Principles of Test

Construction
Principles of Test Construction

For a psychological test to be acceptable it must

fulfill the following three criteria:

1. Standardization
2. Reliability
3. Validity
 Standardization
Standardization:
Standardizing a test involves administering the test
to a representative sample of future test takers in
order to establish a basis for meaningful
comparison.
• AP Exams, Intelligence Tests, and SAT (your
scores are ranked in average of many test takers)

Norms:
Average score for a designated group of people
Normal Curve

Standardized tests establish a normal distribution

of scores on a tested population in a bell-shaped
pattern called the normal curve.
 Reliability
A test is reliable when it yields consistent results. To
establish reliability researchers establish different
procedures:
• Test scores should be same or similar each time the same person
takes it.
• 1st attempt SAT – 1800 , then 2nd attempt 1000 is this reliable?

1. Test-Retest Reliability: Using the same test on two

occasions to measure consistency.
2. Split-half Reliability: Dividing the test into two
equal halves and assessing how consistent the
scores are.
3. Reliability using different tests: Using different
forms of the test to measure consistency between
them.
Example of Test Re-test Method
• when determining the effect of coloring on stress, a
researcher may choose to evaluate the subjects'
blood pressure before, during, and after coloring a set
number of pages. This gives researchers a base line
from which to draw later conclusions.
 Validity

Reliability of a test does not ensure validity.

Validity of a test refers to the degree in which the
test is supposed to measure or predict.

1. Content Validity: Refers to the extent which a test measures

your definition of the construct or behavior of interest?

• Does a physical test measure your knowledge of psychology?

• Does the AP Psychology exam measure your knowledge of
psychology?
• Does a physical test measure how athletic you are?
Validity
2. Criterion-related validity: Relationship between scores on a
test and actual performance
• Predictive Validity: Refers to the function of a test in predicting
a particular behavior or trait.
• SAT Scores to College Grade

• Convergent Validity: The degree to which scores on a test

correlate with (or are related to) scores on other tests that are
designed to assess the same construct.
• Is your SAT scores similar to ACT scores? Is there a relationship
(correlation)?
Validity
3. Construct Validity refers to the ability of a
measurement tool (e.g., a survey, test, etc) to actually
measure the psychological concept being studied.

For example, if we want to know our height we would

use a tape measure and not a bathroom scale because
all height measurements are expressed in inches and
not in pounds.
Intelligence Test
Assessing Intelligence
Psychologists define intelligence testing as a
method for assessing an individual’s mental
aptitudes and comparing them with others using
numerical scores.
 Alfred Binet
Alfred Binet and his
colleague Théodore
Simon practiced a more
modern form of
intelligence testing by
developing questions
that would predict
children’s future
progress in the Paris
school system.
 Alfred Binet
Alfred Binet’s goal
became measuring
children’s mental age, the
level of performance
typically associated with
a certain age.

Identify children who

will need extra need
(Special Ed.)
 Lewis Terman
In the US, Lewis Terman
adapted Binet’s test for
American school
children and named the
test the Stanford-Binet
Test. The following is the
formula of Intelligence
Quotient (IQ),
introduced by William
Stern:
Stanford-Binet scale
• Stanford-Binet scale
• Terman’s adaptation of the Binet-Simon
scale
• introduced the “I.Q.” score
• A score of 100 is considered average
Sample Questions
If a four-year-old girl correctly answered questions on an intelligence
exam similar to a five-year-old girl, she would be said to have a mental
age of five. In this case her intelligence quotient (IQ) would be:
Sample Question
• If a five-year-old girl correctly answered questions on an intelligence
exam similar to a 6-year-old girl, she would be said to have a mental
age of five. In this case her intelligence quotient (IQ) would be:
 Aptitude and Achievement Tests

Aptitude tests are intended to predict your ability to

learn a new skill
achievement tests are intended to reflect what you
have already learned.
Examples
Where would these exams fall under?
1.Pop Quizzes
2.Unit Exams
3.SATs (Enter College)
4.ACTs (Enter College)
5.ASVAB (Enter Military)
6.AP Exams
Examples
• Achievement Tests
• Unit Exams and AP Exams (Assess what you have learned)
• Aptitude Tests
• SATs and ACTs (predicts how well you will do in your first year in college)
• ASVAB (What kind of job would be appropriate for you in the military)
 David Wechsler
Wechsler developed the
Wechsler Adult
Intelligence Scale (WAIS)
and later the Wechsler
Intelligence Scale for
Children (WISC), an
intelligence test for
preschoolers.
WAIS
WAIS measures overall intelligence and 11 other
aspects related to intelligence that are designed to
assess clinical and educational problems.
Intelligence Tests
• The Wechsler Intelligence Scales
• The Wechsler Adult Intelligence Scale – Third Edition is the most commonly
used test of intelligence for adults
• WAIS-III is divided into to parts, one that focuses on verbal abilities and one
that focuses on performance skills
• Also a version for children, Wechsler Intelligence Scale for Children – Third
Edition
Intelligence Tests
• Individual Tests
• Intelligence tests that are taken by one individual
• IQ Tests and Personality Tests
• Group Tests
• Intelligence tests that can be given to large groups
• Advantages
• Quick scoring
• No examiner bias
• Easier to establish norms
• Disadvantages
• Less likely to detect someone who is ill or confused
• Might make people nervous
• Learning disabled children often perform worse
Intelligence Tests
• Speed Tests
• Large number of questions are given in a short period time
• How quickly can you solve the problems?

• Power Tests
• Consists of items that get progressively more difficult
• To what degree of difficulty can you solve the problems?
Intelligence Tests
• Performance tests
• Tests that minimize the use of language
• Used to test very young children or people with retardation
• Also can be used to test those unfamiliar with English
• Culture-fair tests
• Tests designed to reduce cultural bias
• Minimize skills and values that vary from one culture to another
Flynn Effect

In the past 60 years, intelligence scores have risen

steadily by an average of 27 points. This
phenomenon is known as the Flynn effect.
Review
 OA 3
1. What % of people will receive 0-100 on the IQ test?
2. What % of people will receive 85-100 on the IQ test?
3. What % of people will fall under standard 1 deviation On
the IQ test?
4. What % of people will receive 70-100 on the IQ test?
5. What % of people will fall under standard 2 deviation on
the IQ test?
6. What % of people will receive 130- up on the IQ test?
Nature AND Nurture

Both Genetics & Environment play a

role in our intelligence
Genetic Influences
Studies of twins, family members, and adopted
children together support the idea that there is a
significant genetic contribution to intelligence.
 Environmental Effects
Differences in intelligence among these groups are
largely environmental, as if one environment is
more fertile in developing these abilities than
another.
 Reaction Range
• Intelligence may increase or decrease as a result of
environment
• Extent to which traits, abilities, or IQ scores may increase
or decrease as a result of environmental factors
• Have you received tutoring as a child?
• Have your attended Preschool?
• Do you think these factors would have had an impact on your
abilities or IQ scores?
 Early Intervention Effects
Early neglect from caregivers leads children to
develop a lack of personal control over the
environment, and it impoverishes their intelligence.

Romanian orphans with minimal

human interaction are delayed in their development.
 Schooling Effects
Schooling is an experience that pays dividends,
which is reflected in intelligence scores. Increased
schooling correlates with higher intelligence scores.

To increase readiness for schoolwork,

projects like Head Start facilitate leaning.
 Stereotype Threat
• Situational predicament in which people are or feel
themselves to be at risk of confirming negative
stereotypes about their social group.
• Black students who was proctored by a white
teacher scored poorly than black students who was
proctored by a black teacher.
• Real or imagined sense of discrimination