NEW YORK UNIVERSITY, 1985

                         PSYCHOLOGY, CLINICAL (0622); GERONTOLOGY (0351)
 

                         The current study differentiated and measured the concepts of emotional and social loneliness as
                         conceptualized by Weiss (1973) by devising reliable, independent and meaningful scales. Weiss saw
                         emotional loneliness as resulting from the lack of a close, intimate attachment to another person. Social
                         loneliness emerges from the lack of a social network. The subjects studied consisted of 88 elderly
                         people. This sample was chosen given the vulnerability of this age group to the loss of an intimate
                         attachment and shifts in social contact. The emotional loneliness scale that was developed has a reliability
                         of .74 while the social loneliness scale had a reliability of .88. The correlation between the scales was
                         small (r = .26), reflecting their independence. The following variables which related both to the elderly
                         and loneliness were correlated with the scales: attachment, perceived control, social contact,
                         depression, health, affective factors and demographics. Certain variables were significantly related to
                         social loneliness and distinguished between types of loneliness. As predicted, interpersonal control was
                         inversely related to social loneliness. Opposite to prediction, parental attachment was more closely
                         associated with social loneliness. Certain social contact variables (number of friends one has, having or
                         having had a best friend, and satisfaction with one's closest family member) were inversely related to
                         social loneliness. Two affective factors were only related to social loneliness in the regression analyses:
                         feelings of 'marginality' and 'panicky helplessness.' All these variables, including depression, explained
                         68% of the variance in the scale. The strongest predictor was interpersonal control (35%). In the
                         regression analysis, emotional loneliness was predicted by four variables: satisfaction with the quality of
                         friendships, depression and two affective factors ('empty abandonment' and 'impatient/angry
                         vulnerability'). Forty percent of the total variance was explained with depression emerging as the
                         strongest single predictor (11%). The findings support the idea of two clear-cut and meaningful types of
                         loneliness. Social loneliness emerged as a well-predicted scale. Emotional loneliness was not as strongly
                         predicted. Possible reasons for this are discussed.

Social Systems Simulation Group P.O. Box 6904 San Diego, CA  92166-0904 Roland Werner, Principal Phone/FAX  (619) 660-1603