Emerging in the 1960s as an area of English language teaching (ELT), English for Specific Purposes (ESP) has been designed to instruct students to meet their academic needs. ESP is characterized by focusing on learners' needs, content knowledge, appropriateness of language used, single-skill instruction (usually reading), and its difference with English for general purposes. The purpose of the present study was to determine the reliability and validity of translation test as a Writing ability assessment in relation to the method of testing, direct Writing. To answer the research questions of this study, 100 Iranian ESP students were selected and tested. The results indicated that translation test was not only reliable, but also a valid testing method for assessing Writing ability of ESP students (0.85, 0.82). Due to the second research question, the effects of proficiency level (between-subject factor) and test method (within-subject factor), the results clearly indicated that ESP students' Writing performance was influenced. In other words, translation compositions were scored better and higher than direct compositions based on the effects of both factors (66.70, 79.86, 89.14), respectively. To see the interactional effects of both factors on the students' Writing ability, the ANOVA results and the Scheffe test revealed that the two factors (test method proficiency level) had an F of (322.01) and (61.89), respectively. In sum, one can clearly observe an increase due to all proficiency levels in the new test (translation composition test). Regarding the third research question, raters' perception on the usefulness of testing types, they strongly believed that higher-level students more enjoyed in direct Writing, while low-level subjects benefited more from translation test type. Implications are discussed.