The purpose of this study was to analyze the reliability and validity of three achievement tests, with emphasis given to instructional validity. The tests were used to assess the performance of Business Computer Applications (BCA) students on their achievement of knowledge of computer application software concepts and vocabulary for word processing, spreadsheet, and database.

Five concepts and vocabulary categories addressed included: vocabulary, access software, data/text entry, editing, and formatting. Data were also collected from BCA teachers of the students. Teachers were asked to rate each item for the three software types of the achievement test as to whether the concepts and vocabulary corresponding to that item had been taught.

A selected sample was used in the study. Classes were divided into three approximately even student groups. Each group completed one of the tests for each software type. Percentage of items correct for the three software types were 53%-word processing, 46%- spreadsheet, and 36%-database. The Kuder-Richardson 20 (KR-20) estimates for the software types were .92--word processing, .92--spreadsheet, and .76--database. Due to a wide dispersion of KR-20 values for items within the 5 concepts and vocabulary categories by software, the items are useful only for group measures.

Most of the 8 teachers did not teach a substantial part of the concepts and vocabulary included in the validated test items. This outcome indicates the curriculum related to the items needs to be examined and revised so that teachers can emphasize the most important concepts and vocabulary with their students.

Analyses outcomes indicated that 59 of 60 word processing items, 51 of the 59 spreadsheet items, and 45 of the 59 database items were instructionally valid. Teachers selecting items for purposes of examining the achievement of their students related to software concepts and vocabulary cam effectively use items determined to be instructionally valid in this study.