Abstract
Based on test comparisons involving psychotic depressives and chronic paranoid schizophrenics, Carter (J. Nerv. Ment. Dis. 174: 336-341, 1986) inferred that cognitive disorder may reflect a general psychosis factor rather than a differential diagnostic marker. Review of the design and analysis raises methodological and conceptual issues that obviate such interpretation. Problems include absence of control for IQ and chronicity of illness, sample bias incurred by selectively matching groups for education, use of parametric statistics on ordinal data, measures that do not differentially assess cognitive deficit, plausible alternative explanations for nonsignificant outcome, and inadequate data base for generalizing about larger diagnostic groups or psychosis itself. Significant differences found by Carter and others suggest, instead, some fundamental distinctions in the nature if not extent of cognitive disorder in these two groups. It was proposed that a general psychosis factor may be a pervasive interfering variable that obscures diagnostic differences in cognition and thus needs to be rooted out by proper statistical or research design.

This publication has 0 references indexed in Scilit: