Recently, there has been a great deal of interest in understanding the latent organisation of the phenomenology of schizophrenia through examination of the fit of dimensional models to observed symptoms date. A group of 66 DSM-IV paranoid schizophrenic in-patients were assessed three times using the SAPS, SANS, BPRS and PAS. The interrelations between individual symptoms of each scale were examined by means of principal component analysis. The results of factor analysis of the findings from SANS and SAPS confirm the three-factor model, composed of a negative, disorganisation and psychotic factor. Extending the range of symptomatology using BPRS resulted in a five-factor model, composed of the following factors: paranoid, negative, affective, cognitive and disorganised behaviour. In view of the findings based on Strauss' work (1974) the PAS has been added to the SANS, SAPS and BPRS, whose results were examined by factor analysis. The findings indicate that it is possible to consider a six-factor model, composed of the following dimensions: paranoid, negative, affective, cognitive, disorganised behaviour and premorbid social adjustment deficits. The number of factors that best reflect the structure of the symptomatology of paranoid schizophrenia depends on the range of the symptoms under study, i.e., on the type of scales. It follows from our study that six-factor model appears to be the most suitable and clear model in rendering the multidimensionality of paranoid schizophrenia phenomenology.