PROBLEM
I n experimental studies on college populations, Christie and Geis (2) have demonstrated that persons who endorse Machiavellian attitudes are less scrupulous in their behavior than others who do not have this attitude pattern. The most salient behavioral difference between Machiavellians and persons who reject Machiavellian attitudes is that Machiavellian individuals successfully manipulate others, while resisting attempts by others to manipulate them.
At the same time that Machiavellians are by definition and, apparently, often in fact, wily, cunning, deceitful, the endorsement of Machiavellian attitudes and the manifestation of Machiavellian behaviors may be, to a certain extent, "healthy" responses. Assuming a varying but typically positive correlation between attitudes and behavior(3), it can be suggested that the nonmanipulative behavior of the schizophrenic person, described as lack of purpose and interpersonal isolation ( 5 ) 1 is accompanied by a rejection of manipulative or Machiavellian values. i f schizophrenic persons are less Machiavellian than members of the normal (nondiagnosed) population, then it further may be suggested that remission of schizophrenic behavior is associated with an increase in manipulation of the social environment on the one hand and an increase in endorsement of Machiavellian attitudes on the other.
This study compares a normal (Le., nondiagnosed) population with a clinical population on endorsement of Machiavellian attitudes as measured by responses to Christie's Mach (Machiavellian) Scales. It was predicted that a normal population would endorse more strongly Machiavellian attitudes than would a clinical population with a diagnosis of schizophrenia, and that ex-hospital patients would endorse more strongly Machiavellian attitudes than would in-hospital patients. I n studies involving nonpsychiatric adult populations, males score consistently higher on the Mach Scales than females, and scores on the forced-choice form of the Scale are reliably higher than scores on the Likert form. A sex effect and a test form effect were therefore also expected.
*This study was conducted while the authors were members of the profeysiond s!,aff of Horizon House, Inc., Philadelphia. The authors would like