Validity Definition
accuracy of measurement
how well a test measures what it purports to
equation for an observed score
true score plus error
what is intelligence
multifaceted capacity that includes: acquire and apply knowledge, reason logically,plan effectively, grasp visual concepts, using correct words, cope with novel situations
reliability
consistency in measurementent
Parallel and alternate forms tests
parallel- must have equal variance and means
alternate- must be the same, not as strict as parallel
3 ways validity is often conceptualized
Content
Criterion-related
Construct
less error variance means what for reliability
greater reliability and vice versa
1st person to link heritability to intelligence
sir francis galton
Reliability Coefficient
the proportion that indicates the ratio b/w true score variance and total variance
Item response theory
way to model that a person with X ability will be able to perform at a level of Y
validity coefficient
correlation coefficient that provides a measure of relationship between test scores and scores on criterion measure
what is error in reliability
component of the observed score that does not have to do with the test takers true ability or trait being measured
what is interactionism
how much your environment impacts your intelligence and growth
Internal consistency reliability
the degree of relatedness of items on a scale or test
what is test development
everything that goes into making a test
conceptualization
construction
tryout
analysis
revision
Incremental validity
the degree to which an additional predictor explains something about the criterion measure that is not explained by predictors in use
what is measurement error
all factors associated with the process of measuring some variable other than the variable being measured
2 factor theory of analysis
horn and cattell
crystallized: skills and knowledge dependent on formal and informal education
fluid: nonverbal, culture-free
Content validity
evaluation of subjects topics or contents of test
norm-referenced
going off of what others have done
good items tells that a correct answer means better
criterion-related validity
standard against which a test score is evaluated
uncontaminated
Explain the 2 separations of measurement error in reliability
Random Error- the source of error in measuring targeted variable caused by unpredictable variables
Systematic Error- a source of error in measuring a variable that is constant
core and supplemental subtests
Supplemental: provide additional clinical information
Criterion-related validity
obtained by evaluating the relationship of scores obtained on the test to scores on other tests or measures
method of paired comparisons
test-takers must choose between two alternatives to some rule.