Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
DIFFERENCES IN RESPONSE BEHAVIOR FOR HIGH AND LOW SCORERS AS A FUNCTION OF CONTROL OF ITEM PRESENTATION ON A COMPUTER-ASSISTED TEST
Abstract
The purpose of this research was to determine whether the format of a computer administered test, specifically the ability to return to previous items, has any effect on test taking behavior. The test formats used in this research included a cyclical format, where the subject was allowed to return to an item given earlier in the test, versus a traditional format, where the subject was required to answer each question as it was presented. In order to determine whether behavior would differ as a result of ability, high and low ability groups, as determined by self-reported mathematics ability, were included in the analysis. Eighty subjects, primarily sophomore and junior college students, completed all materials used in the experiments. Each of the four treatment groups was compared on score, attitude toward computer administered tests, anxiety, and self-reported test taking behaviors. Subjects in the cyclical group did tend to complete more difficult questions later in the test, but did not differ in overall score. Differences were found in attitude and anxiety across the two forms. Subjects who took the traditional computer administered test had a more positive attitude toward the computer administered test, and a smaller increase in anxiety. This was hypothesized to be due to the increased cognitive load needed to deal with the extra directions and/or decisions necessary with the cyclical format. Few differences were found for high and low ability groups. One difference noted, however, was in the area of answer checking behavior. The cyclical format appeared to encourage re-thinking a problem when checking for the high ability group, while no differences were noted for the low ability group. It would appear, from these results, that there are no practical benefits to using a cyclical format. Further research, however, is necessary in answering this question. The most obvious limitation of this research is the lack of consequences based on test performance for these subjects. Effects due to the time interval between successive presentations of an item, practice effects, and different item types and orders should also be examined.
Subject Area
Educational evaluation
Recommended Citation
HARVEY, ANNE LYNN, "DIFFERENCES IN RESPONSE BEHAVIOR FOR HIGH AND LOW SCORERS AS A FUNCTION OF CONTROL OF ITEM PRESENTATION ON A COMPUTER-ASSISTED TEST" (1987). ETD collection for University of Nebraska-Lincoln. AAI8717253.
https://digitalcommons.unl.edu/dissertations/AAI8717253