THE EVALUATION OF A MICROCOMPUTER SCORING PROGRAM TO MARK EXAMINEE WRITE-IN RESPONSES

A study was undertaken to determine the feasibility of using a compute r program to mark examinee write-in responses. It compared the perform ance of a computer against that of physician markers. A new examinatio n designed to assess clinical decision-making was administered to eigh teen residents. The write-in responses on twenty-three questions were marked both by physicians and a microcomputer. The study found the com puter to be more consistent in marking, equally as cost effective, and as valid as physician markers. Since the computer provided question s tatistics (i.e., means, frequencies of examinees' responses) it was ev aluated to be better and more helpful than physician markers by provid ing data useful for improving examination questions and learning exper iences.