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.