Integrating Generative Ai Tools Into Efl Writing Evaluation: Teachers’ Perceptions Of Hybrid Assessment Approaches
Auteurs : Dif Imane . Lahmer Mouad .
Résumé
The rapid growth of generative Artificial Intelligence (AI) tools such as ChatGPT, Grammarly GO, and Gemini has transformed writing practices in English as a Foreign Language (EFL) context, raising important questions about academic integrity, authorship, and assessment validity. This study investigates EFL teachers’ familiarity with generative AI tools, their perceptions of institutional policies and ethical considerations surrounding AI use, and their views on integrating AI into writing instruction and assessment. A convergent mixed-methods design was employed, using an online questionnaire administered to 51 EFL teachers from several Algerian universities. The questionnaire included Likert-scale items, multiple-choice questions, and open-ended responses to capture both quantitative and qualitative data. The findings indicated that teachers demonstrated moderate familiarity overall with generative AI tools and acknowledged their potential to support writing development, particularly in grammar correction and revision. However, teachers also expressed concerns about academic integrity, authenticity of student work, and the lack of clear institutional guidelines. Overall, the results highlighted the need for balanced pedagogical approaches that combine responsible AI use with traditional assessment practices to maintain fairness and promote critical thinking in EFL writing assessment.
Mots clés
Academic integrity ; AI in education ; AI-assisted writing ; EFL writing assessment ; Hybrid assessment
