This set of analyses is but one example of how Q sort data may lead to analyses yielding information about intra-individual personality patterns. Thc intercsted personalit>y researcher undoubtedly will think of many sets of assumptions built around different Q sort structures. To the extent that patterns derived in this manner are replicablc within specified groups, this approach provides a worthwhile extension of iioriiiativc iiieasurcment. As an example of this, Nahinsky ) denionstrated that estimates of p(c) derived froni intra-sort patterns revealed significant differcriccs between diagnostic groups for this masure. The groups rcprcscnting psychiatric diagnoses showed higher average 1' ( c ) values than did the general popu1at.ion controls for the ideal self sort, suggrsting a greater conflict of values for disturbed individuals than for normals. Thus a measure accessible only to intraindividual analysis resulted in a significant bctwcen-groups variation.Nahinsky (') reported an analysis of variance model for Q sort data involving inter-group difference which enables ail integration with the study of intra-individual patterns. A coxisideration of cach individual as a replicable cxpcrimcrit can lead to a cumulative classification of individuals by types of patterns. The approach coiisidcrcd here might thus lead to series of rcplicablc individual experiments corresponding to typrs with cach cxpcrinicnt providing its own statistical criteria. This in t,urn could bring the more traditional group analyses into the picture. In vicw of the fact that students of personality view it in the context of an intraindividual systciii, would this not be a more direct strategy? REFERENCES 1.
3.
4.
NAHINSKY, I. 1). Q sort approaches in stiirlyingsex, socioeconomic status, and psychopathological NAHINSKY, I. 1). The analysis of v:rriance of Q sort data. J . exp. Educ. (in press). RAWN, M. L. An experimental study of transference and resistance phenomonea in psycho~TErHENSoN, w. T h e S t d y of Behavior.
PROBLEMConsidcrahlc c h f f ort has been devoted to assessing thc personality characteristics of parents of emotionally disturbed children as a prime aspect of the child's socialization milieu. The MMPI has heen used for this purpose with increasing frequency as thv iiccd for a validated objective method which lends itself to routine clinic use has become more apparent. Six studies usirig thc MMPI have found reliable diff crcnc(~s bctwcw t hc parcrils of disl urbcd children and parents of control groups(', 2 p 7 * 9 b 12* 1 3 ) . Orily the last tbrcc have bccri published. While these studies cross validate cach other wc. 11 on this lypc of finding, Liverant's('*) attempt to cstahlish relationships between parwts' hIR/IPI profiles and four groups of their disturbcd children classcd by typc of sympl om was unsuccessful and Ha~ivik's(~) siriiiiar attempt rcsultctd in only a few suggcstivc findings.The prcscwt study had two major aims: a) to compare the MMPI profiles of parciits of clinically disturbed childion with thosc of paren...