# Psychology 2800E Lecture Notes - Lecture 4: Inter-Rater Reliability, Concurrent Validity, Face Validity

## Document Summary

Carry-over effect: effect that carries over from one condition to another. Whenever you have a within subjects design, there is a possibility of carry-over effects. Individual differences only apply to between subjects design. Construct validity - how well are they operationally defined? (whether measured of manipulated) R may be thought of as a measure of linearity. Direction (sign) [slope] and strength (value) [spread] The closer the points in a scatterplot approximate a straight-line the closer r will be to +1 (positive correlation) or -1 (negative correlation) If there is no relationship, r will be close to 0, but not vice versa. An r close to 0 does not mean no relationship (it could instead mean that the relationship is not linear) Measure same set of participants at least twice. It r is positive and strong (above . 5) then good test- retest reliability. For observational measure only! o o o o same time. Have to observers rate same participants at the.