Hello
I have estimated a single bounded spike model in Contingent valuation; but the sign of the coefficient of bid variable is positive. can anyone help me to interpret this coefficient?
I reviewed some articles and in some of them the sign is negative and in some others are positive.
for example in the paper I shared the sign is positive.
The spike model is a way of dealing with zero responses in CVM surveys, which can arise when respondents are not willing to pay anything for the public good or service, or when they refuse to answer the question. The spike model assumes that there are two types of respondents: those who have a positive willingness to pay (WTP) and those who have a zero WTP. The model also assumes that the WTP distribution is lognormal for the positive WTP respondents.
The coefficient of bid variable is the parameter that measures the effect of the bid amount on the probability of a positive response. In other words, it shows how sensitive the respondents are to changes in the price of the public good or service. A negative coefficient means that as the bid increases, the probability of a positive response decreases, which is consistent with the law of demand. A positive coefficient means that as the bid increases, the probability of a positive response increases, which is counter-intuitive and may indicate some problems with the survey design or data quality.
A positive coefficient of bid variable can occur when there are protest responses, strategic behavior, warm glow effects, or information bias among the respondents. These factors can affect the validity and reliability of the WTP estimates. Therefore, it is important to check for these issues and correct them if possible before applying the spike model.