" Psychology and Life " [US]
Richard Grieg, Federer Zimbardo,
Keywords: intelligence test difference reliability Standardized
Psychological measurement -
refers to the degree of testing of individual differences
Used to detect abilities, behaviors and human traits
Targets the differences between
individuals and others in a specific dimension Differences and similarities between people
In Western psychology
the development of measurement is a relatively new field
It was only widely used in the early 20th century
Galton was the first to propose
four ideas on intelligence measurement
① The difference in intelligence is quantified according to the degree of intelligence
That is, the level of intelligence can be quantified
② Individual differences in intelligence are distributed in a bell-shaped curve
or in a normal distribution
The intelligence value of most people lies in the middle In the middle
a small number of people are considered geniuses and intellectually deficient
③Intelligence or mental ability
can be measured by objective tests
The correct answer to each question in the test is unique
④The correlation between the two sets of test scores The degree of correlation
can be determined by statistical analysis
Galton's ideas have long-lasting value
He also proposed controversial ideas
For example, genius is hereditary
Geniuses will appear in the same family
Upbringing has little impact on intelligence
Intelligence is related to people's moral values
He initiated the eugenics movement
advocated the application of evolutionary theory
to encourage heterosexual mating among superior people
In order to achieve the goal of improving human species
Formal measurement procedures should meet three requirements
Reliability, validity and standardization
If the requirements are not met in these three aspects
It is difficult to confirm the credibility of the measurement results
Formal measurement applies to all psychological measurements
Reliability -
Mention the degree of credibility of the consistency score given by a certain measurement tool
Whether the measurement tool is credible
htm Whether l0 is consistent with the underlying concept being measuredThere are three methods to detect the reliability of
① Test-retest reliability
② Parallel test
③ Internal consistency test
Test-retest reliability -
is a method of testing whether a test is reliable
That is, using the same test on the same person
The degree of correlation between the test results in two situations
Parallel test -
uses a modified parallel test
instead of using the same test twice
This method can reduce the practice effect of the test
Internal consistency test -
is conducted on a single test
For example, comparing the scores of odd-numbered items and even-numbered items
In a test with good reliability
the scores of the two parts are equivalent
It is called good internal consistency
has split-half reliability
validity -
refers to the degree to which the test can measure what is to be measured.
Validity reflects the related behaviors and results
. The ability to accurately predict
has three important types of validity.
① Content validity
② Criterion validity
③ Construct validity
Predictive validity -
is also called criterion validity
Compare someone's test scores
with their scores on other related standards
Norms and standardization
Use norms to explain different test scores
Understand the approximate range of scores by looking at the norms of the test
Norms are helpful in evaluating a person
h The relative position of tml0 in the normal populationwill be very useful for individual performance
Standardization -
means testing everyone in the same way under the same conditions
