By adopting the logistic model to estimate the probability of choosing the non-GM food,
the econometric model can be specified as the logit model:
y = αk + βp + ε
(14)
where y = <
rɪ
if the respondent chooses non-GM food product
otherwise,
Also, k is a vector of explanatory variables and p is price factor. In our empirical model, the
price factor is defined as the “price difference” between non-GM and GM food in order to
capture the price effect and WTP can be estimated as the expected premium for non-GM food.
The Survey and Data
A mail survey was conducted in the Columbus Metropolitan Area, Ohio in March 2001.
A three-wave procedure combined with mailing and telephone was used in order to maximize the
response rate. The sampling frame was obtained from the Center for Survey Research at The
Ohio State University. The center randomly selected 650 telephone subscribers in the Columbus
Metropolitan Area based on the zip code. The questionnaires and postage-paid return envelopes
were mailed to these randomly selected households. In total, 141 completed survey
questionnaires were returned, along with 120 undeliverable returning questionnaires, yielding an
overall response rate of 26.6%. Four versions of questionnaire with different prices in the CV
section were equally distributed among the 650 mailings. Among the 141 returned respondents,
we collected 39 copies of version 1, 28 of version 2, 26 of version 3, and 48 of version 4. (These
four versions of prices will be discussed later.)