Abbreviation, Standardization and Computerization of the Raven Intelligence test - Standardized Progressive Matrices - on university students in Yemen

Abstract

This study aimed to abbreviate the marked Raven test of general intelligence (Raven standard matrices), then standardize the short test and set standards for Yemeni university students. The sample of the abbreviation test was (400) male and female students from Taiz University. The sample of standards and rationing for the short test was (3242) male and female students, from four Yemeni universities; Sana'a, Taiz, Ibb and Thamar, from both natural sciences and humanities, and from all academic levels. The comparison sample between the paper and pen test and the computerized test was (148) students from the fourth level in the Faculty of Education from the scientific and human disciplines, for the standard Raven matrices test by the researchers, and the standard Raven matrices test for the computerized ones by the researchers. The research concluded that groups (A) and (B) are very easy for university students in Yemen, and therefore they can be omitted. Thus, the test became shortened from 60 items to 36 items by deleting groups (A) and (B). The short test was standardized (36) items by making sure of its validity and reliability using different types of validity, including the validity of the items using ease and difficulty, as well as distinguishing the items and their relevance to their fields and the total score of the test, as well as the link of the fields to each other. All the previous results have shown the validity of the short test, and the stability of the test was confirmed by using split half and retest. Thus, the abbreviation is reliable and consistent, and this confirms that the abbreviation process was correct, and the abbreviated test became a valid and usable test. Standards for university students in Yemen were also reached according to the categories of achievement rate, specialization, gender, age, academic level, rural and urban areas, by extracting the percentiles, the skewed intelligence ratio, and a seven-year gradient to interpret the level of the raw score.