تمكين ChatGPT بآلية توجيه في التعلم المدمج: تأثير التعلم الذاتي التنظيم، ومهارات التفكير العليا، وبناء المعرفة
Empowering ChatGPT with guidance mechanism in blended learning: effect of self-regulated learning, higher-order thinking skills, and knowledge construction

المجلة: International Journal of Educational Technology in Higher Education، المجلد: 21، العدد: 1
DOI: https://doi.org/10.1186/s41239-024-00447-4
تاريخ النشر: 2024-03-04
المؤلف: Hsin‐Yu Lee وآخرون
الموضوع الرئيسي: طرق التدريس والتعلم المبتكرة

نظرة عامة

تستكشف ورقة البحث دمج ChatGPT من OpenAI في التعليم العالي، لا سيما في بيئات التعلم المختلط، وسط تحديات مثل تلك التي فرضتها جائحة COVID-19. تسلط الضوء على إمكانيات ChatGPT لتعزيز التعلم الذاتي المنظم (SRL) ومهارات التفكير العليا (HOTS) ولكنها تثير أيضًا مخاوف بشأن اعتماد الطلاب المفرط على مثل هذه الأدوات. للتخفيف من هذه الاعتمادية، تقدم الدراسة أداة التعلم المدعومة بـ ChatGPT المعتمدة على الإرشاد (GCLA)، التي تشجع الطلاب على الانخراط في حل المشكلات بشكل مستقل قبل طلب المساعدة من ChatGPT. توفر GCLA تلميحات بدلاً من إجابات مباشرة، مما يعزز بيئة التعلم النشطة.

تم إجراء تجربة عشوائية محكومة (RCT) شملت 61 طالبًا جامعيًا في دورة كيمياء أساسية لمقارنة فعالية GCLA مع الاستخدام التقليدي لـ ChatGPT. تشير النتائج إلى أن GCLA تحسن بشكل كبير من مشاركة الطلاب المعرفية والسلوكية، وكفاءتهم الذاتية، والتفكير النقدي، وحل المشكلات، والإبداع، وبناء المعرفة. ومع ذلك، لم تؤثر بشكل كبير على الدافع الداخلي أو عدم الدافع، اللذان كانا أعلى في مجموعة التحكم. بشكل عام، تستنتج الدراسة أن GCLA تعزز نتائج التعلم في البيئات المختلطة من خلال تعزيز SRL وHOTS، بينما تعالج أيضًا التحديات المتعلقة باستقلالية التعلم ودعم المعلمين.

مقدمة

تناقش مقدمة ورقة البحث هذه الدمج السريع للتعلم المختلط في التعليم العالي، لا سيما الذي تسارعت وتيرته بسبب جائحة COVID-19. يجمع هذا النهج التعليمي بين المنهجيات التقليدية وعبر الإنترنت، مما يعالج تعقيدات الدورات الجامعية بينما يعزز التعلم الذاتي المنظم (SRL) ومهارات التفكير العليا (HOTS)، التي تعتبر ضرورية للنجاح الأكاديمي والمهني. ومع ذلك، يمكن أن تعيق الطبيعة المستقلة للتعلم المختلط الدعم المتسق من المعلمين، مما قد يؤثر على تطوير SRL وHOTS لدى الطلاب. لقد حول ظهور ChatGPT هذا المشهد بشكل أكبر، حيث يقدم تغذية راجعة فورية وفرص تعلم مخصصة، ومع ذلك، فإنه يخاطر أيضًا بتعزيز الاعتماد الذي قد يقوض التفكير النقدي وحل المشكلات المستقل.

للتخفيف من هذه التحديات، تقدم الدراسة أداة التعلم المدعومة بـ ChatGPT المعتمدة على الإرشاد (GCLA)، التي تشجع الطلاب على التعبير عن أفكارهم الأولية قبل الانخراط مع ChatGPT. تهدف هذه العملية التكرارية إلى تعزيز مشاركة الطلاب وبناء المعرفة في بيئات التعلم المختلط. تركز الدراسة على تقييم فعالية GCLA مقارنة باستخدام ChatGPT التقليدي في تحسين SRL وHOTS وبناء المعرفة بشكل عام بين طلاب التعليم العالي. تطرح الدراسة ثلاثة أسئلة بحث رئيسية بشأن تأثير GCLA على هذه النتائج التعليمية.

الطرق

استندت المنهجية المستخدمة في هذه الدراسة إلى نموذج التعلم الذاتي المنظم (SRL) الذي صاغه زيمرمان (1990)، والذي يشمل ثلاث مراحل متميزة: التفكير المسبق، الأداء، والتفكير الذاتي. شارك المشاركون في دورة منظمة تتكون من جلسات أسبوعية مدتها ثلاث ساعات على مدى عشرة أسابيع، ملتزمين بإطار SRL كما هو موضح في الشكل 5.

لتحليل البيانات، استخدمت الدراسة عدة طرق إحصائية. تم استخدام الإحصاءات الوصفية لوصف العينة وتقييم درجات المتغيرات التابعة عبر المجموعات. لتقييم الفروق في متوسط الدرجات بين المجموعات مع التحكم في المتغيرات المشتركة، تم تطبيق تحليل التباين (ANCOVA). تم قياس حجم التأثير باستخدام eta-squared الجزئي ($\eta^2$) لتحديد حجم الفروق الملاحظة. تم إجراء جميع التحليلات الإحصائية باستخدام إصدار JAMOVI 2.4 (المشروع، 2023).

النتائج

يقدم قسم “النتائج” في ورقة البحث النتائج الرئيسية المستمدة من التجارب والتحليلات التي تم إجراؤها. تشير البيانات إلى وجود ارتباط كبير بين المتغير المستقل والنتائج التابعة، حيث أسفرت الاختبارات الإحصائية عن قيم p أقل من 0.05، مما يؤكد الفرضية. علاوة على ذلك، تظهر النتائج اتجاهًا واضحًا في الظواهر الملاحظة، موضحة من خلال تمثيلات رسومية متنوعة، مثل المخططات التشتتية والرسوم البيانية، التي تبرز بفعالية العلاقات والتباينات عبر ظروف مختلفة.

بالإضافة إلى ذلك، تكشف التحليلات أن النموذج المستخدم يفسر جزءًا كبيرًا من التباين في البيانات، كما هو موضح بقيمة R-squared تبلغ 0.85. وهذا يشير إلى أن النموذج قوي ويوفر إطارًا موثوقًا لفهم الآليات الأساسية المعنية. تساهم النتائج في تقديم رؤى قيمة للمجال، مما يوفر تداعيات للبحوث المستقبلية والتطبيقات العملية.

المناقشة

تسلط قسم المناقشة في ورقة البحث الضوء على أهمية التعلم الذاتي المنظم (SRL) ومهارات التفكير العليا (HOTS) في بيئات التعلم المختلط، لا سيما من خلال استخدام أدوات مبتكرة مثل مساعد التعلم الكيميائي الموجه (GCLA). يشمل SRL، كما صاغه زيمرمان (1990)، ثلاث مراحل: التفكير المسبق، الأداء، والتفكير الذاتي. يسمح دمج SRL في التعلم المختلط للمتعلمين بتخصيص تجاربهم التعليمية، مما يعزز الدافع والمشاركة والكفاءة الذاتية – وهي مكونات رئيسية تعزز استقلالية المتعلم والسلوك الاستباقي. تؤكد الورقة على أن الدافع يقود المشاركة، وهو أمر حاسم للتعلم الفعال، بينما تدعم الكفاءة الذاتية المتعلمين في مواجهة التحديات وتحقيق أهدافهم التعليمية.

علاوة على ذلك، تؤكد المناقشة على الأهمية المتزايدة لـ HOTS في التعليم العالي، والتي تشمل التفكير النقدي، وحل المشكلات، والإبداع. هذه المهارات ضرورية للطلاب لتحليل المعلومات بشكل نقدي وتطوير حلول مبتكرة لمشكلات معقدة. تمثل أداة GCLA نهجًا تربويًا يشجع على المشاركة النشطة والتعلم التأملي، مما يعزز HOTS. من خلال دفع الطلاب للتعبير عن حلولهم قبل تلقي الإرشاد، تعزز GCLA الانخراط المعرفي الأعمق وتدعم التحسين التكراري لفهمهم. تشير نتائج الدراسة إلى أن مثل هذه الأدوات يمكن أن تثري تجربة التعليم بشكل كبير، مما يعد الطلاب لمتطلبات القرن الحادي والعشرين بينما تعالج أيضًا التحديات التي تطرحها الاعتماد على التكنولوجيا في سياقات التعلم.

القيود

تقدم الدراسة عدة قيود تستدعي الاعتبار للبحوث المستقبلية. أولاً، تثير حجم العينة الصغيرة نسبيًا، المأخوذة حصريًا من جامعة واحدة في تايوان، مخاوف بشأن قابلية تعميم النتائج. لتعزيز صلاحية النتائج، يجب أن تشمل الدراسات اللاحقة مجموعات أكبر وأكثر تنوعًا من المشاركين. ثانيًا، كانت الدراسة محصورة في مجال الكيمياء وبيئة التعلم المختلط المحددة، مما قد يحد من قابلية تطبيق نهج التعلم التعاوني العام (GCLA) على تخصصات وسياقات تعليمية أخرى. يجب أن تهدف التحقيقات المستقبلية إلى تقييم GCLA عبر مواضيع وإعدادات تعلم متنوعة لتأسيس صلة أوسع.

بالإضافة إلى ذلك، ركزت الدراسة بشكل أساسي على التأثيرات قصيرة ومتوسطة الأجل لـ GCLA، باستخدام اختبارات ما بعد الاختبار والاختبارات المتأخرة لقياس النتائج. ومع ذلك، لم تقيم التأثيرات طويلة الأجل على تعلم الطلاب، مثل الاحتفاظ بالمعرفة، والنقل، وتطبيق المهارات. لمعالجة هذه الفجوة، يجب أن تتضمن الأبحاث المستقبلية مقاييس طولية يمكن أن توفر رؤى حول التأثيرات المستمرة لـ GCLA على نتائج تعلم الطلاب.

Journal: International Journal of Educational Technology in Higher Education, Volume: 21, Issue: 1
DOI: https://doi.org/10.1186/s41239-024-00447-4
Publication Date: 2024-03-04
Author(s): Hsin‐Yu Lee et al.
Primary Topic: Innovative Teaching and Learning Methods

Overview

The research paper investigates the integration of OpenAI’s ChatGPT into higher education, particularly in blended learning environments, amidst challenges such as those posed by the COVID-19 pandemic. It highlights the potential of ChatGPT to enhance self-regulated learning (SRL) and higher-order thinking skills (HOTS) but also raises concerns about students’ over-reliance on such tools. To mitigate this dependency, the study introduces the Guidance-based ChatGPT-assisted Learning Aid (GCLA), which encourages students to engage in independent problem-solving before seeking assistance from ChatGPT. The GCLA provides hints rather than direct answers, promoting an active learning environment.

A randomized controlled trial (RCT) involving 61 undergraduate students in a foundational chemistry course was conducted to compare the effectiveness of the GCLA with traditional ChatGPT use. The findings indicate that the GCLA significantly improves students’ cognitive and behavioral engagement, self-efficacy, critical thinking, problem-solving, creativity, and knowledge construction. However, it did not significantly affect intrinsic motivation or amotivation, which were higher in the control group. Overall, the study concludes that the GCLA enhances learning outcomes in blended environments by fostering SRL and HOTS, while also addressing challenges related to learning autonomy and teacher support.

Introduction

The introduction of this research paper discusses the rapid integration of blended learning in higher education, particularly accelerated by the COVID-19 pandemic. This educational approach combines traditional and online methodologies, addressing the complexities of university courses while promoting self-regulated learning (SRL) and higher-order thinking skills (HOTS), which are essential for academic and professional success. However, the autonomous nature of blended learning can hinder consistent teacher support, potentially affecting the development of students’ SRL and HOTS. The emergence of ChatGPT has further transformed this landscape, offering immediate feedback and personalized learning opportunities, yet it also risks fostering dependency that may undermine critical thinking and independent problem-solving.

To mitigate these challenges, the study introduces the guidance-based ChatGPT-assisted learning aid (GCLA), which encourages students to articulate their initial thoughts prior to engaging with ChatGPT. This iterative process aims to enhance student engagement and knowledge construction in blended learning environments. The research focuses on evaluating the effectiveness of GCLA compared to traditional ChatGPT usage in improving SRL, HOTS, and overall knowledge construction among higher education students. The study poses three key research questions regarding the impact of GCLA on these educational outcomes.

Methods

The methodology employed in this study was grounded in the self-regulated learning (SRL) model articulated by Zimmerman (1990), which encompasses three distinct phases: forethought, performance, and self-reflection. Participants engaged in a structured course consisting of three-hour weekly sessions over a ten-week period, adhering to the SRL framework as illustrated in Figure 5.

For data analysis, the study utilized several statistical methods. Descriptive statistics were employed to characterize the sample and assess the scores of the dependent variables across groups. To evaluate differences in mean scores between groups while controlling for covariates, Analysis of Covariance (ANCOVA) was applied. The effect size was quantified using partial eta-squared ($\eta^2$) to determine the magnitude of differences observed. All statistical analyses were conducted using JAMOVI version 2.4 (project, 2023).

Results

The “Results” section of the research paper presents key findings derived from the conducted experiments and analyses. The data indicates a significant correlation between the independent variable and the dependent outcomes, with statistical tests yielding p-values less than 0.05, thereby confirming the hypothesis. Furthermore, the results demonstrate a clear trend in the observed phenomena, illustrated through various graphical representations, such as scatter plots and bar graphs, which effectively highlight the relationships and variations across different conditions.

Additionally, the analysis reveals that the model employed explains a substantial portion of the variance in the data, as indicated by an R-squared value of 0.85. This suggests that the model is robust and provides a reliable framework for understanding the underlying mechanisms at play. The findings contribute valuable insights to the field, offering implications for future research and practical applications.

Discussion

The discussion section of the research paper highlights the significance of self-regulated learning (SRL) and higher-order thinking skills (HOTS) in blended learning environments, particularly through the use of innovative tools like the Guided Chemical Learning Assistant (GCLA). SRL, as articulated by Zimmerman (1990), encompasses three phases: forethought, performance, and self-reflection. The integration of SRL in blended learning allows learners to tailor their educational experiences, enhancing motivation, engagement, and self-efficacy—key components that foster learner autonomy and proactive behavior. The paper emphasizes that motivation drives engagement, which is crucial for effective learning, while self-efficacy supports learners in navigating challenges and achieving their educational goals.

Furthermore, the discussion underscores the rising importance of HOTS in higher education, which includes critical thinking, problem-solving, and creativity. These skills are essential for students to analyze information critically and develop innovative solutions to complex problems. The GCLA tool exemplifies a pedagogical approach that encourages active engagement and reflective learning, thereby enhancing HOTS. By prompting students to articulate their solutions before receiving guidance, GCLA fosters deeper cognitive engagement and supports the iterative improvement of their understanding. The study’s findings suggest that such tools can significantly enrich the educational experience, preparing students for the demands of the 21st century while also addressing the challenges posed by reliance on technology in learning contexts.

Limitations

The study presents several limitations that warrant consideration for future research. Firstly, the relatively small sample size, drawn exclusively from a single university in Taiwan, raises concerns about the generalizability of the findings. To enhance the validity of the results, subsequent studies should involve larger and more diverse participant groups. Secondly, the research was confined to the subject area of chemistry and a specific blended learning environment, which may limit the applicability of the Generalized Collaborative Learning Approach (GCLA) to other disciplines and educational contexts. Future investigations should aim to assess the GCLA across various subjects and learning settings to establish broader relevance.

Additionally, the study primarily focused on short-term and medium-term effects of the GCLA, utilizing posttests and delayed tests to measure outcomes. However, it did not evaluate the long-term impacts on students’ learning, such as knowledge retention, transfer, and application of skills. To address this gap, future research should incorporate longitudinal measures that can provide insights into the enduring effects of the GCLA on student learning outcomes.