To understand how intelligence relates to educational achievement it needs to be defined; we need, in other words, to know what intelligence is before we can examine how it can be measured and subsequently related to different levels of achievement.
what is intelligence?
Although on the face of things intelligence might appear relatively easy to describe and demonstrate, Sternberg (1987) suggests that not only is it hugely difficult to define its meaning is also vigorously contested, in that “There seem to be almost as many definitions of intelligence as there are experts asked to define it”. For convenience, however, we can group these definitions into two broad categories:
1. Capacities: This general categorisation of intelligence is based on the idea humans have certain faculties, aptitudes and competences that allow them to behave “more intelligently” than other animals. Sternberg’s (1986) triarchic theory, for example, argues intelligence has three related components:
• Meta-components involve the capacity to solve problems and make informed, correct, decisions.
• Performance components involve the ability to actually carry-out meta-component actions, such as seeing a relationship between two or more ideas.
• Knowledge-acquisition components refer to the capacity to acquire new information and make logical choices between different options.
Intelligence, in this respect, is defined in terms of the ability to use each capacity to process information and choose appropriate responses, depending on the particular situation.
2. Abilities: This is a narrower definition, one that focuses on the ability to perform particular tasks or solve specific problems. Binet (1916), for example, the creator of one of the first tests to measure intelligence, defined it as “the capacity to judge, reason and comprehend well”. Jensen (1973) argues that while human intelligence is complex and difficult to precisely define, it is possible to test and quantify some important aspects of what he calls “general intelligence” or the “g” factor, a subset that relates to “abstract reasoning ability”. By focusing on abilities, therefore, it’s possible to develop tests that measure the extent to which individuals are able to do things like identify rules, patterns, reasons and logical principles in three particular areas:
• Verbal or Comprehension
While categories based on capacities or abilities recognise different dimensions to “intelligence” it is still something conceptualised in the singular – it’s something you have or you do not. Gardner (2003), however, has introduced a further layer of complication through his theory of multiple intelligences. This questions the assumption “intelligence is a single entity” passed between generations, with children inheriting their parents’ general intelligence.
Gardner (1999) argues there are at least 7 distinct types of intelligence ranging from the conventional linguistic, mathematical and spatial abilities through musical intelligence to interpersonal IQ – the extent to which an individual can empathise with others. This latter form is sometimes called emotional intelligence and, as Ogundokun and Adeyemo’s (2010), study of Nigerian secondary school students found, it is possible to measure and quantify. They found, for example, a strong correlation between levels of emotional intelligence and academic achievement.
While there is a general agreement that something called “intelligence” exists and can be defined, there are serious disagreements about what it is and how it should be defined. In addition, as we’ve seen, although it’s possible to measure some aspects of intelligence, such as mathematical or verbal abilities, disagreements remain about what is being measured and how it can be reliably and validly measured. In an educational context, however, the most common tests use a quantitative calculation of an intelligence quotient (IQ); a measure, not a definition, of individual intelligence, where an IQ of 100 is taken as the average for any population. Such tests generally attempt to measure some or all of the following:
• verbal reasoning – the ability to think constructively.
• abstract/visual reasoning – the ability to solve non-verbal or spatial problems
• quantitative reasoning – the ability to solve mathematical problems.
• working memory – the ability to process information and make connections between aspects of knowledge
• fluid reasoning – the ability to think logically and to solve problems.
While question-types are many and varied their normal format is multiple-choice because this allows relatively straightforward quantification. Questions usually vary in complexity the further the individual goes in the test and each test normally has a strict time-limit.
A typical comprehension question, for example, might take the following format:
Q. Which of the following below is closest in meaning to “Reassuring”?
Mathematical comprehension, on the other hand, may involve questions that require identification of a number pattern:
Q. Which of the following numbers follows this series: 5 – 6 – 10 – 6 – 15 – 6 –
The purpose of IQ testing in education varies around the world; in the UK, for example, it was extensively used between 1950 and the mid-1970s to separate children at 11 into different schools; those who passed the 11+ were eligible to attend grammar schools and follow a broadly academic curriculum, while those who failed attended secondary modern schools that followed a broadly vocational curriculum. Although most children currently attend Comprehensive education, for which there is no formal entry test, grammar schools still exist in some parts of the country and 11+ IQ tests are used to control entry to these schools. In America, on the other hand, entrance to higher education is partly controlled through Scholastic Aptitude Tests (SATs) that cover the skills of critical reading, essay writing and mathematics.
Although IQ tests are simply one of a range of tests used in education systems their significance lies in the special status they claim as objective tests of innate intelligence; the claim IQ tests not only reliably and validity measure “intelligence”, but that they do so independently of cultural influences such as class, gender, age or ethnicity. For their advocates, therefore, IQ tests are an important tool for revealing the natural variations in intelligence within and between different individuals and populations; people are born with a certain level of intelligence inherited from their parents that does not vary greatly throughout their lifetime – although there is always some level of variation; tests scores are usually expressed as a range (between 90 and 110, for example), rather than a single number such as “an IQ of 100”.
These ideas have significant implications for the relationship between intelligence and education achievement; if an education system is meritocratic – pupils are neither disadvantaged nor discriminated against on the basis of anything other than their individual merits, regardless of their class, gender or ethnicity – any differences in achievement can be explained by natural differences in intelligence.
Internal criticisms of intelligence testing focus on the construction and application of the tests themselves, largely in terms of arguments about their claim to be “culture-free” or “culture-neutral”; the argument here is that rather than being a measure of “natural intelligence” IQ tests actually measure cultural learning.
Where IQ tests are timed for example – a student has to answer questions within a strict time limit – those very familiar with the question formats have a significant advantage over those who are not In other words, students who practice answering IQ tests have an in-built advantage because to answer a question the student needs to initially understand what it’s asking; those who have to spend time thinking about how to answer a question have less time in which to actually answer it. Murray (2013), for example, argues that the ability to pay for private test tuition gives children “from more affluent families an unfair advantage over those from less well-off backgrounds“.
A further criticism is that tests of verbal reasoning and comprehension, especially in societies with a diverse range of ethnicities, make unwarranted cultural assumptions that disadvantage children from ethnic minorities. These relate, for example, to questions that assume knowledge widely known or “taken-for-granted” in the dominant culture is similarly widely known in minority cultures. In America, for example, the idea of “plastic money”, such as credit cards, is part of the cultural discourse – something “everyone is assumed to know”. Students from cash-based countries in Eastern Europe, such as Belarus, however, would not be familiar with this type of assumption.
While these types of cross-cultural assumptions may be important, a more significant criticism of IQ tests relates to subcultural assumptions about cultural knowledge; that is, differences in cultural learning between those born and raised within a particular society who are separated by differences in class and ethnicity. Dove (1971), for example, demonstrated that IQ tests assume particular kinds of cultural knowledge that bias them towards particular (in this instance, white American) ethnic groups; one consequence of this bias is that those familiar with the cultural assumptions contained in the tests have higher measured IQ scores. He also showed how, by reversing these assumptions in favour of minority, culturally different and disadvantaged groups, levels of measured intelligence could be reversed. Dove’s “Chitling IQ Test”, for example, asked the following types of question:
1. A “handkerchief head” is:
a. a cool cat
b. a porter
c. an Uncle Tom
d. a hoddi
e. a preacher.
2. Which word is most out of place here?
An ability to answer these questions correctly depends on an assumed level of cultural knowledge not equally shared throughout a tested population – something that biases both the test and the outcome (as you were probably wondering, “c” is the correct answer to each question, although as to why I’m afraid I have no idea. Which either means I lack the intelligence to know or that the test is culturally biased…).
A different type of internal criticism relates to the idea that the higher the level of a student’s intelligence the less likely they are to be able to correctly answer some types of question; they consider possibilities not seen as problematic by the setter. The following question, for example, assumes there is only one possible distinction and therefore only one possible correct answer:
Identify the Odd-one-out:
HOUSE IGLOO BUNGALOW OFFICE HUT
In this question the “correct” answer is “office” because it’s not a dwelling – but equally valid, if incorrect, answers could be:
• “hut”, because it’s the only one made of mud
• “Igloo” because it’s the only one made from snow.
Kaplan (1998), in this respect, concludes that “How well a person does on an IQ test depends on a variety of factors besides intelligence“. These include:
• reading habits
• experience with and attitudes toward taking tests
• cultural upbringing
• mental and physical health.
He concludes “Count me among those who regard the study of intelligence as more pseudo-science than science“.
External criticisms of IQ testing relate to wider ideas about the validity of such tests and take a number of forms relating to both the nature of “intelligence” and what, if anything, is measured by IQ tests.
Burden (2004), for example, argues the concept of “intelligence” is both problematic – it is too complex to be reduced to simple forms of “intelligence testing” – and context-dependent:
“It’s a hypothetical construct psychologists have used to describe how people behave. How I would behave in the Amazon jungle is a lot less intelligently than I would in my job as a psychology professor. And if something goes wrong with my car I open the bonnet and hope someone will come and help me“.
Other types of criticism focus on the construct validity of IQ tests. Flynn (1987), for example, argues “psychologists should stop saying IQ tests measure intelligence. They should say that IQ tests measure abstract problem-solving ability (APSA), a term that accurately conveys our ignorance“. In other words, while IQ tests used in education systems claim to measure something called “intelligence”, what they actually measure are two possible types of intelligence; those that involve linguistic and logical-mathematical abilities.
This is significant in the context of understanding the relationship between intelligence and educational achievement because it suggests that out of all the possible measures that could be used, decisions have been made about which are considered valid. The implication here is that what counts as “intelligence” and how it relates to educational achievement are culturally-loaded constructs; any relationship between the two is biased in favour of those social groups who have the power to both define how intelligence is measured and how it is realised or expressed in terms of educational achievement.
The concept of achievement is also something that can’t be taken-for-granted because it is culturally defined; what counts as achievement is both:
• relative across different societies: in the UK, for example, it is officially measured as passing 5 GCSEs at grades A* – C while in India it is more-likely to be measured in terms of literacy rates. Ramachandran (2007), for example, notes “Only 66% per cent of the Indian people are literate – 76% of men and 54% of women.
• variable within societies: since 2011 official UK government measures specify valid GCSE achievement can only be in the following subjects: English, mathematics, history or geography, two sciences and an ancient or modern foreign language.
What IQ tests measure, therefore, is at best a subset of “intelligence”. Just as convicted criminals are a subset of everyone who commits a crime, studying this group may tell us something about their characteristics and motives. We can’t, however, safely generalise from convicted offenders because we don’t know how representative they are of all offenders. Similarly, while testing and measuring a limited subset of intelligence doesn’t make IQ tests invalid or methodologically “biased”, it does raise important questions about why they are designed to measure these abilities but not others.
One explanation is that they measure those abilities most valued by powerful social groups, such as a ruling class in Marxist terminology; IQ testing is part of a process of cultural reproduction whereby the power to define and “objectively measure” intelligence” is a valuable social resource, for two reasons:
1. Intelligence is defined in ways that reflect the particular class, gender or ethnic interests of powerful groups.
2. If subordinate social groups accept, or are unable to challenge, this definition it both cements their lower position (they are “less intelligent”) and justifies any differential treatment they receive. Children of the upper and middle classes, for example, achieve more not because of their privileged position but because they are simply “more intelligent” and hence more deserving. It similarly serves as a justification for both social and academic segregation because it is based on “objective” criteria. In the UK, for example, the tripartite system placed children with different levels of measured IQ in different schools and this tended to produce a form of class segregation; grammar schools selected mainly middle class pupils while secondary modern schools were largely the preserve of working class pupils.
From this perspective IQ tests and the concepts of intelligence they embody are part of the ideological state apparatus. If people can be convinced natural intelligence differences exist and can be objectively measured this represents a powerful form of social control. Gardner, for example, calls this “the IQ way of thinking: that people are either smart or not, that there’s nothing much you can do about it, and that tests can tell you if you are one of the smart ones“.
Additional criticisms of the cultural biases of IQ tests focus on their ecological validity. One problem with IQ testing is that it takes place under artificial conditions – not just those of “the test” itself but also, as Flynn argues the underlying premises on which such a form of intelligence testing is based: “We know that people solve problems on IQ tests; we suspect that those problems are so detached, or so abstracted from reality, that the ability to solve them can diverge over time from the real-world problem solving ability called intelligence; thus far we now little else“.
To put this another way, to paraphrase Cohen and Shelley (1982), what IQ tests measure is the ability to complete IQ tests.
Part 2 of this short series looks at three different takes on the relationship between educational acheivement and intelligence.