Validity Flashcards
What is Validity
Degree to which a measure measures what it is supposed to measure.
Types of Validity
Construct validity Convergent validity Divergent validity Content validity Face validity Criterion-related validity Concurrent validity Predictive validity Ecological validity Internal Validity External Validity
What is Construct Validity
extent to which the operationalization of a construct measure a construct as defined by a theory. This subsumes all other types of validity.
What is Convergent Validity
degree to which the operationalization is similar to the operationalizations that it theoretically should be similar to
What is Divergent Validity
degree to which the operationalization is not similar to other operationalizations that it theoretically should not be similar to
What is Content Validity
extent to which the operationalization covers a representative sample of the construct content
What is Face Validity
whether the measure appears to be a good measure or not
What is Criterion- related Validity
degree to which operationalization correlates with a criterion (variable) that has already been validated.
What is Concurrent Validity
degree to which the operationalizations corresponds to previously established measurements for the same construct. Both tests are administered at the same time.
What is Predictive Validity
degree to which operationalization predicts something that it should theoretically predict. Usually has a time lag between predictor and criterion, leading to stronger
support for causation
What is Ecological validity
construct’s relationship with other constructs generalize to samples outside population I’ve drawn from
What is Internal Validity?
is the experiment interpretable
What is External Validity?
generalizability
Threats to Internal Validity
- History/events between first and second measurements
- Maturation process within participants over time
- Testing effects of second testing
- Instrumentation changes in calibration of observers/scorers
- Statistical regression (towards the mean) if groups selected based on extreme scores
- Selection biases used for forming groups
- Experimental mortality or loss of participants
- Interaction between any of these
Threats to External Validity
- Reactive/interaction effect of testing; pretested population vs. not pretested population
- Interaction between selection biases and experimental variable
- Reactive effects of experimental arrangements/setting
- Multiple treatment interference (effects of prior treatment not erasable)
