Contextuality and Dichotomizations of Random Variables

The Contextuality-by-Default approach to determining and measuring the (non)contextuality of a system random variables requires that every variable in be represented by an equivalent set dichotomous variables. In this paper we present general principles justify use dichotomizations determine their choice. main idea choosing is if possible values endowed with pre-topology (V-space), then allowable split space into two linked subsets (“linkedness” being weak form pre-topological connectedness). We primarily focus on types most often encountered practice: categorical real-valued ones (including continuous variables, greatly underrepresented contextuality literature). A (one finite number unordered values) all its dichotomizations. If are real numbers, they dichotomized intervals above below cut point.