Analisis Item, Pengujian Reliabilitas, dan Konsistensi Internal Subtest Informasi dan Subtest Hitungan pada Test Inteligensi Wais Adaptasi UGM

The Wechsler Adult Intelligence Scale (WAIS) UGM Revision has been long and widely used in Indonesia as an instrument for measuring IQ. Nevertheless so far it has not much been done to improve the quality of it's items and the test itself. The purpose of this study was to examine the quali...

Full description

Bibliographic Details
Main Authors: Azwar, Syaifuddin, Rustam, Amrizal, Harjito, Purba
Format: Other
Language:English
Published: Fakultas Psikologi Universitas Gadjah Mada 1989
Subjects:
Online Access:https://repository.ugm.ac.id/275758/1/supriyanto_201307041_saifudin%20azwar%20%20JUDUL.pdf
Description
Summary:The Wechsler Adult Intelligence Scale (WAIS) UGM Revision has been long and widely used in Indonesia as an instrument for measuring IQ. Nevertheless so far it has not much been done to improve the quality of it's items and the test itself. The purpose of this study was to examine the quality of the WAIS Verbal Subscale, spesifically the Information and the Arithmetic Subtests by conducting item analysis procedures, computing reliability coefficients, and assessing the internal consistency indexes of the subtests. Based on a sample size of N = 100 (mostly were SLTA newly-graduates), for the Information subtest, it could be generally concluded that (a) difficulty-wise, the items of the subtest were not orderly arranged. Some i terns were so easy for the subjects. that they could be considered inappropriate; (b) based on discriminating power most items were proved to have very low validity coefficients, and (c) the reliability of the subtest was not convincing. For the Arithmetic subtest, the general conclusions were (a) the items were fairly well arranged according to the item difficulty indexes. The variety of item difficulty was good; (b) most items have reasonably high validity coefficients, and (c) although the reliability coefficient of the subtest was much better than one of the Information subtest it was still considered low. Improvements of the quality of the items and the subtests are recommended.