Factors contributing to success in mathematical estimation in preservice teachers: Types of problems and previous mathematical experience

Variables that may affect success in mathematical estimation in preservice elementary teachers were studied. A multiple regression analysis was performed in order to predict success on a 25 problem estimation test. Subjects' answer to the question, “Are you good at math?” was the best predictor of success on the estimation test. Other important variables were college grade point average, years of study of mathematics, and enjoyment of mathematics. Not only were average mathematics grades not related to estimation score, but they were negatively correlated to estimation score. Half the subjects saw problems in an applied format, and half saw the same problems in a computational format. The number of subjects answering problems correctly in the application format was significantly greater than the number of subjects answering questions correctly in the computational format. This result was contradictory to results usually found for finding exact answers with pencil and paper. A reason for greater success on problems in the application format may be related to the fact that the applied problems involved money, a very relevant type of problem in which to quickly estimate an answer.