On Construct Validity of the Concept of Interpersonal Distance

Abstract

In recent years a great deal of research has been done in the area of interpersonal distance. As noted in a review of the literature ( 2 ) , much confusion has characterized the findings. One source of the confusion may lie in the fact that a variety of measuring instruments have been used to assess interpersonal distance. Further confusion may result from the related fact that interpersonal distance as typically defined comprises both the wish for closeness with others and the actual distance maintained in vivo, thus comprising both projective and objective elements. Only a few previous studies have included more than one measure of interpersonal distance; two studies (3, 4 ) reported significant relationships between projective and objective measures, while one study found no significant relationships between diverse measures ( 1 ) . The present investigation was designed to study the degree of relationship between some of the more commonly used measures of projective and objective interpersonal distance and to also explore whether the pattern of relationship between measures was affected by psychopathology. The subjects of this study were therefore 61 process schizophrenic male patients in a Veteran's Administration psychiatric hospital between the ages of 18 and 55 yr. (process-reactive classification determined by Ullmann-Giovannoni scale scores). Patients who evidenced organic involvement, who were given a secondary d~agnosis, who had electroshock therapy within 1 mo. of the experimental procedure, or who were hospitalized for longer than 6 mo. were not included in this study. Thirty male hospital employees (a normal control group) were also included. The three subject groups did not differ significantly in either mean age or racial composition. Four previously reported measures of interpersonal distance were used. These were the Kuethe Felt Figure Technique (using male, female, and rectangular felt figures), the MAPS test (using four figures in flve scenes), a live measure of distance behavior (chair placement with respect to a rdrget person on entering the experimental room), and the Psychological Distance Scale (.I paper-and-pencil measure ,requiring,,the subject to place an X in one of five spaces bemeen concepts such as s~ster or stranger). Pearson correlations were computed between the four distance measures for each of the three subject groups. Of the 18 resulting coefficients, only two were statistically significant ( r = .31 between the Psychological Distance Scale and the Kuethe technique, r = .31 between the Psvchological Distance Scale and MAPS test distance, both obtained with the process sch~zophrenic subjects, p < .05). While the two significant correlations obtained would suggest at least a small degree of relationship benveen a projective and a paper-and-pencil measure of interpersonal distance for process schizophrenics, the results taken as a whole are highly discouraging as regards the construct validity of interpersonal distance. It is clear that interpersonal distance cannot be conceptualized as a simple, straightforward measure of behavior determinable by a variety of methods. In future research considerable attention must be paid to teasing out method variance in the measurement of interpersonal distance.