The purpose of this inquiry was the construction of a useful, valid, and reliable test instrument to be used for the prediction of the future academic success of students in the Iranian guidance cycle which occurs when students are approximately fourteen years of age. It was assumed that due to the lack of objective measuring instruments and standardized tests, the new guidance cycle would not function efficiently.The first instrument for the purpose of evaluating students' talents and abilities was constructed on the basis of items selected from several widely used American test batteries, such as the DAT and W-G CTA tests. These were translated into the Persian language. The experimental instrument constituted a total of 304 items with which selected Iranian students were tested.The one hundred students in the population selected for this study were randomly selected from eighth grade students at three junior high schools, two boy's schools and one girls' school in Tehran. The items were answered by the students in three sessions and then were returned to the United States for item analysis.Item analyses were run to identify those test items which clearly discriminated between the academically talented and less able students previously identified by teachers in the schools. Since the primary purpose of this experiment was to select the items which would discriminate most effectively, an ordinal index of difficulty was run in an effort to secure the appropriate selection of items. Indices of discrimination were also carried out. The most appropriate experimental approach, with respect to the external criterion (teachers' judgments concerning the students' academic ability), was judged to be the high-low group (HLG) design, with a wide separation between superior and less qualified students. The summary statistics of the item analyses run to determine the difficulty as well as discrimination of test items reflected the relatively high accuracy of the teachers' judgments in distinguishing between academically talented and non-academically talented students in this experiment. It appeared that the tests, on the whole, discriminated between the academically talented and non-academically oriented students, and thus, will likely function more effectively after the less discriminating items have been deleted and the tests are further refined.The items used in this experiment are related to the following seven interrelated competencies and appeared to be a valid measure of the general academic abilities of Iranian students:1. Verbal Reasoning2. Numerical Ability3. Inference4. Recognition of Assumption 5. Deduction6. Interpretation7. Evaluation of ArgumentsIn the first selection, eighty-six items fell within the criteria represented in the item analysis technique designed to strengthen test validity and reliability. An additional number of twenty-four items was selected in the second review of the items, thus making the total number of appropriate items one hundred and ten. The items selected appeared to show high validity and reliability for future testing with Iranian students. These will, hopefully, in the future constitute an assessment measure which will yield high correlation with the future academic success of Iranian students.This kind of test allows the guidance counselors of Iran to feel reasonably certain of the identification and prediction of student talents and abilities associated with academic or technical and vocational studies in the guidance cycle as well as in the subsequent planning of curricular programs. However, it would be difficult to reach final conclusions about the students? Academic success on the basis of test scores alone. There are a number of social and psychological factors which affect the academic success of students. Predictions made about students or their selection should be temporary, flexible, and, in effect, for a short term of time.
| Identifer | oai:union.ndltd.org:BSU/oai:cardinalscholar.bsu.edu:handle/176384 |
| Date | January 1972 |
| Creators | Ghazi, Ghassem, |
| Contributors | Barnes, Donald L. |
| Source Sets | Ball State University |
| Detected Language | English |
| Format | vii, 1558 leaves : ill. ; 28 cm. |
| Source | Virtual Press |
| Coverage | a-ir--- |
Page generated in 0.0018 seconds