Computer Learning and Appropriate Testing: A First Step in Validity Assessment

AbstractThis study looked for evidence that the type of test used to assess computer-mediated learning can influence research findings. The findings from much of the recently published research on computer-mediated learning have been based on pencil-and-paper evaluation of criterion variables. In the present study, 90 university undergraduates served as subjects in order to assess the effects of testing mode on paired-associate learning. A 3 x 2 design was used to examine the effect of testing mode congruence with presentation mode on learning outcomes. One third of the subjects were presented with simple line drawings and associated nonsense names on computer screens, and one third were presented the same information in study books. One half of each treatment group was tested using the computer. The remaining half of each group was tested using pencil and paper. One third of the subjects served as a control group and were tested without a learning session. The interaction between study mode and testing m...