지불용의액 추정을 위한 조건부가치평가 대안으로서 일방응답모형의 실험평가 (An Experimental Evaluation of One Sided Response Model as an Alternative toContingent Valuation Method for Estimating the Willingness to Pay)

서지정보

· 발행기관 : 한국마케팅학회
· 수록지 정보 : 마케팅연구 / 22권 / 3호 / 121 ~ 133페이지
· 저자명 : 박주헌

초록

출시 전 신제품에 대한 소비자들의 지불용의액(willingness to pay)을 조건부가치평가방법으로 추정하면 과대 추정
되는 편의가 발생하는 경향이 있음은 널리 알려져 있다. 가상적 편의(hypothetical bias)라고 불리는 이러한 편의는
조건부가치평가방법 뿐만 아니라 가상거래를 상정하고 조사된 서베이 데이터를 이용하는 방법론에서 공통적으로
발견되는 문제다. 왜냐하면 서베이 응답자들은 자신들의 지불용의액을 과장시키는 경향이 있기 때문이다. 일방응답
모형은 가상적 편의를 줄여 비교적 정확한 지불용의액을 추정하는 대안모형이 될 수 있다. 본 연구는 두 개의 실험데
이터를 이용하여 일방응답모형이 전통적인 조건부가치평가방법의 대안으로서의 가능성을 평가한다.

영어초록

Marketers often want to assess potential customers’ willingness to pay (WTP) for a product, especially when
developing a new product. The contingent valuation method (CVM) has been extensively used for estimating the
WTP for a non-market good. Since a new product has been never traded yet, it can be interpreted as a non-market
good. Therefore, the CVM can be used for estimating potential consumers' WTP for a new product.
However, the CVM has been long criticized to be a poor estimation method for the actual WTP because
respondents are suspected to exaggerate their true WTP. It is well known that estimates of consumers' WTP for
new products or services by the CVM are notoriously biased upward. This kind of bias, called as the hypothetical
bias, would be common for any methods using the survey data collected from the hypothetical transaction.
There have been many attempts to reduce the hypothetical bias. Recognizing that the hypothetical bias is
attributed to poorly designed survey instruments or administration procedures, many researchers have made
efforts to improve survey methods such as sampling design, instrument development, questionnaire structure, etc.
But the results are not satisfactory. As the hypothetical bias might be caused by respondents' exaggeration
tendency that may be due to new product enthusiasm, a desire to please the interviewer, or the tendency for people
to be less sensitive to total costs in a survey than in an actual purchase, we cannot expect to be able to remove it by
improving survey methods.
This paper addresses an issue of how to correct the hypothetical bias by introducing a new CVM called the one
sided response model (OSRM) which is developed by revising the CVM proposed by Hsiao and Sun (1998). The
OSRM assumes that an individual would make two independent preliminary choices before making a final
choice, one is based on the true preference and the other is randomly drawn independently of the true preference.
The individual then selects the larger one as the final choice. As a result, the OSRM integrates the respondents'
exaggeration behavior within the model. A problem is how well the OSRM corrects the hypothetical bias.
In this paper, two experiments are presented to prove the effectiveness of the method. As expected, we obtained
considerably lower WTP estimates from the OSRM than those from conventional CVM. An even better news is
that the OSRM gives a mean estimate of WTP quite close to the mean estimate obtained from the real responses
that are believed to reflect the true WTP, even though using the hypothetical responses that are suspected to be
contaminated by respondents' exaggeration tendency.
The OSRM will reduce the hypothetical bias of the CVM because it will clearly estimate the WTP lower than
the conventional CVM by the model structure. But we cannot exclude a possibility that the OSRM may
underestimate the WTP. If respondents correctly report or under-report their WTP, then the OSRM will
necessarily produce estimates that are biased on the low side. However, we believe that this will hardly ever
happen in a real world situation because respondents have little incentive to do so. Rather, we are concerned about
the possibility of over-correction.
We contend that in most cases, managers would desire to be conservative when estimating WTP, so the
proposed approach would be preferred to estimating WTP high, thereby risking unexpectedly low demand for
new products.

참고자료

· 없음