Test results correlate more highly with estimated intelligence than with grades, including Starch's .75 versus .30 at Wisconsin¶

Claim · All claims
Evidence · 1 study · associational r1 · q2
Currency · newest study 1919 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Across studies, correlations of test results with estimated intelligence exceed those with grades: Starch obtained .75 versus .30, and faculty rankings uncorrelated with test results correlated .55 with the vocabulary test alone. → Caldwell 1919

Evidence¶

Caldwell 1919¶

Caldwell, Helen Hubbert. (1919). Adult Tests of the Stanford Revision Applied to College Students. The Journal of Educational Psychology, X(10). https://doi.org/10.1037/h0074484

q2 · i? · associational · r1

Reported coefficients from Dr. Starch (unpublished) correlating Stanford Adult tests with University of Wisconsin standing: .30 with grades and .75 with estimated intelligence. Table Ib compiles estimated-intelligence correlations: Kitson .57, King .41, Downey .527, Caldwell .47.

"Test results with grades . 30 Test results with estimated intelligence . 75"

Discussion¶

Learner Variables¶

• General Cognitive Ability — predictor: learners who differ on it differ in outcomes

Related Claims¶

• Stanford Adult test results correlate .44 with college grades when Juniors and Sophomores are re-ranked as one group — related
• True-false tests correlate with intelligence ratings (.406) more highly than essays (.344) or written work (.255), and combining true-false tests raises the intelligence correlation while combining essays does not — related
• Group intelligence scores correlated only modestly with school marks (freshmen: 0.41 Thurstone, 0.56 Brown; junior high: 0.25 Illinois, 0.32 Otis), which the author attributes to marks depending on many non-intelligence factors — related