When the "father of intelligence tests", Alfred Binet, was designing the first intelligence test, he had a clear vision of how he thought they should be used. Paris had just passed a law that required every child to attend school and suddenly had to determine the intellectual abilities of every child. Binet's goal was to design a test to help identify the children who might need a little extra attention to give them the best chance to learn. Since then, there have been a lot of criticisms of intelligence testing and whether it may be more harmful than helpful. So, where did things go wrong?
- What are the disadvantages of intelligence tests?
- What determines the reliability of intelligence tests?
- What determines the validity of intelligence tests?
- Is there cultural bias in intelligence tests?
- What are the criticisms of intelligence tests in children?
Disadvantages of Intelligence Tests
We've all taken them. Intelligence testing is essentially the cornerstone of education, especially in the United States. You must take an intelligence test to pass a class, move to a different grade, or get accepted into a school.
An intelligence test is a type of assessment to measure an individual's capacity to learn, think, and reason compared to others. This test uses a numerical score.
Since Binet's test, several other intelligence tests have been formed, including the Stanford-Binet (a revision of Binet's original test), the Weschler Adult Intelligence Scale (WAIS), and the Weschler Intelligence Scale for Children (WISC). Intelligence testing has proved helpful in many ways, but intelligence tests have several disadvantages.
For one, think about how intelligence tests are created. Many intelligence tests report the rating using a scale called the intelligence quotient (IQ)
Intelligence quotient (IQ): originally defined as the ratio of mental age to chronological age, then multiplied by 100 to remove the decimal point.
Any rating scale needs a baseline, but who determines that for intelligence testing? The answer is relative to how other people do on the same test. This can be a significant disadvantage of intelligence testing as a person's "intelligence" can change based on context.
Fig. 1 - Different types of intelligence
But it's important to note that intelligence is not the same as a person's IQ. A person can be intelligent in several different ways. A major criticism of intelligence testing is that it does not measure other essential types of intelligence, such as:
emotional intelligence
social skills
moral development
motivation
There is no way to quantify intelligence perfectly. Intelligence testing only tells part of the story and only offers a relative measure of intelligence.
Reliability of Intelligence Tests
Another criticism of intelligence testing is in cases where they lack reliability. Measuring a person's intelligence relative to a standardized group is one thing. But these results won't say much about a person's intelligence if the tests lack reliability.
Reliability refers to the overall consistency of an intelligence test.
For a test to be reliable, it must give consistent scores when it's retaken. The best way to test the reliability of an intelligence test is to administer it several times and in several different ways. Here are some methods researchers may use to read minister the same test to check reliability:
If the correlation between the original and retest scores is high, then we can say that the test's reliability is high, and vice versa.
Say that researchers tested a group of 7-year-olds and then retested them a few decades later. They found a correlation coefficient of about +.85. Since the correlation coefficient is high, we can say that the intelligence test they administered has high reliability.
Validity of Intelligence Tests
Reliability is not the only hurdle intelligence tests must overcome. They also must have validity, or more specifically, predictive validity.
Validity refers to how well a test measures what it's supposed to measure.
Predictive validity refers to how well a test can predict future behavior the way it's supposed to predict.
In other words, does the intelligence test predict future performance how it's supposed to?
Even if a test has high reliability, it does not necessarily mean high validity.
If you were to use a scale that wasn't properly calibrated (i.e., it read 5 pounds when weighing a 3-pound object), your results would likely have high reliability (they're consistent), but they would not be valid.
We can see if a test is valid by comparing a person's intelligence score to their later achievement. If the correlation is high, we can say the test has high validity. A common criticism of intelligence testing is that the predictive validity seems to peak during a person's early school education and then diminishes or weakens.
Fig. 2 - Intelligence testing with multiple choice
When comparing intelligence scores to school performance, psychologists have made the following observations:
- Ages 6 to 12 had a correlation of +.6
- Between the ages of 11 and 16, correlation peaks at +.81. Predictive validity is highest at this point)
- When taking the SAT in the late teens, the correlation begins to decline to +.5.
- Finally, by early to mid-adulthood, when one may be taking the GRE, the correlation decreases even more to +.5. Predictive ability is lowest at this point.
Cultural Bias in Intelligence Tests
Cultural differences can significantly impact several areas of psychology, including intelligence testing. Another criticism of intelligence testing is they are prone to cultural bias.
Cultural bias: refers to factors that interfere with the results of intelligence tests across cultural groups.
If an intelligence test is riddled with cultural bias, it will be impossible to measure a person's intelligence successfully. Cultural differences can impact how we feel, experience, and think about the world. An intelligence test can have three types of cultural bias -- construct bias, method bias, and differential item functioning.
1. Construct bias -- everything does not have the same meaning in every culture. Something could be good in one country and bad in another. Construct bias refers to the systemic differences in meaning that exists between cultures.
One culture may define a "good husband" differently than another culture. Intelligence tests should be careful to acknowledge these differences to avoid creating cultural bias.
2. Method bias -- this type of bias refers to situations in which a test procedure differs across cultures.
Individuals from one country might be used to inputting data into a computer while individuals from another country might be used to writing everything down. Intelligence testing should consider these differences.
3. Differential item functioning -- a bias occurs when there is a significant gap in the test scores of individuals from different cultures who display the same abilities. Even if a group of people has the same understanding of a certain topic, their cultural differences may affect how they respond. Intelligence testing that ignores this is vulnerable to cultural bias.
On a Colorado history exam, students who grew up in Colorado may have different test scores than individuals who grew up elsewhere, even though they are all equally familiar with its history.
Culture-free intelligence tests are being made available to avoid cultural bias. These tests help produce a more neutral assessment of intelligence that keeps potential cultural differences in mind.
Criticisms of Intelligence Tests in Children
Remember Binet's original vision for intelligence testing, to identify students who needed additional attention in order to optimize their learning? While intelligence tests are still used for this purpose today, they are also often used to identify the highest-scoring children as well. While this may seem well-intentioned, some critics say that this runs the risk of forcing children into rigid boxes.
For example, many schools have a "gifted child" program in which children with higher scores are placed into special classes. Criticisms of intelligence tests in children say that grouping children by aptitude can create a self-fulfilling prophecy.
Self-fulfilling Prophecy: a phenomenon in which a person's prediction of another or their own behavior actually causes that very behavior.
So children in the "gifted" classes are given more academic enrichment opportunities that are not available to their peers—labeling children in this way and denying others the same enrichment opportunities risks widening the achievement gap in schools.
This type of grouping can also affect minority and low-income students because they are more likely to be placed in lower academic groups. This can, in turn, contribute to the occurrence of prejudice and segregation in education.
Criticism of Intelligence Testing - Key takeaways
- A major disadvantage of intelligence tests is that they can be relative in nature. They also do not measure other factors of intelligence including emotional intelligence, social skills, moral development, and motivation.
- Reliability refers to the overall consistency of an intelligence test. For a test to be reliable, it must give consistent scores when it's retaken.
- A common criticism of intelligence testing is that the predictive validity seems to peak during a person's early school education and then subsequently diminishes or weakens.
- If an intelligence test is riddled with cultural bias, it would make it impossible to successfully measure a person's intelligence. Cultural differences can impact how we feel, experience, and think about the world. An intelligence test can have three types of cultural bias -- construct bias, method bias, and differential item functioning.
- While intelligence tests are still used for this purpose today, they are also often used to identify the highest-scoring children as well. While this may seem well-intentioned, some critics say that this runs the risk of forcing children into rigid boxes.
Explanations 
Exams 
Magazine 
