Estimating premorbid IQ from demographic variables: A comparison of a regression equation vs. clinical judgement

Abstract

To assess the accuracy with which clinicians estimate premorbid IQ from demographic variables and compare it with a regression equation which uses the same information. Repeated measures and correlational. Sixty participants were administered the WAIS-R and had their demographic variables recorded (age, sex, years of education and occupation). Eight clinical psychologists estimated the participants' IQs from the demographic variables. Estimated IQs were also obtained using a regression equation developed by Crawford and Allan (1997). The correlation between obtained IQ and the equation-based estimate was significantly higher than the correlation between obtained IQ and the clinicians' estimates. Further, mean estimated IQ from the regression equation did not differ significantly from obtained IQ whereas the means for four of the eight clinicians' estimates did differ significantly. Demographic-based regression equations can provide unbiased and useful estimates of premorbid IQ; these estimates can be modified in the light of the additional qualitative information available to the clinician.