Translation processes and products in L1-to-L2 and L2-to-L1 translations: Insights from keylogging data

Leveraging technological tools in educational research expands investigations into participants’ behaviors and provides insights into learning patterns and teaching effectiveness through innovative data collection and analysis methods. Using keylogging data software, this study explored trainee translators’ process and product behaviors under two scenarios, namely, L1-to-L2 and L2-to-L1 translations of journalistic articles. Based on certain criteria, the sample of the study (N = 18) was selected from an undergraduate language translation program to translate two journalistic texts from and to English and Arabic. The results highlighted differences in the performance and translation processes between the two scenarios (L1-to-L2 and L2-to-L1 translations), although some measures were statistically significant while others were not. Furthermore, the results showed that L2-to-L1 translations were more challenging in orientation, end revision, and cumulative performance than L1-to-L2, whereas L1-to-L2 was more effortful in the measure of clarity. A close reading of the findings also revealed that the trainees’ management of the translation process in L2-to-L1 was better than in L1-to-L2, given that orientation and revision were not given the attention they needed in the L1-to-L2 translation. The study concludes with insights into the inherent complexities in bidirectional translation processes of journalistic articles, which could enhance the effectiveness of bidirectional competence in translation training programs.