| he central purpose of this study was to investigate various methods of test administration and their effect on the reliability of scores on a Science Assessment Test. Six methods of test administration were developed. These methods varied in media type (computer-based and paper-and-pencil), print size, page/screen format, auditory stimulus and response recording. The six methods were administered to 342 fifth grade children. The resulting sets of scores were analyzed to compare the concurrent validity, the reliability, and their relationship to reading ability and science knowledge.;Three research questions were investigated. Does the concurrent validity coefficient of the Science Assessment Test significantly differ under the six methods of test administration? Does the reliability of the Science Assessment Test significantly differ under the six methods of test administration? What is the relationship of reading ability and science knowledge to the scores on the Science Assessment Test under the six methods of administration?;Using the Comprehensive Test of Basic Skills/Form U (CTBS/U) Science Test as a criterion measure, the difference in the concurrent validity of the scores on the computer-based methods (r =.68) and the concurrent validity of the paper-and-pencil methods (r =.51) was significant (p... |
