Self-concepts are increasingly being taken into account in explanations of
the efficacy of seen as explanatory factors in the context of psychotherapy
eutical efficacy. In the area of evaluating psychotherapeutic interventions
and in developmental and diagnostic studies they are considered as variabl
es relevant variables offer predicting behavioural predict:ion. A prerequis
ite for this, however, is that are the instruments used were evaluated in c
linical samplesed in the clinical setting and meeting test-statistical crit
eria. The aim of the present study is the test-statistical evaluation of th
e "Frankfurter Selbstkonzeptskalen" (FSKN, Deusinger, 1986a), by focussing
on the dimensions of the scale. The FSKN is a multidimensional scale, often
used for clinical and research purposes, which gives for a differentiated
description of the self-concept which is often used for clinical and resear
ch purposes. We examined 152 schizophrenic inpatients and 180 healthy contr
ols were examined with using the FSKN. The data of both the two samples wer
e analysed according to by classical (traditional?) test-statistical method
s as well as by non-metrical multidimensional scaling (NMDS) on at the item
level. In both samples the Cronbach-values, indicating the internal consis
tency of the scales, were within a satisfactory range for only some of the
scales only. The mean for each scale was comparable to the data gathered by
Deusinger's data (1986a), and very high inter-correlations of the scales c
ould be demonstrated. Neither for the clinical group nor for the healthy co
ntrols could the FSKN scales of the FSKN could be replicated by means of us
ing the NMDS. The results for both groups refer to indicate an over-differe
ntiation of the FSKN. They results were are discussed against the backgroun
d of with respect to the methodological advantage of questionnaire evaluati
on and development by means of provided by the NMDS, can provide in particu
lar in conjunction with the facet theory.