Abstract
This research aims to study segmentation (i.e., translation unit (TU)) in students' translations of cultural references and assess the translation quality, using Translog software and a translation task. This product and process-oriented research examines the same independent variable, that is, segmentation, from two angles, namely, student production and translation output. Ten female postgraduate students aged between 28 and 30 performed a computer-based translation task on cultural references. Their translations of English cultural references into Arabic revealed process and production problems (i.e., unsystematic management of translation stages, high mean TUs and time, low mean scores, and deviant translations). The study recommended student training in translation process and product management by employing innovation technological tools (e.g., Translog software and eye-tracking device). Moreover, presenting the cultural dimension should go beyond written translation materials and include audio and visual materials. Sound methodology of training should be adopted, one that draws upon the computational model of human translation, linguistic and cultural models, and models of translation quality assessment.