The increasing global demand for skilled translators necessitates improved translation education and training. Translation quality assessment (TQA) is crucial for achieving this, yet the assessment practices of translation instructors remain under-researched, especially in Arab countries. This study addresses this gap by documenting the TQA practices of faculty members in Arab universities. The complexity of translation, encompassing cognitive, social, cross-linguistic, and cross-cultural factors, makes TQA challenging. The subjective nature of quality assessment and the lack of universally accepted translation competence frameworks further complicate the issue. Previous research has yielded mixed results on TQA methods, highlighting the need for further investigation into instructor practices and the theoretical underpinnings of their assessments. This study aims to identify these practices and their theoretical bases, contributing to improved translation training and curriculum development in the Arab world.

The literature review emphasizes the complexities inherent in TQA. The definition of "quality" in translation is a contested concept, with subjective judgments influencing assessment practices. Researchers have attempted to replace subjectivity with objectivity in translation assessment, but the inherent subjectivity of the task remains a significant challenge. Existing research highlights several issues: the lack of a common definition of translation quality, the absence of widely accepted structures for translation competence, the ongoing debate about the level of objectivity desirable in TQA, the lack of standardized terminology for evaluation, and the diversity of assessment procedures arising from differing theoretical perspectives on translation. Previous studies, such as those by Firoozkoohi et al. (2012) and Waddington (2001), have shown inconsistencies in assessment criteria used by translation instructors and a reliance on holistic or error-analysis approaches. These inconsistencies underscore the need for this study to explore TQA practices in Arab countries, contributing to a better understanding of assessment literacy among translation educators.

This quantitative study employed a three-part online survey administered to 98 faculty members from 16 universities in Arab countries. Part 1 collected demographic data (gender, job title, experience, specialization). Parts 2 and 3 used assessment instruments: a modified version of the Approaches to Classroom Assessment Inventory (ACAI) to assess general assessment (GA) practices, and a newly developed Translation Quality Assessment Practices Inventory (TQAPI) to assess specialized TQA practices. The ACAI covered assessment purposes, processes, fairness, and theory, while the TQAPI focused on assessment methods, the role of errors, rubric use, objectivity, and focus of assessment (subskills vs. end product). The survey was piloted before main data collection. Data were analyzed using descriptive statistics, exploratory factor analysis, one-way ANOVA, independent samples t-tests, and chi-square tests. Cronbach's alpha and Kolmogorov-Smirnov tests assessed the reliability and normality of the data.

The study found that participants highly endorsed both GA and TQA practices. For GA, mean scores ranged from 3.80 to 4.54, with the highest scores for communication aspects of the assessment process and "assessment for learning." The lowest scores were for assessment of learning and standardized assessment fairness. In TQA, mean scores ranged from 2.98 to 4.45. The most endorsed items related to the significance of errors, assessment methods (especially full-text translation tasks), and objectivity. The least endorsed items involved rubrics, subjective assessment, and less frequently used assessment methods (e.g., fill-in-the-blank, true/false). A significant positive correlation (r = 0.555, p < 0.01) existed between GA and TQA practices. No significant differences in GA or TQA practices were found across gender, job title, or major. However, gender did significantly affect TQA practices, with male faculty members showing higher endorsement of TQA practices. A higher percentage of participants had taken GA-related courses than TQA-related courses, suggesting a need for specialized TQA training.

The high endorsement of both GA and TQA practices indicates a general commitment to effective assessment. The emphasis on communicating assessment results aligns with best practices in feedback provision. However, the lower endorsement of standardized fairness in GA and the emphasis on objectivity in TQA highlight the tension between fairness and the inherent subjectivity of translation quality. The strong endorsement of full-text translation tasks reflects traditional assessment practices, while the lower preference for rubrics might suggest a need for more effective rubric training and integration. The lack of significant differences across demographic variables may be due to the sample's homogeneity or the nature of unified assessment practices. The disparity in training between GA and TQA emphasizes the need for specialized TQA professional development to improve translation assessment practices.

This study provides valuable insights into the TQA practices of translation faculty in Arab countries. The high endorsement of TQA, coupled with identified areas for improvement (e.g., rubric training, objective assessment measures), informs future professional development initiatives. The need for tailored TQA training programs is evident, especially considering the lack of dedicated TQA training within existing translation curricula. Future research should focus on developing and refining the TQAPI instrument, expanding the sample size, and exploring comparative studies across different contexts. Further investigation into the relationship between TQA practices and student outcomes is warranted.

The study's limitations include its relatively small sample size, which focused on instructors specifically teaching translation courses and may limit the generalizability of findings to other contexts. The reliance on self-reported data may introduce bias. The TQAPI instrument, while useful, requires further development and validation. The study's focus on faculty perceptions does not directly measure the effectiveness of their assessment practices on student learning outcomes.