Select test items by how distinctly they separate a known special-class group from a known regular-class group¶

Strategy · All strategies
Evidence · 1 claim (1 mixed) · 1 study (1 theoretical), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description¶

This is a concrete procedure for validating a screening test: after drafting candidate tests, administer them to two criterion groups and keep the items that discriminate. Pressey writes: "The obvious procedure is to give the tests to (1) a group of children who are known, from long and intensive study of them, to belong in special classes, and (2) another group known to belong in the regular classes. The tests which separate these two groups most distinctly are, then, the tests to include in the final scale."

Design Implications¶

Context¶

Requirements¶

• Access to two criterion groups whose placement is already known from long and intensive study

Constraints¶

• The strategy is stated for school-placement prognosis problems; its usefulness for other purposes must be determined by actual trial with the concrete problem

Target Learners¶

• school-age children at placement decision points

Target Learning Goals¶

• accurate classification of children for special or advanced placement

Claims¶

• Statistical item selection that maximizes norm-referenced discrimination undermines criterion-referenced test definitions [~W]

Related Strategies¶

Examples¶

-

Key Sources¶

• Pressey, S. L. (1920). Intelligence and Its Measurement. The Journal of Educational Psychology. https://eric.ed.gov/?q=Intelligence+and+Its+Measurement+Pressey