More moderate correlations are obtained between the PS final exam and GE exams and final grades, but this may be explained by the less standardized nature of the tests given in the GE exams. It appears that the PS final exam has at least some predictive ability for future performance in the following English class.
Thus, we can conclude that, at least for this set of tests, the assessment of face validity fairly accurately reflects more objective measures of test quality. The most immediate pedagogical implications drawn from the study largely concern the curriculum unit and testing office of the ZKU Preparatory School.
In turn, the testing office should examine the extent to which these goals and objectives are represented in the achievement tests and examine the conditions under which tests are administered and scored.
Another immediate implication for the Z KU Preparatory School to emerge from this study is that, since the predictive validity of the midterm achievement tests is relatively high, these scores can be employed to predict students’ achievement on the final exam and to identify those students in danger of failing. Such students might be offered extra assistance and support, to improve their chances of success.
There is also some evidence that many students who score poorly on the PS final exam will also perform poorly in GE classes the next year. Indeed, 76 per cent of those scoring below the mean on the preparatory final exam also received a final grade that was below the mean in the first semester of the GE class.
Such a trend underscores the need to provide extra assistance to students who perform poorly in PS, to ensure that their poor performance does not continue in subsequent English classes.
Several implications for other institutions can be drawn from the current study. The questionnaires employed in this study might be used in other institutions to assess their own achievement tests. Furthermore, the study Discount Shoes illustrates a process by which other institutions may examine the appropriateness of such achievement tests.
Finally, with regard to the relationship between face validity and more objective measures, it was shown that the face validity of these achievement tests accurately reflects the aspects of reliability measured in this study, as well as the predictive validity. This might imply that administrators and testing officials may rely on perceptions of face validity in determining the worth of a test.