The purpose of the present study was to investigate the predictive value of gender, attachment dimensions and family environment in explaining loneliness among students. The study included 473 students (281 females, 192 males) from Ankara University. The UCLA Loneliness Scale, Family Environment Assessment Scale and Experiences in Close Relationships - Revised Questionnaire were administered to all participants. The results of the regression analyses demonstrated that the avoidance dimension of attachment accounted for 11%, avoidance and anxiety 14%, all attachment dimensions and family environment 18%, and all variables together (attachment dimensions, family environment, gender) accounted for 19% of variance in loneliness. The present study demonstrated that insecurely attached males with low-coherent families reported the highest degree of loneliness. On the contrary, securely attached females with high-coherent families reported the lowest degree of loneliness.