Abstract: The present study examined how translation teachers
develop final tests as measures for checking on the quality of
students’ academic translation in Iranian context. To achieve this
goal, thirty experienced male and female translation teachers from the
four types of the universities offering the program were invited to an
in-depth 30-minute one-session semi-structured interview. The
responses provided showed how much discrepancy exists among the
Iranian translation teachers (as developers of final translation tests),
who are least informed with the current translation evaluation
methods. It was also revealed that the criteria they use for developing
such tests and scoring student translations are not theory-driven but
are highly subjective, mainly based on their personal experience and
intuition. Hence, the quality and accountability of such tests are under
serious question. The results also confirmed that the dominant
method commonly and currently practiced is the purely essay-type
format. To remedy the situation, some suggestions are in order. As
part of the solution, to improve the reliability and validity of such
tests, the present summative, product-oriented evaluation should be
accompanied with some formative, process-oriented methods of
evaluation. Training the teachers and helping them get acquainted
with modern principles of translation evaluation as well as the
existing models, and rating scales does improve the quality of
academic translation evaluation.