I. Norm-Referenced Measurement

1. Definition - In norm-referenced testing an examinee's performance is compared with the performance of a specific group. A raw score alone tells us nothing. What does saying someone got 17 out of 20 right mean. This allows us to compare a person's score with the typical performance of a group, usually one representative of the population of the country. Then we know how someone is performing compared to other kids their age.


2. Representativeness - This is extent norm group is characteristic of the population. Should match as closely as possible demographics of population. Usually use age, grade level, gender, geographic region, ethnicity, race, and SES.


3. Size - Should be large enough to insure stability of test scores and inclusion of all groups that are represented in population. Should include at least 100 subjects at each age and grade level.

4. Relevance - What is the norm group you are selecting? National norms are usually used in IQ tests. For educational achievement, local norms may be more appropriate. May want to compare only against their race or ethnic background. Depends on what you want to use the results for.


II. Derived Scores

1. Age and Grade Equivalent Scores - This is arrived at by determining the average performance of children at an age or grade. You therefore find out at what age or grade the average child got the score of the examinee and that is their age or grade score. It does not mean that a child is functioning at that level. Grade scores should never be used. They are too misleading.

2. Ratio IQ - This is the first way that IQ's were calculated. It is MA/CA x 100. This generally is not used anymore and if you see a test that uses this procedure you should question its utility.

3. Percentile Ranks - Allows us to determine an individual's position relative to the standardization sample. It is the point in the distribution at or below which that percentage of individuals fall. They are easy for people to understand, especially parents.

4. Standard Scores - The deviation IQ is a version of a standard score. These are raw scores that have been transformed to have a given mean and standard deviation. They express how far an individual's score lies from the mean in terms of the standard deviation.

III. Reliability

1. Definition - Reliability refers to the stability of measurement. That is, it is reproducible and stable. It is expressed by a reliability (correlation) coefficient (range 0 to 1.0). A test should not be trusted if its reliability is low. Generally a reliability of .80 or higher is considered acceptable.

2. Test-Retest Reliability - An index of stability. You administer the same test to the same people at two different occasions, usually a short period of time (2 weeks to a month). The correlation represents the extent to which the score is stable over time.

3. Internal Consistency Reliability - Also called split-half reliability. Divide a test into two equivalent halves (usually odd versus even) and then obtain correlation.


IV. Validity

1. Definition - Generally it refers to does the test measure what it is supposed to measure? How meaningful is your result? Tests can be used for different purposes and therefore the validity of a test can vary depending on its use. IQ tests are a valid predictor of a child's success in school. It is not a valid predictor of your ability to be a janitor. Tests should be used only for those purposes for which its validity has been established. A test must be reliable to be valid but it can be reliable without being valid. Reliability is necessary but not sufficient.

2. Content Validity - Do the items on the test cover the domain that the test purports to measure. If you are measuring intelligence, does test cover all aspects of intelligence. It should not be confused with Face Validity which refers to what it appears to measure. It does not necessarily mean it has content validity.

3. Criterion-Related Validity - Refers to relationship between test score and some type of criterion or outcome such as a rating or other test score. Two types: (a) Concurrent Validity - How it relates to some other currently available measure. If came up with a new IQ test would see how it relates to standard in field, the WISC-R. (b) Predictive Validity - Refers to correlation between test score and performance on relative criterion after a period of time has passed. Is the test a predictor of future performance. Is IQ a predictor of future school performance?

4. Construct Validity - How does the test relate to the theoretical constructs that the test purports to measure. For example, does an IQ test measure the construct of intelligence the way the author designed. Usually done through factor analysis. WISC-R, does it break down to verbal and performance sections.


V. Definitions of Intelligence

1. Binet (1916) - "...judgement, otherwise called good sense, practical sense, initiative, the faculty of adapting one's self to circumstances. To judge well, to comprehend well, to reason well, these are the essential activities of intelligence."

2. Terman (1923) - The ability to carry on abstract thinking.

3. Stoddard (1943) - "...the ability to undertaker activities that are characterized by (1) difficulty, (2) complexity, (3) abstractness, (4) economy, (5) adaptedness to goal, (6) social value, and (7) the emergence of originals, and to maintain such activities under conditions that demand a concentration of energy and a resistance to emotional forces."

4. Freeman (1955) - "...adjustment or adaptation of the individual to his total environment, or limited aspects thereof....the capacity to reorganize one's behavior patterns so as to act more effectively and more appropriately in novel situations...the ability to learn...the extent to which a person is educable...the ability to carry on abstract thinking...the effective use of concepts and symbols in dealing with a problem to be solved..."

5. Wechsler (1958) - The aggregate or global capacity of the individual to act purposefully, to think rationally, and to deal effectively with his environment."

6. Das (1973) - "....the ability to plan and structure one's behavior with an end in view."

7. Humphreys (1979) - "....the resultant of the process of acquiring, storing in memory, retrieving, combining, comparing, and using in new contexts information and conceptual skills."

8. Conclusion - I don't care if you know all the different theories I just gave or the ones given in the book. Just remember that there are two major divisions in theories. Those that view intelligence as a unitary construct and those that see it is having many components.


VI. History of Intelligence Testing

1. Roots - Interest in intelligence and intelligence testing began in earnest during the latter part of the 19th century as psychology began to emerge as a disciple of its own.

2. Galton (1869) - Some regard him as the father of the testing movement. He was concerned with the study of individual differences and began studying mental inheritance.

3. Binet (1905) - With Simon developed the Binet-Simon Scale, the first test of intelligence to have valuable practical application.

4. Terman (1916) - Published the Stanford Revision and Extension of the Binet-Simon Intelligence Scale.

5. Wechsler - Published the Wechsler-Bellevue Intelligence Scale in 1939. Since that time the Wechsler scales (WAIS, WISC, WPPSI) have become the most frequently used tests



VII. Intelligence Classifications

1. Profound Retardation - IQ < 20

2. Severe Retardation - IQ = 20-39

3. Moderate Retardation - IQ = 40-54

4. Mild Retardation - IQ = 55-69

5. Borderline Retardation - IQ = 70-79

6. Dull Normal - IQ = 80-89

7. Average - IQ = 90-109

8. Bright Normal - IQ = 110-119

9. Superior - IQ = 120-129

10. Very Superior - IQ = 130+


V. Intelligence Tests Used Today

1. WISC-3 - Appropriate for children age 6 through 17. It is the most commonly used test today. Provides you with a Verbal IQ (measure of verbal comprehension), Performance IQ (perceptual organizational skills), and Full Scale IQ.

2. WAIS-R - Appropriate for individuals ages 16 through 74. Most commonly used adult test. Provides you with same scores as the WISC-R.

3. WPPSI-R - Appropriate for children 3 to 7 years of age. Also gives VIQ, PIQ, and FSIQ.

4. Stanford-Binet - Revision just came out a little more than a year ago. Appropriate for ages 2 to adult. Provides measures of verbal reasoning, abstract/visual reasoning, quantitative reasoning, short-term memory, and a composite IQ.


VIII. Stability of IQ

1. Infant Intelligence Tests - Tests of IQ during first year of life do not accurately predict IQ performance later in childhood although they are useful in identifying motor problems or extreme IQ deficiencies. These tests are made up primarily of sensorimotor tasks.

2. Later Testing - From about age 8 on the tests become very good predictors of adult intelligence although there is still a lot of variability.

3. Prediction - IQ tests were generally developed to predict how well someone will do in school. They do this very well.


IX. The Genetic Hypothesis

1. Controversial - This is a very controversial area that explains differences between groups (black and white) in intelligence is inherited.

2. Arthur Jenson - The strongest proponent of this position. He believes that lower-class and black children are not as good as middle class and white children in abstract reasoning and problem solving (higher level intellectual abilities).

3. Study - He was initially not looking to test a genetic hypothesis but rather he was looking for a cumulative deficit. This would be an increase in decrement of IQ with age as a result of environmental deprivation.

4. California Study - First attempt to find this. Looked at children 5-10 years. Found age decrement for black children on verbal IQ. Also found that overall, blacks scored about 1 standard deviation below whites.

5. Georgia Study - To further test the hypothesis of a cumulative deficit he looked to find the worst environment he could. Found it in rural Georgia. Tested children in grades K-12 in an integrated school (1/2 white, 1/2 black). Blacks were very low SES, whites were low to low middle SES. Findings: (a) IQ - White = 102 Black = 71 (b) significant age decrement in blacks but not whites in both verbal and performance IQ. He concluded difference due to heredity.

6. Results and Types of Intelligence - Level I intelligence is rote memory. Mostly short-term memory tasks like digit span. Level II intelligence involves more complex cognitive processing and problem solving skills. Jenson found the Black-White difference on Level II only. He also found Blacks to do the worst on the least culturally loaded tests and best on vocabulary and information which are the most culturally loaded. Other studies have not confirmed this however.

6. Criticisms - If true would expect mixed-race persons to be lower than whites. When compared to white children of similar backgrounds there was no difference in IQ. In general there is no evidence to support this or deny it. That is why it continues to be perpetuated.


X. Heredity - Environment Issue

1. Studies in Families - (a) In general, the more the genetic similarity, the higher the correlation. (b) It does however support the role of both heredity and environment..

2. Study With Adopted Children - Look at scores of children adopted before 6 months into middle SES homes. Group I: Mother had IQ<75. Group II: Biological fathers occupation was at very low level. Group III: Biological parents both I and II. Results: Tested children at 8 years. I, IQ=105; II, IQ=110; III, IQ=104.

3. Conclusion - Heredity may set limits but the rest is determined by how much stimulation you get in your environment.