This study investigated the effect of assessment type (self vs. peer vs. teacher) on university students' academic achievement and students' attitudes toward them. In the main study, 82 undergraduate English-as-a-Foreign-Language students in four classes at three universities in Iran were randomly assigned into one of self-, peer- and teacher-assessment or control groups. A pre-test was designed to measure students' initial knowledge in the subject "Teaching Methods". Then experimental groups were subjected to one of the assessment types; however, the control group received no intervention. Thirty-eight students from experimental groups completed an attitude questionnaire. A similarly designed follow-up study was conducted with two classes of BA Chemistry students taking a General English course at Urmia University. The application of one-way analysis of covariance on the main study post-test data indicated differences in performances of all groups, with peer-assessment group performing the best. Quantitative and qualitative analyses indicated that both self- and peer-assessment, as well as the teacher-assessment group, had positive attitudes toward their assessment experiences. The application of one-way analysis of variance on replication study data showed that the experimental group significantly outperformed the control group on the post-test and that students in the former group had positive attitudes toward self-assessment.