for values to be put on equal access.   When there is no consumption or purchase involved, WTP data are considered more valid because there is no reason to give false answers (McFadden, 1988) .  Their perceptions of worth come directly from their experiences as travelers using their normal skills.   At the time of the interview, the subjects were comfortably at home, not dealing with the difficulties of travel.   Most people offered amounts for these scenarios, thus adding to their real-world validity.   

    6.2.2.2. Post-test questions

    The same questions were asked after a long and demanding field experiment, and, in that case, subjects evaluated the WTP for a very concrete and specific technology.  They had just spent up to several hours using the system while making many trips in a crowded and complicated urban transit terminal.   They had been asked to independently perform many tasks and had a very vivid memory of that situation and of what they had just experienced.   In this way, the respondents were self-aware of the effects of the system and how vision loss affects their travel choice and activities, without the researcher having to describe any scenario or perceived benefits.   The perceptions of the worth came directly from their experiences during the experiment.  There was no need for a mental search for facts or experience on which to base their answers.   All subjects had the same experience on which to judge their answers.   They were also highly aware of the trips they do not make and other restrictions on their personal activity space and behavior.   Without the pre-test questions, doubt could be raised about the accuracy of the post-test RIAS condition question.  By including the hypothetical pre-test questions as a control, subjects could be tested as to whether they were affected by attitudes toward the researcher and if the effects of primacy and recency apply.   If they did, there should have been a difference between the field test and the telephone question results.   However, there was no significant difference in monetary valuation between a self-generated scenario (with equal access and independence) in the pre-test questions and the very real world they experienced in the transfer tasks using a specific technology.   By asking such simple and relevant questions and being assured that subjects had the vital information they needed to make a value judgment, the concerns and caveats mentioned above have been addressed.    

    6.2.2.3. Willingness-to-Pay Versus Perceived Benefits

    It was important to establish a monetary amount that subjects placed on the benefit of independent travel and equal access to transit.   The pre-test question dealt with a concept only: what is independent access worth, and what is it worth to have the same access to signs as the sighted public.   However, the true “benefit” of this access is most likely incalculable; perhaps no amount of money can truly capture the benefit of access to all signs and the ability to travel independently.   Although it is the widely held opinion of people with vision impairments that they should not have to pay for access that the general public gets at no extra charge, the question based on what they would pay was asked.   This put the valuation on the level of what they would actually be willing to forgo in order to achieve these goals.   This procedure also made it easier to compare the answers from the concept question to those answers derived from the experiences of an existing technology that they had just experienced.   By asking what they would pay for equal access, a baseline was established for this ideal self-generated manifestation, which, hopefully, revealed the cost of limited access and dependency.

    6.2.2.4. Saliency and Importance of the Scenarios

    WTP data, to be valid, must relate to scenarios that are well understood by the subjects.  They must have full knowledge of the material and how it would affect their lives.   It could be argued that there is little, or nothing, more important to a person with a vision impairment than the ability to independently move through the urban environment and to have equal access to activities and travel.   They are faced daily with restrictions on travel and are acutely aware of what it would be worth to “turn off” these limitations.   There is little reason to doubt that the subjects did not fully understand the question, have complete and current knowledge of the situation, and know what the cost of those limitations meant to them.

    6.2.2.5. Cognitive Illusions

    Subjects fully understood that the University of California and the California Department of Transportation were conducting this research.   No product sale or market analysis was involved.   Cognitive illusions are highly sensitive to context, and they are stronger with unfamiliar tasks.   For a monetary study to have a reasonable probability of success, the “consumer” must be fully informed about the attributes of the commodity and be experienced in making decisions about it or trained in a manner that provides a context that resembles historical experience (McFadden, 1988).

     
    BACK TO OVERVIEW
    		BACK TO TABLE OF CONTENTS
    		NEXT SECTION