Abstract
This paper explores the issue of translation competence evaluation by adopting Analytic Hierarchy Process (AHP), for AHP is a systematic and hierarchic evaluation system including a complex multi-objective decision-making method with multiple criteria, and multiple levels of indicators. Combining the quantitative method and the qualitative method, AHP has the merits of evaluating complex and difficult decision-making problems to obtain the optimal solution by calculating maximum weight, so AHP has become an important analytic evaluation tool in various field. Likewise, translation competence evaluation includes a detailed assessment of subcompetences, such as language knowledge application subcompetence, intercultural communication subcompetence, proofreading subcompetence and so on, of which the components interact with each other and are not feasible to calculate accurately, thus formulating a complex decision-making system. However, AHP has the advantages in this aspect. Therefore, this paper attempts to study the adaptability and practicality of AHP in translation competence evaluation.
Introduction
Evaluating translation competence has long been a challenge in both academic and professional spheres. Unlike purely technical skills, translation is a multi-faceted discipline where linguistic fluency, subject matter expertise, and strategic problem-solving overlap in complex ways. Traditionally, assessments have often relied on qualitative "gut-feelings" or subjective grading that fails to account for the intricate interactions between different sub-skills.
To address this, researchers Jijun Wang and Yuan Tian propose a structured framework using the Analytic Hierarchy Process (AHP). Developed by Thomas L. Saaty, AHP is a decision-making framework designed to decompose complex problems into a manageable hierarchy, transforming qualitative judgments into rigorous quantitative data. This paper explores the adaptability of AHP in creating a weighted, objective model for translation competence evaluation.
Methodology: The AHP Framework
The proposed model organizes the evaluation process into three distinct levels: the Target, the Rules (Criteria), and the Scheme. This hierarchical approach allows evaluators to focus on specific components of competence before synthesizing them into a final score.
- The Target: The primary goal, which is the overall Translation Competence Evaluation.
- The Rules (Criteria): Six core sub-competences derived from international (PACTE) and Chinese (CATTI) standards. These include Bilingual subcompetence (BS), Extralinguistic subcompetence (ES), Knowledge about translation (KATS), Instrumental/Tool subcompetence (IS), Strategic subcompetence (SS), and Translation Proofreading (TPS).
- The Scheme: The specific translation or interpretation texts being evaluated (represented as P1, P2, P3 in the model).
The core of the AHP method involves pairwise comparisons. Rather than ranking all six skills simultaneously, the model compares them in pairs using a 1–9 scaling method. These comparisons form a matrix from which specific weights are mathematically derived. To ensure the reliability of these human judgments, a Consistency Ratio (CR) is calculated; a CR of less than 0.1 indicates that the model's logic is sound and consistent.
Results and Weight Distribution
The study successfully calculated specific weights for each sub-competence, providing a blueprint for a weighted scoring system. The results indicate a clear hierarchy of importance among the various skills required for professional translation.
As shown in the data, Bilingual Subcompetence (BS) holds the highest weight at 0.2222 (22.2%), followed closely by Extralinguistic Subcompetence (ES) at 0.2041 (20.4%). This suggests that while language fluency is paramount, subject matter expertise (knowledge of the specific field, such as law or medicine) is nearly as critical for high-quality translation.
The remaining weights are distributed among Knowledge about Translation (18.4%), Strategic Subcompetence (13.7%), Translation Proofreading (12.9%), and Instrumental Subcompetence (12.2%). The model achieved a high level of mathematical consistency, with a Consistency Ratio well below the 0.1 threshold, proving that this structured approach can yield stable, logical results for abstract human skills.
Conclusion and Future Work
By applying these derived weights to a translator's performance in each category, evaluators can generate a single, mathematically grounded "competence score." This moves the field away from vague assessments toward a more transparent and standardized evaluation process.
However, the authors acknowledge that this is currently a design model. While AHP reduces individual bias, the initial 1–9 scaling still relies on expert judgment. Future research should focus on empirical testing, applying this model to a diverse group of translators and texts to verify if these AHP-derived scores accurately reflect real-world quality differences. The ultimate goal is to refine this tool for use in professional certification and academic assessment.
Summary updated