The aim of the present longitudinal study is to understand the longitudinal relation between depressive symptoms and daily hassles (i.e., general and acculturation hassles) in a high-risk population of unaccompanied refugees. We investigated the validity of three stress-mental health models: the stress exposure model, the stress generation model, and the reciprocal model. Data were collected from 918 unaccompanied refugees in Norway in three waves. Of the initial sample, the majority (82.1%) were male (M
age = 19.01 years, SD = 2.54 years). The data were analyzed with auto-regressive cross-lagged modeling and latent growth curve modeling. The results supported the stress exposure model for the relation between depressive symptoms and acculturation hassles, indicating that acculturation hassles predicted the subsequent levels of depressive symptoms rather than vice versa. On the other hand, the reciprocal model was supported for the relation between depressive symptoms and general hassles indicating a bidirectional, mutual relation. Unconditional latent growth models further showed that depression level remained unchanged over time, while levels of acculturation and general hassles decreased. The implications for clinical practice and immigration policy are discussed.