The role of item presentation, dispatching the answer, and presentation of the processing-time in the context of equivalence of paper-based-testing (PBT) and computer-based-testing (CBT) was analyzed. 298 pupils (M=18.9, SD=2.5 years) were tested two times, one group first with the PBT and several weeks later with the CBT, another group the other way round. The used test battery (revised LPS; originally Horn, 1983) contains 11 subtests measuring crystallized and fluid intelligence, visual perception and cognitive speedness (three-stratum-theory; Carroll, 1993).
The results subsumed for all subtests show, that it doesn’t matter, if the items of the CBT were presented one after another or all together, if the kind of test is unknown by the subjects. If the CBT is the second one, the presentation of all items together will lead to a higher degree of equivalence. If there is a button to accept the marked solution or not as well as the opportunity of changing already given answers plays a minor part for the equivalence. It although plays a tangential role if the remaining processing time is shown or not. For the single subtests the influence of the conditions were partly different, especially when the rate of speed component is high. The higher the mean item-difficulty the smaller the influence of the variation of the condition. But the equivalence of PBT and CBT couldn’t ensured with maximum similarity of both modes as well as it is possible that probably different seeming translations could lead to equivalent tests.