Intelligence and Testing
Chapter 8
What is intelligence?
• intelligence: A general term referring
to the ability to learn and develop
adaptive behaviors.
• intelligence tests: Tests designed to
measure a person’s general mental
abilities.
Theories of Intelligence
• Charles Spearman’s genreal intelligence “g”
• L. L. Thurstone’s Seven Mental Abilities
• Raymond B. Cattell's Two-Factor Theory
• Robert Sternberg’s Triarchic Theory
• Howard Gardner’s Theory of Multiple
Intelligences
Charles Spearman’s Theory
• “g” factor (general intelligence):
Spearman stressed a general mental
energy that can be channeled in various
directions. Intelligent people can
perform well in many areas.
• s factors (specific intelligence):
reflect specific knowledge and abilities
that are only used when performing
specific tasks that have been leaarned.
Thurstone’s Seven Primary Mental Abilities
• In contrast, L. L. Thurstone said primary
mental abilities are “independent” of each
other. A person could excel in one area and
be very average in others.
• Thurstone’s seven primary mental abilities:
spatial visualizations
verbal meaning
word fluency
number facility
memory
reasoning
perceptual speed
Sternberg’s Triarchic Theory
Experiential
Componential
Contextual
We all share these three types of intelligence
but excel in ore or two areas
Sternberg’s Triarchic Theory
• componential: Ability to acquire new
knowledge and solve problems effectively
(intelligence as we think of it).
• experiential: The ability to adapt creatively
in new situations (ability to meet new
challenges).
• contextual: Ability to select “contexts” in
which you can excel (e.g., being able to
make the right career choice).
Gardner’s Theory of Multiple Intelligences
Much like Thurstone, Gardner suggests, seven
“independent” areas (types) of intelligence.
• logical-mathematical - Newton, Einstein
• linguistic - Whitman, Shakespeare
• spatial - Picasso, DaVinci, F. L. Wright
• musical - Mozart, Gershwin, Eric Clapton
• body-kinesthetic - M. Jackson, M. Jordan
• interpersonal - Ghandi,Carter, Nixon
• intrapersonal - to know oneself
Gardner’s Theory
• logical-mathematical: Ability to handle chains
of reasoning, numerical relations, and
hierarchical relations.
• linguistic: Sensitivity to the meaning and order
of words, as well as the functions of language.
• spatial: Ability to perceive the world accurately
and to transform and recreate perceptions.
• musical: Sensitivity to pitch, tone, timbre, and
musical patterns.
Gardner’s Theory
• body-kinesthetic: Ability to use one’s body or
to work with objects in highly differentiated
and skillful ways.
• interpersonal: Ability to notice and make
distinctions among the moods, temperaments,
motivations, and intentions of others.
• intrapersonal: Ability to understand one’s own
feelings and use them to guide behavior.
Daniel Goleman: Emotional Intelligence
• Relatively new area of study
• understanding and managing one’s own
emotions is probably MORE vital to success
than pure intellectual ability
• Many highly successful people are not
brilliant and many geniuses die in poverty
• Understanding the emotions of others is also
important. It allows us to predict how they
will act in various situations
Raymond Cattell: Two Factor Theory
(not in book)
• fluid intelligence: General mental energy (like
“g”). like a fluid, it “shapes” itself to the task
(e.g., math, music, chemistry).
• crystallized intelligence: Knowledge that is
“learned” or accumulated over a lifetime.
• Crystallized intelligence is more stable across
the lifespan whereas fluid intelligence begins a
slow decline in middle adulthood.
Cattell’s Other Contributions
• Personality: A leader in personality theory and
testing as well
• Cattell developed the “16 pf (personality
factor) Inventory”
• He applied the technique of “factor analyses”
(deveoped by Spearman) to the study of
intelligence and personality
• Factor Analysis: statistical (mathematical)
method for identifying the basic “factors” of
intelligence and personality
Intelligence Tests
• The Binet-Simon Scale
• The Stanford Binet
• The Weschler Scales
– WAIS (Weschler Adult Intelligence Scale)
– WISC (Weschler Intelligence Scale for Children)
– WPPSI (Weschler Preschool/Primary Scales of
Intelligence)
• culture fair tests
• performance tests
The Binet-Simon Scale
• Binet-Simon Scale: The “FIRST” intelligence test.
Alfred Binet was commissioned by the French
Government to identify children with special
learning needs
• Binet developed the concept of “Mental Age”
• mental age: a child who scored the same as the
“average” child of a given age on a standardized
test had that “mental age”
• example: if a 10 year old scored the same as the
average 12 year old, he had a mental age of 12
The Stanford-Binet
• Louis M. Terman: working at Stanford
University, developed an “English” version of
the Binet-Simon
• Longitudinal Study: Terman also began a
long-term study of “gifted” children
• Intelligence Quotient (IQ): credited to Terman.
Expesses the relationship between “mental
age” and “chronological age” as a single
number (a “quotient”)
• IQ = mental age/chronological age * 100
Examples of IQ Computation
• a ten year old: scores at the level of the
average 12 year old on a given test
– 12/10 * 100 = 120 (a bright 10 year
old)
• a ten year old: scores at the level of the
average 8 year old on a given test
– 8/10 * 100 = 80 (a below average 10
year old)
The Stanford-Binet yields an
overall IQ plus four “area” IQs
• 1. verbal reasoning
• 2. abstract/visual reasoning
• 3. quantitative reasoning
• 4. short-term memory
The Stanford-Binet overall IQ and area IQs have
a mean of 100 and standard deviation of 15
David Wechsler’s Contribution
• Wechsler noted that the Stanford-Binet was
designed primarily for “children”
• Weshsler developed a test for adults (the
WAIS) Wechsler Adult Intelligence Scale
• The WAIS yields an overall IQ plus two “area”
IQs: “verbal” and “performance”
• Like the Stanford-Binet, the WAIS overall IQ
and area IQs have a mean of 100 and standard
deviation of 15
Wechsler Extends His Work
• Wechsler developed a test for (1) older
children and adolescents (WISC) Wechsler
Intelligence Scale for Children and (2) young
children (WPPSI) Wechsler
Preschool/Primary Scales of Intelligence
• Because each test covers a narrow age range
and is easier to use, the Wechsler scales have
surpassed the Stanford-Binet as the most
widely used IQ tests
Test Administration
• The Stanford-Binet, Wechsler Scales,
and other IQ test are administered
individually (one examiner to one test
taker). One takes about 90 minutes.
• The testing process can be lengthy and
tiring for examiner and test taker
• There are group (pencil and paper) IQ
tests which are less accurate
What does an IQ score mean?
(See Appendix A in book)
• To answer this we need to know a little
bit about “STATISTICS”
• Statistics: branch of mathematics used to
organize and analyze data
• 3 important statistical concepts
– measures of central tendency
– measures of variability
– the “normal” or “bell” curve
Three Measures of Central Tendency
• 1. Mode: the most frequently occurring
score (2,3,5,2,4,6) the mode is “2”
• 2. Median: score that divides the
distribution of scores in half (2,4,5,8,9)
the median is “5”
• 3. Mean: the arithmetic average
(2+3+5+6) / 4 = “4”
Two Measures of Variability
• 1. Range: difference between the highest
and lowest scores in a distribution
(2,5,3,9,7) the range is 9 -2 = “7”
• 2. Standard Deviation: a more useful
measure “the average distance that a
group of scores are from their mean”
• 7 scores (85, 90, 95, 100, 105, 110, 115)
What is the average distance that these
scores are from their mean (100)?
Computing the standard deviation
85 - 100 = -15 (-15)2 = 225
90 - 100 = -10 (-10)2 = 100
95 - 100 = - 5
(-5)2 = 25
100 -100 = 0
(0)2 = 0
105 -100 = 5
(5)2 = 25
110 -100 = 10 (10)2 = 100
115 -100 = 15 (15)2 = 225
 = 700
700 / 7 = 100
100 = 10
The standard deviation is “10”
The “Normal” Distribution (“Bell Curve”)
• Is a “hypothetical” bell shaped curve that
approximates the distribution of scores
(characteristics) found among most “naturally
occurring” variables
• “some naturally occurring variables are height,
weight, personality traits, and “IQ”!
• The normal distribution is “marked off” in
“standard deviations” with 0 at the center
• IN the next slide, the percentages for each part of
the curve are shown. Note, they add up to 100%
The “Normal” Distribution (“Bell Curve”)
marked off in “standard deviations” (SD)
• Intelligence is a “naturally occurring variable.” IQ is,
therefore, “normally distributed.” recall, Mean = 100, SD = 15
•So, for all people in the U.S. (or any other) population:
•68 % have IQs between + and - 1 SD (85 - 115)
•95 % have IQs between + and - 2 SD (70 - 130)
2.5%
13.5%
34%
34%
-2
-1
0
70
85
100
13.5%
2.5%
+1
+2
< SD
115
130
< IQ
a more complete distribution of IQ
scores going out to about + and - 4
standard deviations
Performance and Culture-Fair Tests
• performance tests: Intelligence tests that
minimize the use of language.
• culture-fair tests: Intelligence tests that
eliminate “cultural bias” by minimizing skills
and values that vary from one culture to another
(e.g., what to do if another child hits you?)
Children of different “Social Classes would
answer differently.
• Progressive Matrices:an example of a test that
is both a performance test and a culture-fair test
What Makes a Good Test: Psychometrics
• 1. reliability: Ability of a test to produce
consistent and stable scores.
• 2. validity: Extent to which the test
measures what we think it measures.
• correlation: Statistical measures of the
strength and direction of the relationship
between two variables. Correlation ranges
from -1 to + 1
• Both reliability and validity are established
via correlations
Reliability
• Five people take the same test twice on
Monday and again on Friday
person Monday Score
Friday Score
A
100
80
B
140
90
C
75
120
D
80
130
E
150
90
• Correlation is “low” Reliability is “poor”
Reliability
• Five people take the same test twice on
Monday and again on Friday
person Monday Score
Friday Score
A
100
102
B
140
138
C
75
72
D
80
79
E
150
155
• Correlation is “high” Reliability is “good”
Reliability
• carry over effects: can occur when the
first administration of a test influences
the score on a second testing.
• split-half reliability: A method of
determining test reliability by dividing a
test into two halves and correlating the
two halves. Eliminates carry over
effects.
Establishing Validity of a New Test
• Five people complete our “new IQ test” and
an “established test” (e.g., Stanford-Binet)
person
New Test
Established Test
A
70
102
B
140
90
C
75
135
D
80
122
E
150
75
• Correlation is “low” Validity is “poor”
Establishing Validity of a New Test
• Five people complete our “new IQ test” and
an “established test” (e.g., Stanford-Binet)
person
New Test
Established Test
A
100
102
B
140
138
C
75
72
D
80
79
E
150
155
• Correlation is “high” Validity is “good”
Types of Validity
• content validity: extent to which a test
contains questions that cover “all”
aspects of the subject being tested
• criterion validity: the new test is
compared to a “criterion” or “standard”
to establish its validity
• face validity: what the questions are
measuring is very obvious (e.g., Are you
depressed?). Sometimes a problem in
personality, attitude, or clinical testing
Positive Aspects of Individual Testing
• problems (lack of motivation) can be detected
• tests can be tailored for those with special needs
• tests are quite accurate and reliable
Criticisms of Individual Testing
• possibility of “Halo Effect” (a positive or negative
bias on the part of the examiner)
• tests are time consuming and expensive
• tests can give an unfair advantage to those of
higher SES in school placement situations
• tests may contain cultural-ethnic biases
What Determines Intelligence?
(Nature vs. Nurture)
Heredity
Environment
• Tryon’s maze
• Tryon’s and
bright and maze
dull rats
• IQ correlations in
humans
Rosenzweig’s rat
studies
• H. M. Skeels
study of mentally
retarded orphans
Evidence for Heredity
• Robert Tryon taught rats to run a maze
• He bred the fastest learners with other
fast learners over several generations
• He did the same with the slowest learners
• After several generations, he had two
distinct populations: “maze bright rats”
who learned quickly and “maze dull rats”
who learned slowly
Evidence for Heredity
IQ and Familial Relationships
Evidence for Environment
• Research by both Robert Tryon and Mark
Rosenszweig showed that rats raised in a
“stimulating” environment had more well
developed brains and were brighter than
rats raised in a plain and boring
environment
Evidence for Environment
• H. M. Skeels studied two groups of
below average IQ orphans
• One group was placed in a setting where
they had “attention” from adult patients
(also below average IQ)
• The other group remained in the
orphanage, receiving little attention
• IQ scores INCREASED for the group
placed with the adults
The Flynn Effect
• Professor James Flynn: has noted that IQ
scores worldwide have been increasing
several points per decade since the 1930s
• However, nobody, as yet, is sure exactly
why.
• Are we getting more intelligent, and if so
why?
• Maybe just getting better at taking tests.
Gender Differences in Intelligence
• Recent research finds little difference
in the math and verbal abilities of
males and females.
• Males do have better spatial ability
than females.
• The average male and female IQ is
about the same, but there is a higher
proportion of men at the two extremes.
Extremes of Intelligence
• mental retardation: The condition of
significantly subaverage intelligence (IQ
below 70) combined with deficiencies in
adaptive behavior.
• giftedness: Refers to superior IQ
combined with demonstrated or potential
ability in such areas as academic
aptitude, creativity, and leadership.
Levels of Mental Retardation
Mild - 50 through 70
may be able to function independently
Moderate > 35 through 50
will need some level of care and supervision
Severe > 20 through 35
will need extensive care and supervision
Profound > 0 through 20
incapable of the even the simplest tasks
Some Genetic Causes of Retardation
• Down syndrome: results from an “extra”
defective 21st chromosome (also called
“Trisomy 21”).
• phenylketonuria (PKU): An enzyme
needed for metabolizing certain foods is
absent. Toxins collect and damage the
nervous system
• fragile-X syndrome: due to a defect on
the X chromosome
Other Causes of Retardation
• alcohol or other drugs during pregnancy
• exposure to toxins or radiation during
•
•
•
•
pregnancy
maternal illness during pregnancy
birth trauma
malnutrition/deprivation during childhood
numerous other causes
Creativity
• The ability to produce novel and
socially valued ideas or objects.
• Creativity differs from giftedness
• One can be very creative without
having superior intelligence
Descargar

No Slide Title