Statistical control of halo: A response

Responds to criticisms made by K. R. Murphy (PA, Vol 68:100), C. L. Hulin (PA, Vol 68:95) and R. J. Harvey (PA, Vol 68:94) of an article by the present authors in which Landy et al suggested a method for statistically controlling unwanted variance in performance ratings. The method was based on partial correlation techniques through which the variance in overall ratings was subtracted from variance in dimensional ratings. The technique was criticized on the basis of logic, methods, and conclusions. The present authors consider the methodological criticisms and suggest methods by which more accurate estimates of expected values of dimensional intercorrelations might be obtained. (11 ref)