|
What makes a good intelligence test? Originally, Alfred Binet developed the intelligence test to predict future academic success. The value of an intelligence test to predict achievement depends on several important factors: validity, reliability, standardization, objectivity, and practicality.
Validity, as discussed in chapter 1, "The Study of Psychology," refers to the degree to which a test actually measures what it is intended to measure. A valid intelligence test would measure intelligence, not memory, speed, guessing ability, or vocabulary. Reliability is the degree to which a person's score at one time is the same at a different time. A reliable intelligence test would yield consistent scores from one time to the next. Standardization is the process of obtaining a norm, or sample of scores representative of the population, and is necessary in interpreting a particular subject's score. For example, if a friend of yours tells you that he or she received a score of 179 on a test, how would you know whether this was a good score? It would help if you knew that the test contained a possible 200 points, the mean score was 125, and the range for the class was 75 to 190. Psychologists standardize tests so that they can interpret individual test results. The first step is to define the population serving as the standardization group. For example, we might wish to know how an individual compared to other adult Americans, and so we would standardize the test on the American adult population. Or a population might include all high school seniors in Canada. After the population is identified, we need to obtain a norm, or representative sample, that will allow us to describe what the entire population is like. An intelligence test should be objective. Objectivity ensures that a test's results are not affected by the personal feelings and biases of the examiner. Ideally, a test should be constructed so that any qualified person can administer and score it and obtain the same results as any other scorer. An intelligence test should also be practical. Practicality provides that a test can be administered easily and scored in a reasonable amount of time. Chances are good that psychologists would not be interested in a new test requiring five people to administer and 16 hours to complete. Tests that are valid, reliable, standardized, objective, and practical meet the general requirements for use. Developing intelligence tests that meet all of these requirements is a difficult task. |
| Copyright ©2009 The McGraw-Hill Companies. All rights reserved. Any use is subject to the McGraw-Hill Higher Education is one of the many fine businesses of The McGraw-Hill Companies. Please visit our Technical support website at http://mhhe.com/support. |
 |