Validity

• Internal Validity is the extent that the IV causes the DV
• External Validity is the extent to which the findings of the study can be generalized to a population
• Random sampling is for external validity
• Random assignment is for internal validity
• Face validity is subjective
• Face validity: The degree to which a test appears to measure what it is supposed to measure.
• Content validity: The degree to which a test measures all aspects of the construct being measured.
• Criterion-validity: The degree to which a test has predictive, concurrent, and retrospective validity
• Predictive validity: Predict (forecast) future behavior
• Concurrent validity: Predict current behavior
• Retrospective validity: Predict past behavior
• Construct validity: The extent to which a test has convergent and divergent validity
• Convergent validity is also known as congruent validity
• Convergent validity:

  the test scores correlate with scores from another measure of the same construct
• Divergent validity is also known as discriminant validity
• Divergent validity: The extent to which scores uncorrelate with scores from another measure of different construct
• The Relationship Quality Interview measures relationship in 5 domains. Strong correlations among the 5 parts indicate convergence of same construct (convergent validity)
• Weak correlations between RQI domains and a measure of relationship satisfaction (rather than quality) indicate divergence.
• High correlations are close to 1
• Low correlations (less relationship) are close to 0
• Covariance: Results show that as A changes, B changes either in same or opposite direction
• Temporal precedence: Study’s method ensures A comes first in time before B (A comes before B)
• Internal validity: Study’s method ensures there’s no alternate explanation for change in B
• Reliability is the degree of stability or consistency of measurements (same individuals get same scores every time)
• A valid measure is reliable but a reliable measure is not necessarily valid
• Measure of height is not a valid measure of intelligence but it is reliable because it yields consistent results.
• Accuracy is the degree to which a measure conforms to the established standard
• A speedometer that consistently reads 10 mph faster than actual speed is reliable (yields same results) and valid (measures speed) but not accurate
• Open ended questions: Allow respondents to answer how they choose
• Close ended questions: Limit the possible answers
• Open ended questions are good for gathering rich data but create more code or analysis
• Close ended questions are good when you want limited responses and are easier to analyze but take more work to write good questions and options
• Close ended questions are better when the construct is well defined and have a clear idea about the types of responses people are likely to give
• Not reliable, not valid

  
• Valid, not reliable

  
• Not valid, reliable

  
• Valid, reliable