Evaluation of a networked self-testing program

The use of a computerized, multiple-choice test bank to present practice an d assessment tests on a network was evaluated with 46 men and 119 women fro m a first-year class in psychology. A correlation of .65 (p < .001) between scores on a traditional paper-and-pencil test and scores on a computerized test provided some validity for the computerized assessment. Regression an alysis showed that ability (previous academic performance) and motivation ( number of practice tests taken) accounted for 73% of the explained variance in computerized test scores. Sex differences did not enter the regression equation significantly.