The Effects of Differing Instructions and Guessing Formulas on Reliability and Validity

THE use of correction for guessing formulas has led to conflicting views. The rationale of those who support the use of correction formulas has been summarized by Davis (1964). In general, proponents believe that corrected scores agree more closely with true scores than do simple raw measurements. Theoretical support for the use of correction formulas was given by Lord (1957, 1963) and Cureton (1966) who demonstrated mathematically that guessing produces an increase in variability, thus causing