5 Steps to a 5 AP Psychology, 2010-2011 Edition (58 page)

Performance tests in which there is a correct answer for each item can be divided into two types,
speed tests
and
power tests. Speed tests
generally include a large number of relatively easy items administered with strict time limits under which most test takers find it impossible to answer all questions. Given more time, many test takers would probably score higher, so differences in scores among test takers are at least partly a function of the speed with which they respond. This differs from power tests, which allot enough time for test takers to complete the items of varying difficulty on the test, so that differences in scores among test takers are a function of the test taker’s knowledge, and possibly good guessing.

Another way tests can be categorized is into ability, interest, and personality tests, which are relevant to decision making. General mental ability is particularly important in scholastic performance and in performing cognitively demanding tasks. Interests influence a person’s reactions to and satisfaction with his/her situation. Personality involves consistency in behavior over a wide range of situations. (For personality tests, see
Chapter 14
.) Ability tests include
aptitude tests
designed to predict a person’s future performance or to assess the person’s capacity to learn, and
achievement tests
designed to assess what a person has already learned. For example, the SAT is designed to measure potential to do well in college, whereas the AP Psychology test is designed to measure your mastery of the material in this course of study. Interest tests use a person’s descriptions of his/her own interests to predict vocational adjustment and satisfaction. For example, the current version of the Strong-Campbell Interest Inventory, which is the most widely used vocational interest test, is based on the assumptions that responses that are similar to a particular occupational group and different from people in general provide key information about occupational interests, and that interests can be measured.

Also, there are group tests and individual tests. Standardized tests that can be administered in groups are much more widely used than individual tests administered to one person by a trained professional. Whereas group tests require a test taker to work alone on a structured task and respond to questions; individual tests require social interaction between the examiner and test taker, and require test takers to respond to a person. The test taker needs to view the examiner as trustworthy, competent, and nonjudgmental. The psychologist or other trained professional must use sound professional judgment in eliciting and scoring responses to test items. The differing roles of examiners in individual versus group tests can significantly affect the responses of test takers. Group tests are better standardized and more efficient than individual tests, but individual tests provide more information on test behavior, can be given to test takers who cannot sit for group tests, and can sometimes elicit more creative responses. The most popular individual intelligence test, the Wechsler Adult Intelligence Scale-III and the Stanford-Binet Intelligence Scales, exemplify individual exams. Examples of group tests are the widely used Armed Services Vocational Aptitude Battery (ASVAB) employed by the military to screen recruits and assign them to various jobs, training programs, and career paths; and the SAT and ACT (American College Test).

Because of the potential for abuse, ethical standards guide the development and application of tests. Numerous professional organizations, including the American Psychological
Association, have produced documents detailing appropriate technical and professional standards for construction, evaluation, interpretation, and application of psychological tests to promote the welfare and best interests of the client, guard against the misuse of assessment results, respect the client’s right to know the results, and safeguard the dignity of test takers. Psychologists need to obtain informed consent and guarantee confidentiality in personnel testing, for example. Tests should be used for the purpose for which they were designed by professionals trained in their use.

Because some groups (such as African Americans) have tended to score lower on average than other groups (such as European Americans) on intelligence tests and SATs, critics argue that such tests are biased. Since these tests predict school achievement of all races equally well, the major tests are not biased with respect to predictive validity. However, they do seem biased with respect to performance differences resulting from cultural experience. Biologically oriented theorist Arthur Jensen attempted to succeed where Galton failed in developing a culture-free measure of intelligence by measuring reaction time, but his test is inadequate to represent a measure of intelligence. Several attempts at creating culture-reduced tests that measure general intelligence, such as Raven’s Progressive Matrices, have not succeeded in eliminating the difference in mean scores.
Culture relevant tests
that incorporate skills and knowledge related to the cultural experiences of the test takers may be more successful.

Since intelligence is a construct, it can only be defined by the behaviors that indicate intelligence, such as the ability to learn from experience, solve problems, use information to adapt to the environment, and benefit from training. Because intelligence tests are common and have been used so widely, they have influenced the definition of intelligence; sometimes a score is used to define someone’s intelligence. Intelligence is sometimes reified.
Reification
occurs when a construct is treated as though it were a concrete, tangible object. Intelligence test developer David Wechsler said, “
Intelligence
, operationally defined, is the aggregate or global capacity of the individual to act purposefully, to think rationally, and to deal effectively with his environment.”

Modern ability testing originated with Charles Darwin’s cousin, nativist Francis Galton, who measured psychomotor tasks to gauge intelligence, reasoning that people with excellent physical abilities are better adapted for survival, and thus highly intelligent. James McKeen Cattell brought Galton’s studies to the United States, measuring strength, reaction time, sensitivity to pain, and weight discrimination, using the term “mental test.” Although Galton and Cattell’s measurements correlated poorly with reasoning ability, they drew attention to the systematic study of measuring cognitive and behavioral differences among individuals. At about the same time, French psychologist Alfred Binet was hired by the French government to identify children who would not benefit from a traditional school setting and those who would benefit from special education. He thought intelligence could be measured by sampling performance of tasks that involved memory, comprehension, and judgment. He collaborated with Theodore Simon to create the Binet-Simon scale, which he meant to be used only for class placement.

Binet thought that as we age, we become more sophisticated in the ways we know about the world and that, therefore, most 6-year-olds answer questions differently from 8-year-olds.
As a result of their responses to test items, children were assigned a
mental age
or
mental level
reflecting the age at which typical children give those same responses. Although mental age differentiates between abilities of children, it can be misleading when a 6-year-old and an 8-year-old, for example, have mental ages 2 years below their actual (chronological) ages. The younger child would be proportionally further behind peers than the older child. German psychologist William Stern suggested using the ratio of mental age (MA) to chronological age (CA) to determine the child’s level of intelligence.

In adapting Binet’s test for Americans, Lewis Terman developed the Stanford-Binet Intelligence Scale reporting results as an IQ,
intelligence quotient
, which is the child’s mental age divided by his/her chronological age, multiplied by 100; or MA/CA × 100. A 10-year-old who answers questions typical of most 12-year-olds has an IQ score of 120. Another 10-year-old youngster who answers questions typical of an 8-year-old scores 80. With the development of intelligence tests for adults, the ratio IQ becomes meaningless and has been replaced by the deviation IQ determined as a result of the standardizing process for a particular test. For the fifth edition of the Stanford-Binet Intelligence Scale for Adults, the test has been standardized with a representative sample of test takers up to age 90. Fluid reasoning, visual-spatial processing, working memory, and quantitative reasoning seem to peak in the 30s, whereas knowledge seems to peak in the 50s.

The newest version assesses each of five ability areas, such as knowledge, fluid reasoning, and quantitative reasoning, both nonverbally and verbally. By combining these subtest scores, one IQ score is determined.

David Wechsler developed another set of age-based intelligence tests: the Wechsler Preschool and Primary Scale of Intelligence (
WPPSI
) for preschool children, the Wechsler Intelligence Scale for Children (
WISC
) for ages 6 to 16, and the Wechsler Adult Intelligence Scale (
WAIS
) for older adolescents and adults. The latest edition, the WAIS-III, has a verbal scale including items on comprehension, vocabulary, information, similarities, arithmetic, and digit span; and a performance scale including items dealing with object assembly, block design, picture completion, picture arrangement, and digit symbols. Wechsler based his measures on deviation IQs or how spread out the scores were from the mean of 100 (
Figure 15.1
). Since intelligence has a bell curve distribution, 68% of the population will have an IQ between 85 and 115. These test takers are considered to be low normal through high normal. Test takers who fall two deviations below the mean have a score of 70, typically considered the borderline for
mental retardation
, while test takers two standard deviations above the mean have scores of 130, sometimes considered intellectually gifted, and those three standard deviations above the mean have scores of 145, sometimes considered geniuses. The Wechsler tests are judged more helpful for determining the extremes of intelligence at the mentally retarded and the genius level than the Stanford-Binet. They also help indicate possible learning disabilities when a child’s performance IQ is very different from his/her verbal score.

Some people prefer the term
cognitively disabled
rather than mentally retarded. Degrees of mental retardation vary from mild to profound. To be considered mentally retarded, an individual must earn a score at or below 70 on an IQ test, and also show difficulty adapting in everyday life. Typically, mildly retarded individuals (about 85%) score between 50 and 70 on IQ tests, are usually able to care for themselves, can care for a home, achieve a sixth-grade education, hold a job, get married, and become an adequate parent. In schools, they are often
mainstreamed
, or integrated into regular education classes. Moderately retarded individuals (about 10%) score between 35 and 49 on IQ tests; may achieve a second-grade education; may be given training in skills such as eating, toileting, hygiene, dressing, and grooming so that they can care for themselves; and may be given basic training in home management, consumer, and community mobility skills so that they can hold menial jobs and live successfully in a group home. Severely retarded individuals (about 3–4%), with IQs between 20 and 34, typically develop a very limited vocabulary and learn limited self-care skills. Usually they are unable to care for themselves adequately and do not develop enduring friendships. Profoundly retarded individuals (1–2%), with IQs below 20, require custodial care. Communities have been housing a greater proportion of mentally retarded people than in the past. These people live with their own families or in group homes when possible. This deinstitutionalization is termed
normalization
.

Figure 15.1 The normal curve
.

Is there one underlying capacity for intelligence or do we have different, distinct ways of being intelligent? A contemporary of Alfred Binet, Charles Spearman tested a large number of people on a number of different types of mental tasks. He used
factor analysis
, a statistical procedure that identifies closely related clusters of factors among groups of items by determining which variables have a high degree of correlation. Because all of the
mental tasks had a high degree of correlation, he concluded that one important factor, which he called
g
, underlies all intelligence. Because the correlation wasn’t a perfect 1.0 between all pairs of factors, he also concluded the existence of the less important
s
, or specialized abilities. Louis Thurstone disagreed with Spearman’s concept of
g
. Based on factor analysis of tests of college students, Thurstone identified seven distinct factors he called
primary mental abilities
, including inductive reasoning, word fluency, perceptual speed, verbal comprehension, spatial visualization, numerical ability, and associative memory. J. P. Guilford divided intelligence into 150 different intelligence sets.