This paper reports the development and psychometric evaluations of a multidimensional model of loneliness in Australian adolescents. In the first study a new instrument was designed and administered to 1,074 adolescents (ages 10-18 years, M = 13.01). An exploratory factor analysis from data supplied by 694 of these participants yielded a 4-factor structure (friendship, isolation, negative attitude to solitude, and positive attitude to solitude). Competing measurement models were then evaluated using confirmatory factor analysis with data from the remaining 380 participants; strong support was demonstrated for the conceptual model. Significant main effects were evident for geographical location (rural remote/urban), age and sex. In a second study, involving 235 Australian adolescents (ages 10.0-16 years, M = 13.8) the superiority of the first-order model represented by four correlated factors was confirmed. The findings have clinical and practical implications for professional groups represented by child and adolescent psychiatry, pediatric and clinical psychology services, researchers, and educators. Specifically, the new self-report instrument identifies adolescents who are at risk of loneliness and its associated adverse outcomes and in doing has the potential to offer new insights into prevention and intervention.