Construct Validity of Cognitive-behavioral Treatments for Intimate and Social Loneliness BENEDICT T. McWHIRTER University of Nebraska-Lincoln JOHN J. HORAN Arizona State University The purpose of this study was to compare the effects of two cognitive-behavioral interventions whose contents were matched to distinct types of loneliness (intimate and social) derived from empirical and theoretical literature and from a factor analysis of the Revised UCLA Loneliness Scale. Screened participants were assessed on mea- sures varying in theoretical relevance to the treatment conditions. The hypotheses were that each treatment would favorably influence measures of high theoretical rel- evance (e.g., the social loneliness treatment would reflect changes on measures of social loneliness) and would have comparatively lesser effects on measures of lower theoretical relevance. Analysis of Covariance on pre-, post-, and follow-up data indi- cated that the social loneliness treatment significantly decreased both intimate and social loneliness in the absence of differences on control variables. Issues relevant to the construct validity of the distinct treatments are discussed. Loneliness is a pervasive problem. Although it is not typically the primary problem expressed by clients during initial interviews, practitioners frequently address client loneliness throughout the course of treatment. Hence, the construct of loneliness and appropriate interventions for this experience warrant serious consideration. Cognitive theorists suggest that loneliness evolves from a discrepancy between de- sired and achieved relationships (Peplau, Micheli, & Morasch, 1982; Perlman & Peplau, 1982) and from a constellation of negative, internal, and stable self-attributions about relationship deficiencies (Anderson & Arnoult, 1985; Schultz, & Moore, 1986; Snodgrass, 1987). Behaviorists argue that loneliness derives from deficiencies in skills that are critical for developing appropriate intimate and social relationships (de Jong- Gierveld, 1987; Jones, Hobbs, & Hockenbury, 1982). Both cognitive and behavioral positions have traditionally viewed loneliness as a unidimensional experience. Current research, however, suggests that loneliness is a multidimensional phenomenon (McWhirter, 1990b). Proponents of this perspective argue that people experience not only different degrees but also different kinds of loneliness (Hojat & Crandall, 1987; Richmond & Picken, 1986; Rook, 1984). Weiss (1973, 1987), for example, identified emotional (or intimate) loneliness and social loneliness as two distinct experiences that result from deficits in specific types of Current Psychology: Developmental • Learning • Personality • Social Spring, 1996, Vol. 15, No. 1, 42-52.