Jan 1, 1974
Describes a prototype computer program based on "admissible probability measurement" — a testing procedure which allows a user to express his uncertainty, in terms of probabilities, about the correctness of his answers to multiple-choice questions. A logarithmic scoring system encourages the user to "reproduce" or "reveal" his personal, subjective probability of correctness for each of the choices. The computer system allows the user to explore the relationships between his assignment of probability of correctness to answers and the possible score outcome. In addition, during test administration, the user is provided with feedback about his present state of knowledge and how realistically that level of knowledge is assessed. The ease of use of the computer system and the speed of analysis permit cycling through tests quickly, thus removing a major barrier from the task of learning to make better assessments of the user's knowledge.