Introduction: This paper undertakes to investigate the connection between two approaches to scoring translation by examining 115 test papers of the translation component of the C1 level PROFEX English for Medical Purposes (EMP) exam. The main objective of the study is to reveal whether, and to what extent, the method of assessment influences the score.
Material and method: The test papers were scored independently by two experienced raters according to the marking scale of the PROFEX EMP exam, then holistic scoring was carried out by a third rater, who was uninformed of the official scores for the test papers. Correlations were calculated, first between the holistic scores and the official scores based on the combined holistic and discrete point approach, then between other components of the written part of PROFEX EMP exam (reading comprehension and writing) and the holistic scores and the scores reached with the combined method, respectively.
Results: A strong correlation has been revealed between the scores achieved by the purely holistic method and those assessed with the combined holistic and discrete point approach. The holistic method was shown to be slightly more reliable than the combined approach.
Conclusion: The study has revealed that the method of assessment does not significantly influence the score in the evaluation of translation in the PROFEX EMP exam.