Open AccessTHE TWO‐PREDICTOR VALIDITY CURVE
Author(s) -
Dorans Neil J.,
Kulick Edward
Publication year - 1985
Publication title -
ets research report series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.235
H-Index - 5
ISSN - 2330-8516
DOI - 10.1002/j.2330-8516.1985.tb00108.x
Subject(s) - statistics , mathematics , point (geometry) , criterion validity , predictive validity , correlation , psychology , psychometrics , geometry , construct validity
A visual display known as the two‐predictor validity curve is introduced. Derivations are presented for the general case. Then, for illustrative purposes the two predictors are taken to be SAT‐V (V) and SAT‐M (M), and the criterion of interest is the first year college grade point average (FYA). The two‐predictor validity curve, in this case SAT V&M, allows one to assess the relative importance of SAT‐V in relation to SAT‐M for predicting the criterion of interest. When several V&M curves are placed on a single page, comparisons across gender can be made and trends across time can also be studied. Several illustrations using SAT data are presented.