Midterm Review Flashcards
(39 cards)
Measurement validity includes:
Face Validity
Content Validity
Criterion Validity
convergent validity
Discriminant validty
Face validity
Does it appear on the surface that you are measuring what you are supposed to?
Content validity
Are you measuring every aspect of the construct?
Criterion validity
Does the measure correlate to the behavioral outcome?
Convergent Validity
is the measure of a construct similar to other measures of the same construct?
Discriminant validity
Is your measure different from other measures that are not measuring your construct? They should be
Coding Scheme
Classifying behavior into pre-defined categories
Observer Bias
Skew ratings towards own hypothesis
Observer expectancy effects
When the observer impacts participant behavior
Participant reactivity
Participant is aware that they are being observed and modifies behavior accordingly
Fence sitting
When a participant is neutral for all the questions in a survey
Acquiescence
When a participant agrees to all the statements or says yes to all the questions
random Probability sampling includes:
Simple random Sampling
Stratified random sampling
Cluster sampling
Simple random sampling
Pulling out of a hat -
Systematic sampling -> random number generator
Stratified random Sampling
some of everything
Oversampling - compensation for comparison
Cluster sampling
Randomly sampling whole groups
Multistage sampling -> two random processes
Nonprobability sampling ( biased sampling)
convenience sampling
Quota Sampling
Purposive sampling
Snowball Sampling
Convenience sampling
Right in your backyard
Quota Sampling
Filing requirements
Can think of this as stratified non random sampling
Purposive Sampling
Deliberate recruitment -> need a specialist to pick out appropriate participants
Snowball sampling
2 tell two friends and it snowballs from them
Three criteria for establishing causality
Covariation
Temporal Precendence
Internal Validity
Covariation
variables are related to each other in a systematic way.
necessary for testing association and causal claims
can be captured in correlations and in regression analysis
Temporal Precedence
One variable must undergo change before the othr.
Presumed cause should occur earlier in time than the effect
Longitudinal studies establish this