Abstract
The purpose of this paper was to study the psychometric properties and clinical relevance of the unipolarly scaled Resource Oriented Body Examination (ROBE I). Our aims were to find out if ROBE I could be reduced into a shorter version, labeled ROBE II. We also wanted to know if ROBE II would treat the variation of items in the 10 different subscales in a such a way that it also provided for discrimination between groups of patients with psychosomatic, musculoskeletal and schizophrenic disorders compared to a group of non-patients. A total of 198 subjects were body examined with a ROBE I protocol. The sample data were entered into a separate factor analysis for each domain. Principal components with varimax rotations were used, and the first two factors for each domain were extracted. The original 254 variables were reduced to 144, constituting an instrument for body examination, ROBE II, with 10 subscales. All subscales showed satisfactory internal consistency. Within all but one of the domains, the subscales showed acceptable intercorrelation. All subscales disclosed information of body patterns relevant for tracing psychosomatic symptoms in accordance with Norwegian Psychomotor Physiotherapy (NPMP). The subscales of ROBE II distinguished the bodily characteristics of patients with psychosomatic, musculoskeletal and schizophrenic disorders.