Abstract
This study investigated the psychometric properties of the UCLA Loneliness Scale. Test-retest and internal consistency reliability were found to be highly acceptable, and support for the concurrent related validity of the scale was also indicated. Confirmatory factor analysis provided support for the unidimensionality of the scale. Evidence was found that the scale is not multivariate normal and in fact appears to be bimodal. Implications point to the need to reconsider the nature of the loneliness construct and to develop conceptually or theoretically based models for testing.

This publication has 11 references indexed in Scilit: