مقارنة الخصائص السيكومترية بين الاختبارات التحصيلية الإلكترونية والورقية

الملخص EN

The present study aimed at comparing the psychometric characteristics of both the electronic achievement tests versus the pencil and paper ones through making a comparison of the statistical analysis of the difficulty index and its discrimination index as well as the reliability and the concurrent validity of the achievement test when presenting it in three different ways: computer based testing through the lecture, computer based testing at home and Paper-pencil testing.

The co-efficient reliability was calculated by two different ways of the internal consistency: Cronbach’s alpha and the split half-reliability coefficient, and calculating the concurrent validity of the test through calculating the coefficient correlation of students' scores on the formative achievement test and the summative one.

The formative test consisted of twenty MCQ items with four alternatives in the Thinking Skills Course using the experimental research design.

Three groups were selected from among the Faculty of Education for girls in Imam Abdulrahman Bin Faisal University.

The first group was tested electronically using the Blackboard system during lectures making sure of giving students only one trial to answer the test.

The allotted time was twenty five minutes.

Shuffling questions as well as the alternatives where considered.

All questions were presented in one page as to give students the chance to revise their answers before submission.

The second group was tested asynchronously using the Blackboard system.

Students were allowed to answer the test in one session and there were not allowed to go back to review questions.

Shuffling questions as well as the alternatives where considered as to reduce the opportunities for cheating the answers to minimum, while the third group was tested conventionally (paper-pencil test).

Results revealed that there was a statistical difference between difficulty indexes of the achievement test items in presenting it asynchronously and when presenting it while lecturing and conventionally (paper-pencil test),and there was no statistical difference between discrimination indexes of the achievement test items in presenting it while lecturing and when presenting it asynchronously and conventionally (paper-pencil test) The value of the reliability coefficient and the concurrent validity was extremely high when conducting the test during lectures and it was extremely low when conducting the test asynchronously.

In the light of these results, the present study recommends developing the procedures of managing tests asynchronously as to give teachers more opportunities to monitor students while conducting tests.