PSYC37H3 Lecture Notes - Lecture 6: Concurrent Validity, Construct Validity, Predictive Validity
Document Summary
Face validity: when measure appears to be measuring what it is supposed to, the items seem relevant and on topic, when test looks valid, not really a type of validity. Criterion-related evidence for validity: whe(cid:374) test is effe(cid:272)ti(cid:448)e i(cid:374) esti(cid:373)ati(cid:374)g a(cid:374) e(cid:454)a(cid:373)i(cid:374)ee"s perfor(cid:373)a(cid:374)(cid:272)e o(cid:374) a defi(cid:374)ed out(cid:272)o(cid:373)e measure (outcome = criterion, criterion-> standard against which test is compared. Sat score tells how you did in university (criteria -> 1st year gpa: predictive validity: when test scores can predict criterion that is assessed at future time point. Sats predicting college performance: concurrent validity: when test score and criterion are assessed at approximately the same time, and test score is a good estimate of criterion. Self-report depression measure and current psychiatric diagnosis of major. Depressive disorder (criterion -> diagnosis) if person have depression based on diagnosis, questionnaire should correlate: validity coefficient. Squaring validity coefficient tells percentage of variation in criterion that is explained by test scores.