The 40-sample stereoscopic grading scale is highly reliable: sample grades correlate .98 with values from other judges, and individual placement correlates .92 with true scale placement¶

Claim · All claims
Evidence · 1 study · design r2 · q2 · i3 large
Currency · newest study 1916 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i3 Grades given to scale samples would correlate .98 with scale values derived from judgments of other equally competent judges, and an individual's placement of samples correlates .92 with true scale placement. → Kelley 1916

Evidence¶

Kelley 1916¶

Kelley, Truman Lee. (1916). A constructive ability test. The Journal of Educational Psychology, Vol. VII, No. 1. https://www.google.com/search?q=Kelley+1916+A+Constructive+Ability+Test+Journal+of+Educational+Psychology

q2 · i3 · design · r2

Psychometric evaluation of the merit scale built by the Thorndike handwriting-scale method from rankings by ten judges of structures built by normal and subnormal boys. The printed reliability correlations are r = .98 (quote) and r = .92, "the correlation between an individual's placement of a large number of samples and the true placement in the scale is .92"; practice with 40-50 samples halves the probable error of judgment.

"The grades given to the samples would correlate with other scale values, derived in a similar manner from the judgments of other equally competent judges to the extent of .98."

Discussion¶

Related Claims¶

• Cronbach's alpha equals the reliability of the scale score under classical test theory, domain sampling, and generalizability theory, and should be interpreted as a reliability coefficient — related
• Combinations of two measure types correlate very highly with the composite (.96, .91, .98), with little difference between choices — related