Guess for Success? Application of a Mixture Model to Test-Wiseness on Multiple-Choice Exams

The use of large lecture halls in business and economic education often dictates the multiple-choice exams to measure student learning. This study asserts that performance on these types can be viewed as result process elimination incorrect answers, rather than selection correct answer. More specifically, how students respond a test broken down into fractions questions where no wrong answers eliminated (i.e., random guessing), one answer eliminated, two all eliminated. results from an empirical model, representing mixture binomials which probability choice depends number choices we find, using data final exam principles microeconomics consisting 100 multiple questions, responses characterized by some form guessing, with more 26 percent being completed purely guessing.