A Research of the Reliability and Validity of the PSYGAT on Different Linguistic Backgrounds
The issue of cultural bias in intelligence tests sparks debates every time the latter is created or administered, resulting to many researches into how the reliability and validity of an ability test may differ when assessed on groups from different cultural-linguistic backgrounds. The aim of this study is to test the reliability and validity of the PSYGAT Verbal IQ Test on university students from English-speaking backgrounds (ESB) and non-English speaking backgrounds (NESB) in relation to the Queendom Verbal IQ Test and Cultural Fair IQ Test. 445 third year psychology students aged 19 to 62 were involved in this study; 319 participants were categorized in the ESB group whereas 126 in the NESB group, whom all took three tests; the Queendom Verbal IQ and Cultural-Fair Test, then the PSYGAT Verbal IQ Test. Reliability as internal consistency and a Pearson’s Correlation test was run using SPSS on the data score collected. Although the results showed that PSYGAT is reliable and valid in assessing the verbal ability of culturally-diversified participants, it must be reminded that this is true within a university context only.
Intelligence, an issue easily opened to heated debate, which a concrete definition has yet to be decided upon, from the concept that it is about how able an individual is at learning, reasoning and applying, to how an individual acts rationally and effectively with his/her environment (Varon, 1936; Wechsler, 1958, as cited in Cohen & Swerdlik, 2010). However, it is common to assess intelligence – for the purpose of this study; we term it as “abilities” – with the use of intelligence tests, or a commoner term, IQ (Intelligence Quotient) Tests. A test is defined as a measuring device or procedure, designed to measure a related variable (Cohen & Swerdlik, 2010, p. 5). But as tests are “designed”, by humans, there may be flaws that may or may not, cause it to be biased towards certain group, or groups of test-takers, depending on its construction and its psychometric properties.
Culture, is one of these issues that debates on test biases frequently encircles. Sternberg (2004) defines culture as a set of values, beliefs and behaviors shared by a group of people with language as the main means of communication, and that intelligence (or abilities) cannot be fully investigated outside its cultural contexts. Then again, this is parallel to the notion that test bias is present, when the test score has different implications for different groups or test-takers, often differentiated by age, gender and of course, cultural-linguistic background (Cole, 1989), because distinct cultural groups promote different types of abilities (Cohen & Swerdlik, 2010). Consequently, in consideration of the complex role culture plays in measuring abilities, and whether or not they contribute as sources of biases, researchers often delve into the psychometric properties of a test to investigate whether that particular test is reliable, and valid for measuring particular abilities across diverse cultures.
Here, reliability of a test means how “consistent” a test is at measuring what is proposes to measure, with the reliability coefficient as an index of reliability (Cohen & Swerdlik, 2010, p. 139). There are two notions of reliability which are “reliability as stability over-time” which investigates into the consistency of the test on individual scores when the tests are taken on several occasions, and “reliability as internal consistency” in which the extent of homogeneity (or heterogeneity) of a test items measures a unidimensional (or multidimensional) construct (Cohen & Swerdlik, 2010). Most psychological tests works within the second framework of reliability, as it takes less time and is less costly to be administered. The more homogeneous the test items, the better it is at...