The Effects of Positively and Negatively Worded Items on the Factor Structure of the UCLA Loneliness Scale

The purpose of this study was to evaluate the factor structure of the University of California, Los Angeles (UCLA) Loneliness Scale and examine possible wording effects on a sample of 1,429 students from the United Arab Emirates University. Correlated traits-correlated uniqueness as well as correlated traits-correlated methods were used to examine the scale’s eight-factor models. The results of confirmatory factor analysis with maximum likelihood estimation indicated two models provided the best fit. The first, a one-dimensional model, assumes the correlation of the standard errors in the positive items. The second assumes a one-dimensional model and two method factors that correspond to the negative and positive items.