Besides serving as a research consultant in research design and statistical analysis to faculty and students, I am also interested in developing a new model of “multiple-choice testing”, which I call “Probabilistic Testing”. Instead of the conventional approach of choosing the “correct answer” from a list of 4 or 5 alternatives, the test-taker will assign “subjective probability” to each alternative answer. The goal of such model is to improve “test validity” and “test reliability”.