Application and Usefulness of Theories of Intelligence in Assessment

Research into the concept and assessment of intelligence, clearly indicates that no universal consensus exists. On the contrary, many theories and much research have emerged with understanding continuing to be investigated and debated. Despite the breadth and depth of research there still exists no standard definition of intelligence. As cited by Neisser et al (1996) Stemberg and Detterman (1986) found that “when two dozen prominent theorists were recently asked to define intelligence, they gave two dozen somewhat different definitions.” Some research suggests it is a single general ability, whilst others identify a range of skills and aptitudes. In addition to the disagreement about what constitutes intelligence, is the debate about it, and how, accurate measurements are possible. Since the earliest intelligence tests by Alfred Binet and Theodore Simon in France 1905, which sought to identify those pupils who did not learn effectively in the classroom environment, theorists and researchers have attempted to devise psychometric testing which can effectively measure and quantify cognitive ability. Over time the intelligent quotient (IQ) and numerous similar revised editions of intelligence testing have been used to support key decisions on how children and adults may be differentiated, categorized and supported. There are many different kinds of intelligence tests using a wide variety of tasks. Some tests consist of a single type of task; others contain a broad collection of tasks with different contents. This continues to spur debate and controversy over the use of intelligence tests, cultural biases, influences on intelligence and even the very way we define intelligence. I propose to explain some of the common theories of intelligence, consider how these impact on the particular scenario with which I am presented and discuss why I would consider the candidate’s original assessment as unreliable.

The British psychologist Charles Spearman is generally recognized as a major contributor to intelligence testing. He pioneered a means to measure and numerically express intelligence. He proposed the existence of a general intelligence factor ‘g’. Spearman carried out the first formal factor analysis of correlations between various test tasks. The concept of general ability came from the ‘positive manifold’ (Macgregor & Turner, 2015) observed by Spearman. He observed a trend for all such tests to correlate positively with each other. Spearman found that a single common factor explained the positive correlations among tests. He interpreted it as the core of human intelligence which influences success in all cognitive tasks. This interpretation of ‘g’ as a common cause of test performance is still dominant in psychometrics. According to Spearman, the second ‘s’ factor represented various specific abilities; abstract, verbal and numerical, on each test, “but these were of only incidental interest to him.” (Tager-Flusberg and Plesa-Skwerer, 2009)

Spearman’s theory of general intelligence has been much debated by subsequent theorists. Even within those proponents of the existence of a ‘g’ factor, there still does not exist agreement on what it actually means. As stated by Neisser et al (1996) it has been explained as simply statistical regularity (Thomson, 1939), mental energy (Spearman, 1927), abstract reasoning ability (Gustafsson, 1984) and an index measure of neural processing speed (Reed & Jensen, 1972).

Howard Gardner’s book Frames of Mind: The Theory of Multiple Intelligences (1983) first proposed a theory of Multiple Intelligences (MI). Gardner was concerned that the skills and capabilities which are valued in one culture may not be awarded the same status in another society. His focus was not so much on whether a person is intelligent, but rather in what ways they are intelligent. Gardner initially identified seven different bits of intelligence, subsequently adding an eighth in 1988. His intelligence can be identified as verbal/linguistic, musical, logical/mathematical, visual/spatial, bodily/kinesthetic, interpersonal, intrapersonal and naturalist. This broader range of intelligence challenged the traditionally dominating valued intelligence based on verbal/mathematical intelligence and the view of a fixed intelligence. Gardner’s work continues to impact on the work of educationalists and teaching practitioners. The theory of multiple intelligences supports the case for implementing teaching approaches to suit the multiple profiles of pupils. Gardner’s theory has not been without criticism including its effectiveness in having too many constructs to measure, and Sternberg’s (1985) referral to his MI as talents rather than intelligence.

Robert Sternberg (1985) proposed the triarchic model of intelligence. Li (1996, p.37) argues that this is “a comprehensive theory, more encompassing …. because it takes into account social and contextual factors apart from human ability.” His theory categorizes intelligence into three parts: analytical, creative or synthetic and practical. As indicated by Macgregor and Turner (2015) Sternberg’s theory can be applied to evaluating and teaching pupils and offered a focus on matching instruction to strengths of an individual in each area. This is familiar to teaching practitioners in today’s classrooms.

Through time the idea of general intelligence ‘g’ evolved into a hierarchy, with ‘g’ at the top. Raymond Cattell’s research (1971) led to the concepts of fluid and crystallized intelligence, with the theory further developed with John Horn. The Cattell-Horn theory of fluid and crystallized intelligence argues that intelligence is made from different abilities which interact to produce overall individual intelligence. Crystallized intelligence refers to the knowledge and skills acquired throughout life with fluid intelligence the ability to reason, problem-solve and make sense of abstract information. Fluid intelligence is regarded as independent of learning tending to decline in later adulthood. Crystallized intelligence, however, is directly related to learning and experience which tends to increase as people grow older.

Arguably the most influential theory in the study of human intelligence is The Cattell-Horn-Carroll theory (CHC). This combined two previously established theoretical models of intelligence due to their similarities: the theory of fluid and crystallized intelligence (Cattell, 1941; Horn 1965), and Carroll's Three-Stratum theory (1993). Further expansions on the CHC model were developed in the 1990s. CHC presents a three-stratum model whereby the distinct differences in cognitive ability can be classified: stratum I containing over 70 narrow abilities, stratum II with eight broad abilities and the third stratum of the overall general ability ‘g’. The CHC ability domains are incorporated into contemporary intelligence batteries and generally represent five to seven of the broad abilities of fluid reasoning, comprehension-knowledge, short-term memory, visual-spatial processing, auditory processing, long-term storage and retrieval, cognitive processing speed, reading and writing, and quantitative knowledge. Interestingly CHC sparks much debate on the distinction between ability and attainment, with its inclusion of attainments and skills as narrow abilities. Today, CHC theory has developed as the broad basis for most widely used intelligence tests.
Given the evidence available on the candidate in the scenario presented, it should be fair to assume that her original assessment was undertaken at a time when the Spearman model of psychometric testing was largely practiced. Her intelligence was likely to have been assessed by means of a singular construct assessment given that it took place before the publications of Stenberg, Gardener, and Cattell–Horn–Carroll. Her score at that time was most likely considered and reported as a fixed measure of cognitive. It seems unlikely that any overall cognitive profile was considered or compiled. As argued by Jenny Webb and Simon Whitaker (British Psychological Society, 2012), “the issue is not what somebody’s IQ is, but whether they are able to cope, and this comes down to what society as a whole would accept as a minimum standard of quality of life.”

The questions of reliability and validity must be raised with this candidate’s original assessment. We have no evidence of a test which offered norm samples, confidence intervals or appropriate standardization. Webb and Whitaker (British Psychological Society, 2012), also raises the issue of how accurately IQ can be measured in the low range even using the modern commonly used (WISC IV and WAIS-IV). There is no indication as to why the testing was carried out, what testing was used and whether it was reviewed and evaluated as the most appropriate for this particular candidate. Was this assessment part of a hypothesis testing process? As indicated by Macgregor and Turner (Strand 3 p. 24) we should question if a test explicitly focuses on aspects of development important for successful learning, rather than emphasizing only weaknesses and limitations. In this case, it would appear not. Later success in life highlights the unreliability of the concept of fixed intellect with this candidate and would support fluid and crystallized intelligence which can change over time. It is stated that her IQ was considered to be of such a low level that she was almost considered unsuitable for mainstream education.

As questioned by Webb and Whitaker (British Psychological Society, 2012), how helpful is intelligence testing in enabling us to make judgments or predictions about the nature or degree of the supports people need to achieve a good quality of life? Did the original assessment demonstrate utility or ecological validity? It is indicated by Birch, Cline & Gulliford (2015), that knowing the IQ of a child with MLD or SLD does not offer a teacher useful information on ways that child will learn or find it difficult to learn. Hence, a low IQ score, despite any decision on accessing mainstream education or otherwise, would have offered very little in supporting this candidate’s needs to develop her potential. In fact, this early labeling may even have subjected the candidate to bias in subsequent learning and teaching experiences and possibly had a long-term negative impact on her attitudes toward herself and her learning. Adey et al (2007) propose that offering an early low fixed ability score can possibly discourage teachers from trying to develop potential. It could also be questioned how the original score awarded to the candidate might reflect in today’s context bearing in mind the ‘Flynn effect’ phenomena whereby mean IQs have increased a full standard variation over the past 50 years with the rate of gain possibly increasing. (Neisser et al, 1996). As discussed by Webb and Whitaker (British Psychological Society, 2012), “this uncertainty of the Flynn effect means that IQ scores cannot be corrected with confidence and that in effect there is an additional chance error in the scores of the order of about 0.3 of a point per year.”

Bias and fairness issue to be further questioned in relation to the presented scenario. Macgregor and Turner (2015) use the definition by Cleary (1968) which “argues that tests are biased if they predict performance on a criterion measure differently for different groups; what is commonly referred to as differential validity.” Bias concerns are further highlighted by evidence of societal mean IQ increases over time with changes in educational opportunities alongside the narrowing of the gap between black and white IQ scores over time with the improvement of socio-economic conditions for African-Americans. (Neisser et al, 1995). The Russian psychologist Lev Vygotsky (1985) argued that intelligence is social in origin and has potential to develop throughout life. He thought that "language and thought first appear in early interactions with parents, and continue to develop through contact with teachers and others" (Neisser et al., 1996, p. 80). Alyeska (2010, p.1) cites an example of some elementary schools using Wechsler Pre-school and Primary Scale of Intelligence, the Standford-Binet Intelligence Scales and Otis-Lennon School Ability Test to measure the intelligence of children before they can be admitted to the schools.

These assessments, however, are intended to test the intelligence of children who are four years old in a way that is not affected by socioeconomic conditions and the cultural background of the person. Alyeska further explains that administering IQ tests to very young children do not rule out the influence of socio-cultural influence which can influence the test results. There is also no indication if the candidate spoke English as a first or additional language, which would also have an impact on decisions for appropriate assessment. Neisser et al (1996, p.96) notes that “the mean intelligence test scores of Hispanic Americans are somewhat lower than those of Whites, in part because Hispanics are often less familiar with English.” Coard (1971) as referenced in Macgregor and Turner (2015) claimed that IQ tests were the only representative of white middle-class males, as the vocabulary and style of the test were inaccessible to different races and cultures. Another example used in Neisser et al evidences women shoppers in California who had no difficulty in comparing product values at the supermarket were unable to carry out the same mathematical operations in paper-and-pencil tasks (Lave, 1998). We do know that the candidate is female and, although APA 1996 (cited in Macgregor and Turner (2015) indicated no important sex differences, that substantial differences do appear for specific abilities. As there no evidence on the tests administered, it may be possible that the assessment concentrated on visual-spatial abilities and mathematical skills, typically scored higher on by males. Although socio-economic and/or cultural background of the candidate is not stated, it must be considered that the original test administered may have been inaccessible to her and not provided a true representation of her intelligence or abilities.

At the time of the assessment, the candidate indicates that she was experiencing significant stress at home. As cited by MacGregor and Turner (2015), Thorndike concluded in his research any test is subject to variability in outcomes due to factors outside the control of the test situation. Included in these are health, hunger, fatigue and emotional strain. There is no indication that any provision was made for the possible existence of any of these factors.

With a candidate exhibiting delayed physical development, the question must also be asked whether any reasonable adjustments or accommodations were made to ensure that she was not unfairly disadvantaged. The candidate also reports poorly developed social skills. This would raise concerns as to what extent, if any, consideration was given to ensuring the candidate had experience of positive interactions and relationship with the test administrator. Was she provided with an environment where she felt comfortable?

As SENCO and Vice Principal of a school recently awarded the Marjorie Boxall Quality Award for Nurturing Schools, I would be very concerned over the assessment process, conclusions drawn, feedback given and subsequent decisions made on appropriate interventions for a child who had been assessed under such circumstances. The limited information on the candidate at the time of assessment indicates delayed physical development, poorly developed social skills and that she was experiencing significant stress at home. It would appear in today’s context that a global assessment of this child’s social, emotional, behavioral and learning development, alongside appropriately selected intelligence testing with suitable adjustments to accommodate physical needs, would be much more worthwhile. This would allow for more accurate triangulation of information. An assessment of this kind would certainly incorporate a variety of qualitative and quantitative data from a variety of sources and relevant personnel, to compile an overall profile which would ensure the opportunity to display strengths to enhance potential, as recommended in CHC.

Although the case of the candidate presented in this scenario certainly raises many questions and areas for discussion over theories of intelligence, accurately assessing intelligence and appropriately supporting the particular strengths and areas for development of individuals, one must also be cautioned to consider evaluations and conclusions carefully. The context of changing times and constantly developing research, knowledge, and understanding must be considered. Intelligence will no doubt continue to exist as the controversial topic it has always been throughout the history of psychology.

As stated by The American Psychological Association (APA) (1996) and cited in Macgregor and Turner 92015):

“The study of intelligence does not need politicized assertions and recriminations; it needs self-restraint, reflection, and a great deal more research.”