Convergent and discriminant validity of various self-efficacy measures was examined across two studies. In Study 1, U.S. high school students (n=358) rated their self-efficacy in 6 school subjects with reference to specific problems or general self-efficacy statements on the Motivated Strategies for Learning Questionnaire (MSLQ) (P. Pintrich and E. De Groot, 1990). In Study 2, Korean female high school students (n=235) judged their perceived efficacy with reference to specific problems, specific task descriptions, and MSLQ statements in 3 school subjects. Across both studies, the first-order confirmatory factor analyses provide support for both convergent validity of different self-efficacy responses and discriminant validity of perceived self-efficacy across different subject areas. The second-order confirmatory factor analyses confirmed the discriminant validity of self-efficacy beliefs. Substantial method effects were also observed. The problem-and task-referencing methods correlated with each other to a greater extent than they did with the MSLQ self-efficacy scale. (Contains 1 figure, 6 tables, and 45 references.