Earticle

This study explores the differential role of creativity in Iranian EFL learners’ performance on argumentative and narrative written tasks. The study involved the measurement of learners’ (N=70) creativity using the Abbreviated Torrance Test for Adults (ATTA) in terms of features of fluency, elaboration, flexibility and originality as well as the elicitation of the complexity, accuracy, and fluency of their performance on written narrative and argumentative tasks. Whereas the argumentative task required that learners wrote an argumentative essay giving their opinions concerning the effect of technology on human’s life, the narrative task involved learners in narrating a story based on a picture as a visual cue. With regards to the argumentative task, the results shows negative relationships between two measures of argumentative task fluency and students’ creative originality score. However, the syntactic complexity of argumentations was affected by creative elaboration. Concerning the narrative task, significant positive correlations were observed between creative fluency and the three fluency measures of narrations. Besides, all three fluency measures correlated positively with the total creativity score. Yet the number of words was found to be negatively correlated with the creative originality. The findings are discussed in the context of language teaching and learning.

The purpose of this study is to examine the effects of task complexity on the quality of L2 learners’ argumentative writing using both global measures of complexity, accuracy, and fluency (CAF) and task-specific measures, namely, for conjunctions. A group of 110 Korean high school students in South Korea performed either a simple or complex argumentative writing task. Task complexity was manipulated by +/?reasoning demands and +/?few elements. A set of 110 argumentative essays were analyzed on 6 global measures of CAF and 2 task-specific measures. The results showed that task complexity affected the fluency of the argumentative writings, in that the complex task group produced more fluent writings than the simple group. However, task complexity did not affect accuracy or syntactic complexity of the argumentative writings. In the task-specific measures, task complexity affected neither frequency nor target-like use of conjunctions. These results have pedagogical implications for task design to help learners develop their L2 proficiency.

본 연구의 목적은 토의⋅토론 중심 대학 논증 글쓰기 수업에서 교수자가 조성하는 자율성 지지, 구조 제공, 관계성 지원의 수업 환경이 학습자의 자기조정학습 전략 조정과 어떠한 관련성을 보이는지를탐구하는 것이다. 또한, 자기조정학습 전략 조정이 수행 지속성과 수행 양상에 어떻게 연결되는지, 수행 경험에 대한 전략 성찰과 전략 변화가 자기조정학습 전략 조정과 어떠한 관련성을 보이는지를질적 다중 사례연구를 통해 살펴보고자 하였다. 본 연구는 동일 교과목 내 두 개의 수업 맥락에서목적표집을 통해 사례를 선정한 질적 다중 사례연구로 설계되었다. 연구 참여자는 D대학 <대학 글쓰기> 수강생 7명이며, 개요 및 서론 작성 자료, 발표 및 토의⋅토론 자료, 장문 논증 글쓰기 과제, 성찰지등을 수집하여 사례 내 분석과 사례 간 비교 분석을 실시하였다. 연구 결과, 자율성 지지적 수업 환경은학습자의 자기조정학습 전략 조정과 수행 지속성 및 수행 양상과 관련된 것으로 나타났다. 학습자들은목표설정, 시간관리, 도움요청, 자기점검 전략을 각기 다른 방식으로 조정하였다. 이러한 전략 조정은논증 확장, 단계적 수행 지속, 참여 태도 변화, 생성형 AI 활용 방식 등의 차이로 나타났다. 또한, 수행 경험에 대한 전략 성찰은 자신의 전략 사용과 수행 과정에 대한 재인식으로 이어졌으며, 일부사례에서는 이러한 성찰이 논증 구성 방식과 수행 전략의 재조정으로 연결되었다. 이는 자기조정학습전략 조정이 수행 경험과 성찰을 바탕으로 지속적으로 수정⋅재구성되는 순환적 과정으로 나타날수 있음을 시사한다.

The purpose of this study was to explore the relationship between the autonomy-supportive learning environment created by instructors?consisting of autonomy support, structure, and relatedness support?and students’ self-regulated learning (SRL) strategy adjustment in discussion- and debate-based university argumentative writing courses. In addition, this study examined how SRL strategy adjustment was associated with task persistence and performance patterns, as well as how strategy reflection and strategy readjustment following task experiences were related to SRL strategy adjustment. A qualitative multiple-case study design was employed. Seven students enrolled in a university writing course at D University were selected through purposive sampling from two instructional contexts within the same course. Data sources included outlines and introductory drafts, presentation and discussion materials, extended argumentative writing assignments, and reflective journals. Both within-case and cross-case analyses were conducted. The findings revealed that the autonomy-supportive learning environment was closely related to students’ SRL strategy adjustment, task persistence, and performance patterns. Participants adjusted goal-setting, time-management, help-seeking, and self-monitoring strategies in different ways. These differences were reflected in variations in argument development, stepwise task persistence, changes in participation attitudes, and patterns of generative AI use. In addition, reflection on task experiences led students to reexamine their strategy use and learning processes. In some cases, such reflection resulted in the readjustment of argument construction and task-performance strategies. The findings suggest that SRL strategy adjustment can be understood as a cyclical process that is continuously revised and reconstructed through task experiences and reflective practices.

본 연구는 논증적 글쓰기 평가에서 과제 맥락에 따라 달라지는 평가 준거의 적용 논리를 형식적으로 구조화하기 위하여 쓰기 평가 온톨로지(Writing Assessment Ontology; WAO)를 설계․구현하였다. 이를 위해 논증적 작문 과제를 네 유형으로 유형화하고, 내용․조직․표현으로 구성된 질적 평가 요인을 과제 맥락과 연계하여 클래스, 객체 속성, 공리 체계로 모델링하였다. WAO는 Protégé 환경에서 OWL 2 DL로 구현되었으며, 시뮬레이션된 학생 글 인스턴스를 활용한 추론 실험을 통해 과제 유형별 필수 요소의 충족 여부에 따라 정상 응답과 결함 사례를 구조적으로 구분할 수 있음을 확인하였다. 본 연구는 쓰기 평가 준거를 온톨로지 수준에서 형식화하려는 초기 시도로서 평가자의 판단 논리를 구조화 가능한 지식으로 전환하기 위한 기초 틀을 제시한다.

This study designs and implements a Writing Assessment Ontology (WAO) to formally represent how evaluation criteria in argumentative writing assessment are applied differently according to task context. Argumentative writing tasks are typologized, and qualitative evaluation factors?content, organization, and expression?are modeled as ontology classes, object properties, and axioms in relation to task context. The proposed WAO is implemented in an OWL 2 DL environment using Protege, and reasoning experiments demonstrate that it can logically distinguish well-formed responses from defective cases across task types. This study presents a foundational framework for transforming evaluators’ judgment processes into structurally representable knowledge.