There’s a major question that has stumped psychologists for quite a long time: « How do you measure or assess intelligence in an unbiased and accurate manner? » Is it even possible? One of those psychologists is Dr. Gary A. Plank who became particularly interested in American Indian children/adolescents and the methods of intelligence testing they had to go through. He, like most other psychologists, knew that there is no such thing as a completely unbiased or culture-free test because no matter how you look at it, the very essence of our intelligence is embedded and defined by specific cultures. So what’s the problem with culturally influenced cognition tests? Quite simply, they’re unfair. Children who haven’t been exposed to the same mainstream cultural experiences as others end up with very low scores, not due to a lack of intelligence but because they lack the cultural experience.
That’s exactly what happened with American Indian children. They were tested using the Wechsler Intelligence Scale for Children (WISC). Their scores on the Performance Scale were similar to those of their Caucasian, Black and Hispanic counterparts. The problem was with their Full Scale scores, which were low, and their Verbal Scale scores which were extremely low, by about 20 points. This resulted in these American Indian children being labeled as « learning disabled ». But come on, is it really that likely that all American Indian children were learning disabled, or is it more plausible that there was something wrong with the method of testing? Dr. Gary set out to study the root of the problem along with Dr. Gerner, a school psychologist from Arizona, who suggested they use other tests instead of Wechsler. They tried out 3 others: the Stanford-Binet-Fourth Edition, Kaufman Assessment Battery for Children, and Kaufman Adult Intelligence Test. Sure, these tests might have been a bit less verbally/culturally influenced but they still had several disadvantages and were far from perfect. In the end, they reached an important conclusion: « one intelligence test is not enough » and from there, they began exploring the Gf-Gc Cross Battery approach.
The Gf-Gc cross battery approach was based on the Cattell-Horn-Carroll Gf-Gc theory which stated that there are several types of intelligence, not just Gf (fluid intelligence) and Gc (crystallized intelligence). There’s no single overall broad intelligence (G). Instead, intelligence is composed of a wide, complex array of human abilities. That’s why a cross-battery approach is absolutely necessary because it outlines which subtest from each battery best measures each type of intelligence. The Cattell-Horn-Carroll model reached the conclusion that there are 10 different areas of intelligence: (Gc), (Gf), (Gsm), (Gv), (Ga), (Glr), (Gs), (Gt or CDS), (Gq), and (Grw). Out of these 10 broad cognition areas, only 7 can be accurately assessed. That means you can’t just grade intelligence based on one or two scales (verbal and performance) like the Wechsler Scale does. Sure, the Wechsler Scale does a good job of measuring Gc and Gv but as for the other 5 areas, not so good. That’s probably why American Indian children scored poorly on the test because Gc is largely influenced by culture and language, so by giving American Indian students the Wechsler test, you’re basically testing their English proficiency and their cultural experience, not their intelligence. This of course, led to American Indian children being diagnosed with learning disabilities and that led to poor educational planning.
Psychologists soon realized that this was unfair for LEP (Limited English Proficient) students and decided to use Performance Scales, instead. Unfortunately, that still didn’t solve the problem because it measured both Gr and Gc which meant the language and culture factor was still an issue. American Indian students were still scoring lower than they should be. Since another option had failed, the cross-battery approach began to look even more necessary. Once again, psychologists turned to the Cattell-Horn-Carroll Theory. It had identified exactly 10 broad areas and 69 narrow areas and had set a rule: in order to measure a broad area, you needed to adequately measure 2 different narrow areas. That rule right there proved the Wechsler Scale was insufficient because it only had one narrow-area test, the Digit Span », for the broad area, Gsm. As for the cross-battery approach, all you need is 2 subtests from each battery to measure a certain type of intelligence. For example, with American Indian children, you can ideally measure 2 subtests for Gv or Gc and that would be enough to fairly assess the broad area of intelligence. No need to continue with 4-5 more subtests which means less administration time is required.
Psychologists acknowledge that there is no perfect test that taps into all broad areas of cognition. They have to make do with what they have. The best option not only for testing American Indian or culturally/linguistically different students, but for practically all individuals, is the cross-battery approach. Dr. Gary proved this for example, when he reevaluated 8 American Indian students who had originally scored an average of 58-69 on the WISC-III. Having scored lower than 70 points, they were labeled mentally retarded. When they were retested using the cross-battery approach, they scored an average of 90-105, which means these students were definitely not retarded or mentally impaired. See how the method of testing can completely change a student’s life?