Present study attempts to find the relationship between tolerance of ambiguity and the students' language proficiency. To this aim, 135 students were randomly selected among Shahed University students. Subjects took a CELT (A Standard Language Proficiency Test) and filled the MacLain's (1993) Second Language Tolerance of Ambiguity questionnaire. They were divided into two groups of low and high tolerant of ambiguity. For processing the data, T-Test and multivariate regression were used and the results indicated that there was a significant relationship between students' English grades in University Entrance Examination as well as their grades of General English course at the university with their tolerance of ambiguity. Moreover, the results from multi variable regression showed that in a linear relation the Entrance exam scores can predict 61% of the variance related to the subjects' English grades. On the other hand, the degree of the tolerance of ambiguity had no significant relations with sex, field of study, Shahed martyrs' first degree relatives or non-Shahed attributes in different groups.