Dependent and independent variables used in selection and
outcome models in potatoes, peas, green beans, strawberries or
cherries are given in Table 1. In the first stage, selection equations
were formed and were solved with a probit model. The main reason
for using a probit model was to calculate the sample selection
equation or, in other words, the Heckman correction coefficient.
This variable is represented by l and is calculated by making use of
the error term of the solved probit model.