The dummy variable represents the shift in demand for
frozen potatoes. The coefficient of the dummy variable
was negative and statistically insignificant. The fact that
this coefficient was not statistically significant indicates
that there was no demand shift towards frozen potatoes
consumption. Therefore, the price premium that the
consumers would be willing to pay was not calculated. l
was found to be statistically significant. This indicates
sample selection bias and that Heckman correction was
required. A model that considers only the consumers who
purchase frozen potatoes will yield biased results. It
indicates that the consumers who did not buy frozen
potatoes had a potential effect on the tendency of
purchasing potatoes, and both the group of consumers
who purchased frozen potatoes and also those who did
not, should be considered and evaluated collectively in
the models.