Immigrant and nonimmigrant women: factors that predict leaving an abusive relationship.

This research used logistic regression to test components of Choice and Lamke's (1997) two-part decision-making model and Hamby's (2008) holistic risk assessment as predictors of the decision to separate from an abusive partner, comparing significant predictors for immigrant (n = 497) and nonimmigrant (n = 808) women. Findings demonstrated that immigrant women reported higher levels of perceived risks/barriers to leaving, provided some support for the use of a holistic risk assessment in understanding women's decisions to leave, and demonstrated that immigrant and nonimmigrant women have both similarities and differences in the factors that predict leaving. Clinical and policy implications are addressed.