| Sumario: | A study was conducted to compare a computerized perceptual/clerical speed test with a paper-and-pencil form in order to update established predictors of clerical work behaviors. Twenty different versions of the test were developed, in which type of text, feedback, and user friendliness were varied. Findings revealed the presence of a very strong common factor of speediness among all versions of the test as well as a second factor capturing unique method variance. Although they were closely related, it was noted that the two were not equivalent. In addition, the type of text and presence of feedback was found to influence performance.It was concluded that a number of factors have to be considered when trying to obtain convergent validity for computer-based perceptual/clerical speed tests.
|
|---|