Looking Forward: A Simulation of Decision Aids in Tomorrow's Classroom

Abstract

Although there are many examples of successful application, the introduction of decision aids into the home, school and workplace has frequently failed to produce expected gains. The present study sought to determine if undergraduates could effectively use a decision aid to improve their scores on a multiple-choice test. The design was a 2 (Aid Accuracy: 60%, 80%) × 3 (Trial Type: Aid Given On Request (OR), Aid Given On Every Trial (OET), No Aid Available). Results revealed a statistically significant Aid Accuracy × Trial Type interaction. Providing undergraduates with an aid that whose recommendations were correct on 60% of the trials did not enhance their test scores. However, equipping students with an was 80% accurate increased performance above the no aid condition. More molecular analyses yielded a series of results that further defined the effects of the decision aid on student performance. The most important of these was the discovery of a “mere presence” effect. Thirty 30 of 44 students requested no advice in the OR arrangement. With respect to information received, their OR trials were equivalent to the No Aid trials. Nevertheless, these students performed substantially better on the OR trials than on the No Aid trials. Thus, the decision aid had effects on performance that cannot be attributed to the information it supplied.