Where Does Intelligence Come From? (page 2)

By — John Wiley & Sons, Inc.
Updated on Jan 1, 2011

What do Intelligence Tests Look Like?

The majority of intelligence tests are developed to measure problem-solving skills in two major areas: verbal ability and visual ability. Tasks that measure verbal ability are usually presented orally and require the child to provide verbal responses. Areas evaluated can include measurements of vocabulary knowledge, verbal reasoning, and practical information. Children who score high on verbal tasks usually have better-developed communication skills, good listening skills, and are likely to be avid readers. On the other hand, visually oriented tasks require hands-on performance (completing puzzles, constructing block designs) or visual reasoning (selecting one picture or design from several visual alternatives). Visual tasks require minimal verbal expression. Some visual tasks are timed (using a stopwatch), resulting in either bonus marks for speed or no points for solving the problem after the given time limit.

How Reliable are IQ Tests?

Any test score can fluctuate from assessment to assessment because of many outside factors, such as illness, fatigue, lack of sufficient rapport, or performance anxiety. However, IQ tests are highly reliable and provide results within a 5 to 10 percent range of accuracy. In the psychological report, there should be a statement to the effect that:

There is a 90 to 95 percent chance that the scores reported are a true indicator of the child's ability, and that the child's IQ score falls within the range of ___ to ___.

Because different IQ tests sample different types of abilities, it is possible that a child's IQ on one test can differ from that obtained on another IQ test. In addition, different environmental conditions present at the time of testing can also contribute to variation in IQ scores. Nonetheless, research has shown that individually administered IQ tests are among the most reliable measures that we have and are a strong predictor of a child's ability to learn academic subjects.

What do the Test Scores Mean?

IQ scores can be reported in several ways: standard scores, percentiles, and age equivalents. Each of these scores is obtained by comparing the child's performance with the available range of scores for his or her peer group.

Here Come the Numbers: Put on Your Safety Belts—We Are in for a Numbers Ride!

In this section, we will talk about how scores from IQ tests can help us understand a child's ability in terms of strengths and weaknesses, and help us form possible expectations for performance. Standard scores are a common measurement that allow us to compare scores across a number of different types of assessment instruments.

Standard Scores

IQ tests report their scores as standard scores. In our initial example, we discussed how measuring a child's temperature provides a comparison with the normal standard score, which for temperature is 98.6 degrees on the Fahrenheit scale. So regardless of which thermometer you use, the numbers will represent the same scale. Similarly, scores on different tests (for example, achievement tests and IQ tests) can be compared if the scores are all using the same scale. Standard scores are based on the normal distribution of scores that look like the bell-shaped curve presented in Figure 7.1. In this distribution, the majority of scores fall in the middle or the average range. The average score (called the mean) is 100, which falls in the exact middle of the bell. If we were to test 2,000 people, we would find that that 50 percent of the people would score above 100 and 50 percent would score below 100. However, we would also find that the majority of people who took the IQ test had scores that deviated from the mean (called the standard deviation) by 15 points on either side. Roughly 68 percent of the population would score within one standard deviation of the mean:

Where Does Intelligence Come From?

  • 34 percent would obtain an IQ score between 85 and 100 (100 – 15);
  • 34 percent would obtain an IQ score between 100 and 115 (100 + 15).

What about 2 standard deviations above or below the mean? How many would score in that area? Well, if we look again at our bell shape, we see that the biggest concentration of population is in the center within one standard deviation on either side. However, as we move toward the ends of the bell on either side, we are losing more and more people. If we looked at these narrow portions under the bell, representing 2 standard deviations (15 + 15) or 30 points from the middle on either side, we would find about 2 percent of the population would score within two standard deviations of the mean:

  • 2 percent would obtain an IQ score at or below an IQ of 70
  • 2 percent would obtain an IQ score at or above an IQ of 130.

View Full Article
Add your own comment