The above is the link to wikipedia’s info about Preference (economics). One thing that is worthy to think, is that, what you prefer, cannot simply be illustrated as what you will choose, say, A over B, and B over C. But if you ask the questions in the other way: which one you will drop first, A, B, or C, people can have different answers from choosing A, B, or C.
When you drop one choice, the opportunity cost of elimination will directly hit your head. And, the notion is: the opportunity cost of choosing one over another is not symmetrical to the opportunity cost of dropping one first over another, as opportunity cost is not comprised of just a number on the number line, but can be a partial differential functions comprised of various parameters, that can be transformed to a matrix, that distributive law is not applicable.
If so, perhaps even the questionnaire on preferences and choices should be structured not just in preferring what over others, but preferring dropping what first over others, for a more comprehensive, and perhaps even precise, result.