PSYCH 209 Lecture Notes - Lecture 8: Criterion Validity, Convergent Validity, Discriminant Validity
4/13/18 Lecture Notes: Psychometrics: Identifying Good
Measurement
Activity: Reliability
• Researchers place unobtrusive video recording devices in the living rooms of 20 children. Later,
coders view tapes of the living areas and code how many minutes each child spends playing video
games.
o Interrater
o Scatterplot
• Person 1 vs. Person 2 (axes)
▪ Strong correlation (close to imaginary 45 degree line)
▪ Weak correlation (unorganized scattering of points)
• Clinical psychologists have developed a 7-item self-report measure to quickly identify people who
are at risk for panic disorder.
o Internal
• Look at alpha value
o Test-retest
• Results should be stable over time
• Scatterplot
▪ Time 1 vs. Time 2 (axes)
• A restaurant owner uses a response card with four items in order to evaluate how satisfied
customers are with the food, service, ambience, and overall experience.
o Internal
Relationship Between Reliability and Validity
• Can a measure be reliable but not valid?
o Yes
o Example:
• Shoe size as an intelligence test (reliable, not valid)
• Number of children you have as a measure of interest in children (reliably measured,
but correlated with interest?)
Construct Validity
• Is this operationalization appropriate for measuring the construct?
Validity of Measurement
• Validity of measurement = construct validity
• How appropriately the construct is measured; how well the measure represents the construct
• Types of construct validity:
o Face validity and content validity (subjective)
o Criterion validity (empirical)
o Convergent validity and discriminant validity (empirical)
Subjective judgments of validity
• Face validity
o Does it look like it measures what it's supposed to, overall?
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Activity: reliability: researchers place unobtrusive video recording devices in the living rooms of 20 children. Later, coders view tapes of the living areas and code how many minutes each child spends playing video games. Strong correlation (close to imaginary 45 degree line: weak correlation (unorganized scattering of points, clinical psychologists have developed a 7-item self-report measure to quickly identify people who are at risk for panic disorder. Look at alpha value: test-retest, results should be stable over time. Scatterplot: time 1 vs. time 2 (axes, a restaurant owner uses a response card with four items in order to evaluate how satisfied customers are with the food, service, ambience, and overall experience. Relationship between reliability and validity: can a measure be reliable but not valid, yes, example: Shoe size as an intelligence test (reliable, not valid: number of children you have as a measure of interest in children (reliably measured, but correlated with interest?)