Despite numerous technical treatments in many venues, analysis of covariance (ANCOVA) remains a widely misused approach to dealing with substantive group differences on potential covariates, particularly in psychopathology research. Published articles reach unfounded conclusions, and some statistics texts neglect the issue. The problem with ANCOVA in such cases is reviewed. In many cases, there is no means of achieving the superficially appealing goal of "correcting" or "controlling for" real group differences on a potential covariate. In hopes of curtailing misuse of ANCOVA and promoting appropriate use, a nontechnical discussion is provided, emphasizing a substantive confound rarely articulated in textbooks and other general presentations, to complement the mathematical critiques already available. Some alternatives are discussed for contexts in which ANCOVA is inappropriate or questionable.

True-false scales were devised to measure anhedonia, the lowered ability to experience pleasure. The scales were Physical Anhedonia (40 items) and Social Anhedonia (48 items). After scale development using 371 college students, the final version was given to 505 normal adults stratified by social class, age (18-45), and sex, and to 123 male schizophrenics. The potential artifacts of social desirability, acquiescence, and random responding were ruled out. Coefficient alpha values for Physical Anhedonia and for Social Anhedonia were .74 and .85 for male normal subjects and .82 and .85 for male schizophrenics. The schizophrenics scored more anhedonic than the normal subjects on both Physical and Social Anhedonia. The schizophrenics' scores on Physical Anhedonia appeared to fall into two clusters of scores, one cluster resembling the total distribution of the normal subjects, and a second cluster consisting of scores that were more anhedonic than those of the normal subjects. Anhedonics were more often poor premorbid and hedonics more often good premorbid. The Physical Anhedonia Scale may be useful for testing the hypotheses, advanced by several theorists, that anhedonia is genetically transmitted and that nonpsychotic anhedonics are at high risk for schizophrenia.Anhedonia, the lowered ability to experience pleasure, has been described as a schizophrenic symptom by many psychopathologists, including Kraepelin (1913/1919) and Bleuler (1911/1950, although both Kraepelin and Bleuler viewed the loss of the experience of pleasure as only one facet of the deteriora-

for their assistance in locating subjects and tabulating data.

A 28-item true-false scale was constructed to measure schizophrenic body-image aberration. The scale was standardized on both college students and a nonstudent normal control group. Male schizophrenics reported more body-image aberration than nonstudent normal control subjects, but only a portion of the schizophrenics were deviant. Nonpsychotic clinic clients did not have heightened scores. Correlational findings indicated that schizophrenic body-image aberration is an aspect of a broader perceptual aberration. Scores on body-image aberration were negatively correlated with time since first hospitalization. The Body-Image Aberration Scale had essentially no correlation with the Physical Anhedonia Scale of Chapman, Chapman, and Raulin for schizophrenics. For nonschizophrenics, however, high scores on the two scales accompanied one another significantly less often than expected by chance. The authors suggest that the two scales may identify alternative manifestations of proneness toward the same schizophrenia.

Practicing psychodiagnosticians (JV = 32), when surveyed, failed to report observing Wheeler-Rorschach Signs 7 and 8 as accompanying male homosexuality although research evidence indicates that these are valid signs. They instead reported observing Wheeler Signs 4, S, 16, 19, and 20, which research literature indicates are invalid. These popular invalid signs were found to have much stronger rated, verbal associative connections to male homosexuality than the unpopular valid signs. Six hundred and ninety-three undergraduates (divided among 13 conditions) viewed 30 Rorschach cards on each of which was arbitrarily designated a patient's response and his two symptoms. The 5s "rediscovered" the same invalid Rorschach content signs of homosexuality as the clinicians reported observing in their clinical practice, although these relationships were absent in the experimental materials. They did so regardless of the degree to which the clinically valid signs were valid in the contrived task materials.