دراسة قبول الطلاب واستخدامهم لـ ChatGPT في التعليم العالي السعودي Examining Students’ Acceptance and Use of ChatGPT in Saudi Arabian Higher Education

المجلة: European Journal of Investigation in Health Psychology and Education، المجلد: 14، العدد: 3
DOI: https://doi.org/10.3390/ejihpe14030047
PMID: https://pubmed.ncbi.nlm.nih.gov/38534908
تاريخ النشر: 2024-03-17

دراسة قبول الطلاب واستخدامهم لـ ChatGPT في التعليم العالي السعودي

أبو النصر إي. صبحي إبراهيم أ. الشاعر (د) وأحمد م. حسنين (د)1 قسم الإدارة، كلية إدارة الأعمال، جامعة الملك فيصل، الأحساء 31982، المملكة العربية السعودية؛ ielshaer@kfu.edu.sa (I.A.E.); aabdelrazek@kfu.edu.sa (A.M.H.)2 كلية السياحة وإدارة الفنادق، جامعة حلوان، القاهرة 12612، مصر كلية السياحة وإدارة الفنادق، جامعة قناة السويس، الإسماعيلية 41522، مصر* المراسلة: asobaih@kfu.edu.sa

الاقتباس: صبحي، أ.ع.ع؛ الشاعر، إ.ع.؛ حسنين، أ.م. دراسة قبول الطلاب واستخدامهم لـ ChatGPT في التعليم العالي السعودي. المجلة الأوروبية للتحقيق في الصحة وعلم النفس والتعليم. 2024، 14، 709-721.https://doi.org/10.3390/ejihpe14030047
المحررون الأكاديميون: ماريا ديل مار سيمون ماركيز وهيونسو لي
تاريخ الاستلام: 3 يناير 2024
تمت المراجعة: 10 مارس 2024
تم القبول: 15 مارس 2024
نُشر: 17 مارس 2024
حقوق الطبع والنشر: © 2024 من قبل المؤلفين. المرخص له MDPI، بازل، سويسرا. هذه المقالة هي مقالة مفتوحة الوصول موزعة بموجب الشروط والأحكام لرخصة المشاع الإبداعي النسب (CC BY) (https://creativecommons.org/licenses/by/ 4.0/).

الملخص

تدرس هذه الدراسة قبول الطلاب واستخدامهم لـ ChatGPT في التعليم العالي في المملكة العربية السعودية، حيث يزداد الاهتمام باستخدام هذه الأداة منذ إطلاقها في عام 2022. تم جمع بيانات البحث الكمي، من خلال استبيان ذاتي يعتمد على “النظرية الموحدة لقبول واستخدام التكنولوجيا” (UTAUT2)، من 520 طالبًا في إحدى الجامعات العامة في السعودية في بداية الفصل الدراسي الأول من العام الدراسي 2023-2024. دعمت نتائج نمذجة المعادلات الهيكلية جزئيًا UTAUT والبحوث السابقة فيما يتعلق بالتأثير المباشر الكبير لتوقع الأداء (PE) والتأثير الاجتماعي (SI) وتوقع الجهد (EE) على النية السلوكية (BI) لاستخدام ChatGPT، والتأثير المباشر الكبير لـ PE وSI وBI على الاستخدام الفعلي لـ ChatGPT. ومع ذلك، لم تدعم النتائج الأبحاث السابقة فيما يتعلق بالعلاقة المباشرة بين الظروف الميسرة (FCs) وكلا من BI والاستخدام الفعلي لـ ChatGPT، والتي وُجد أنها سلبية في العلاقة الأولى وغير ذات دلالة في الثانية. كانت هذه النتائج بسبب غياب الموارد والدعم والمساعدة من المصادر الخارجية فيما يتعلق باستخدام ChatGPT. أظهرت النتائج وجود وساطة جزئية لـ BI في الرابط بين PE وSI وFC والاستخدام الفعلي لـ ChatGPT في التعليم، ووساطة كاملة في الرابط بين BI وEE والاستخدام الفعلي لـ ChatGPT في التعليم. تقدم النتائج العديد من الدلالات للباحثين ومؤسسات التعليم العالي في السعودية، والتي تهم أيضًا مؤسسات أخرى في سياقات مشابهة.

الكلمات المفتاحية: شات جي بي تي؛ روبوت الدردشة؛ الذكاء الاصطناعي؛ قبول التكنولوجيا؛ استخدام التكنولوجيا؛ التعليم في المملكة العربية السعودية

1. المقدمة

منذ تدشين نموذج المحادثة المدرب مسبقًا (ChatGPT) في عام 2022، زاد الاهتمام بين الطلاب باستخدام ChatGPT في التعليم [1]. ChatGPT هو روبوت محادثة تم إنشاؤه بواسطة OpenAI، والذي يساعد المستخدمين في إنشاء نصوص تشبه النصوص البشرية من خلال معالجة الأوامر والمحادثات [2]. أظهرت الدراسات حول استخدام ChatGPT في سياق التعليم العالي، انظر على سبيل المثال، المراجع [1-4]، العديد من الفوائد لاستخدام ChatGPT في التعليم، مثل الحصول على خدمة سريعة وفورية، والدعم الأكاديمي وغير الأكاديمي؛ وتطوير مهارات اللغة؛ واستخدام خدمات التحرير. كما أظهرت الأبحاث السابقة أن ChatGPT يساعد الطلاب في واجباتهم ومشاريعهم وأعمالهم المنزلية ومهام البحث. ومع ذلك، هناك بعض المخاوف بين العلماء وقيادة التعليم العالي بشأن دمج ChatGPT في التعليم العالي واستخدامه لدعم التدريس والتعلم [5-8]. تشمل المخاوف الأخرى توليد معلومات متحيزة، واستشهادات مزيفة، ومخاوف تتعلق بالموثوقية وحقوق الملكية الفكرية، مما يثير إنذارات أخلاقية حول المحتوى المقدم من قبل الطلاب. علق حسنين وصبحي [1] على التأثير المدمر لـ ChatGPT على النزاهة الأكاديمية وسوء استخدام المعلومات المقدمة من ChatGPT. علاوة على ذلك، هناك مخاوف أخرى بين العديد من العلماء، مثل المراجع [1،9]، بشأن دمج ChatGPT في التعليم؛ تتعلق هذه المخاوف بنتائج التعلم ومهارات الطلاب، وخاصة التفكير النقدي، والتي هي
من المتوقع أن يتأثر سلبًا. لذلك، تم الإشارة إلى أنه يجب على المؤسسات الأكاديمية اتخاذ الحذر لمنع الاعتماد المفرط من قبل الطلاب على استخدام ChatGPT لأغراض أكاديمية [1].
لقد كانت هناك بعض المحاولات من قبل العلماء، مثل المراجع [2،6]، لفهم اعتماد الذكاء الاصطناعي، مثل ChatGPT، في التعليم. ومع ذلك، كانت هناك محاولات قليلة (على الرغم من تزايد عدد الدراسات) لفهم المتغيرات التي تدفع الطلاب لاستخدام ChatGPT في التعليم [1،4،10]. تناولت دراسة العتيبي والشيهري [10] الفوائد المحتملة والعقبات المرتبطة بدمج نتائج التعلم المدفوعة بالذكاء الاصطناعي ضمن نظام التعليم العالي في المملكة العربية السعودية. وأكدت نتائجهم أن الذكاء الاصطناعي لا يزال في مرحلة تطوير مبكرة في مجال التعليم، ومع ذلك فإن أهميته لمؤسسات التعليم العالي لا يمكن إنكارها. إن تبني هذه التكنولوجيا التحويلية أمر ضروري لمواجهة التحديات التعليمية القادمة. ومن المهم بنفس القدر ضمان أن جميع الطلاب السعوديين يطورون الكفاءات التقنية اللازمة للتفاعل مع مجال الذكاء الاصطناعي والمساهمة فيه في المستقبل.
حدد ستريزليكي [4] فجوة بحثية تتعلق بقبول الطلاب واستخدامهم لـ ChatGPT في التعليم. أجرى حسنين وصبحي [1] مقابلات معمقة لاستكشاف تصورات أصحاب المصلحة الرئيسيين، بما في ذلك الطلاب، حول العوامل التي تدفع لاستخدام ChatGPT في التعليم العالي، وحددوا 12 عاملاً تؤثر على استخدام الطلاب لـ ChatGPT لأسباب أكاديمية. تشمل هذه العوامل سهولة الاستخدام، مساعدة الواجبات المنزلية، الاستجابة السريعة، تحرير اللغة، حل المشكلات، التحضير للامتحانات، تحليل البيانات، دعم البحث، التعلم التكيفي، تعريف المفاهيم، موارد التعلم، وأنشطة التقييم. اعتمد ستريزليكي [4] “نظرية القبول واستخدام التكنولوجيا الموحدة (UTAUT2)” لفحص المتغيرات التي تؤثر على قبول واستخدام طلاب التعليم العالي لـ ChatGPT في بولندا، داعيًا إلى دراسات بحثية مستقبلية في سياقات أخرى من منظور الطلاب. تحاول هذه الدراسة سد هذه الفجوة البحثية في سياق التعليم العالي السعودي واستكشاف العوامل التي تؤثر على استخدام الطلاب لـ ChatGPT في التعلم.
تتناول هذه الدراسة قبول الطلاب واستخدامهم لـ ChatGPT في مؤسسات التعليم العالي في المملكة العربية السعودية، حيث أصبح الطلاب مهتمين باستخدام هذه الأداة لأغراض أكاديمية. سيمكن فهم المتغيرات التي تؤثر على قبول الطلاب واستخدامهم لـ ChatGPT في التعليم قادة الجامعات والباحثين من الحصول على فهم أفضل لهذه العوامل وتنفيذ إدارة متفوقة لدمج الطلاب لأدوات الذكاء الاصطناعي لأغراض أكاديمية. اعتمدت هذه الدراسة إطار UTAUT لفهم أفضل لهذه العوامل بين الطلاب السعوديين في التعليم العالي. لهذا السبب، تستعرض الجزء التالي من المقال إطار UTAUT وتناقش الأساس النظري للدراسة. ثم تشرح الأساليب المعتمدة في الدراسة وتقدم وتناقش النتائج. بعد ذلك، تُقدم الملاحظات النهائية وتُقترح إرشادات للدراسات المستقبلية.

2. مراجعة الأدبيات

2.1. قبول الطلاب لـ ChatGPT ونواياهم السلوكية

يعد نموذج UTAUT خلفية شاملة للتحقيق في قبول واستخدام الابتكارات التكنولوجية في سياقات مختلفة [11-13]. إن تطبيق إطار UTAUT لدمج ChatGPT في البيئات التعليمية يوفر نهجًا منظمًا لفهم نوايا الطلاب السلوكية (BIs) تجاه هذا النموذج اللغوي المتقدم [14،15]. وفقًا لـ Venkatesh وآخرين [13]، تشمل العوامل الرئيسية التي تؤثر على سلوكيات قبول المستخدمين واستخدامهم توقعات الأداء (PE) وتوقعات الجهد (EE) والتأثير الاجتماعي (SI) وظروف التيسير (FCs). يرتبط PE بإدراك الأفراد لقدرات النموذج في تعزيز تجربتهم التعليمية [13]. تؤثر تصورات الطلاب حول قدرات ChatGPT وإمكاناته في المساهمة بشكل إيجابي في تجاربهم التعليمية بشكل مباشر على نواياهم السلوكية [15-20]. علاوة على ذلك، كشف Al-Emran وآخرون [21] أنه عندما يعتقد الطلاب أن ChatGPT يمكن أن يساعدهم بفعالية في تعلم مواضيع معقدة، أو تطوير أفكار إبداعية، أو تعزيز التحصيل الأكاديمي العام، فإنهم يكونون أكثر ميلًا لإظهار نية إيجابية لاستخدام ChatGPT لأغراض تعليمية متنوعة. فيما يتعلق بـ EE، فإنه يشير إلى وجهات نظر المستخدمين بشأن درجة البساطة.
أو تعقيد استخدام التكنولوجيا [13،22]. تلعب دورًا رئيسيًا في تشكيل استعداد الناس ونواياهم للتفاعل مع نموذج اللغة [12]. جادل مينون وشيلبا [15] بأن تصور الطلاب لـ ChatGPT كأداة سهلة الاستخدام، وسهلة الفهم، ومتكاملة بسلاسة في الأنشطة اليومية يؤثر بشكل إيجابي على نواياهم السلوكية. علاوة على ذلك، تشير الدراسات الحديثة، مثل المراجع. لقد كشفت الدراسات أن تصور الطلاب قد تأثر بعدة عناصر مثل سهولة التفاعل مع النموذج، وتصميم واجهة المستخدم، والتجربة العامة للمستخدم. عندما يتوقع الطلاب جهدًا minimal في استخدام ChatGPT، فإن ذلك يزيد من احتمال دمجهم لـ ChatGPT في روتينهم التعليمي اليومي. ].
تأخذ SI في الاعتبار تأثير الأقران والمدرسين في تشكيل مواقف الطلاب تجاه ChatGPT يتجلى ذلك عندما يلاحظ الطلاب المواقف والآراء والسلوكيات لمن حولهم فيما يتعلق بإدماج ChatGPT في التعليم [12،14]. وقد جادلت عدة دراسات، مثل المراجع [15،26،27]، بأن الزملاء الذين ينشرون معلومات إيجابية حول استخدام ChatGPT في الأنشطة التعليمية من المحتمل أن يؤثروا بشكل إيجابي على نية استخدام الطلاب الآخرين. فيما يتعلق بالعوامل الميسرة، يقيم هذا العامل توفر الموارد اللازمة لاستخدامه بشكل فعال [13]. ويشمل عناصر مثل إمكانية الوصول إلى التكنولوجيا، وتوفير الأدوات والتدريب اللازمين، والدعم المؤسسي العام لدمجه. يجب أن يكون لهذا تأثير كبير على نوايا الطلاب لاستخدام ChatGPT عندما يدركون أن البيئة التعليمية توفر الموارد والدعم اللازمين للاستخدام الفعال له. وقد جادل دراسة آل عمران وآخرون أن الفصول المقلوبة تتجاوز مجرد الموارد التكنولوجية الضرورية؛ بل تتطلب إنشاء بنية تحتية داعمة تمكّن المستخدمين من التنقل واستغلال ChatGPT بفعالية، مما يعزز نوايا الطلاب للتفاعل مع التكنولوجيا. لذلك، استنادًا إلى هذه المناقشات، نفترض أن
H1: للتعليم، يؤثر PE بشكل إيجابي كبير على BI لاستخدام ChatGPT.
H2: للتعليم تأثير إيجابي كبير على نية استخدام ChatGPT في التعليم.
H3: لدى SI تأثير إيجابي كبير على BI لاستخدام ChatGPT في التعليم.
H4: FC له تأثير إيجابي كبير على BI لاستخدام ChatGPT في التعليم.

2.2. قبول الطلاب والاستخدام الفعلي لـ ChatGPT

لقد أبرزت عدة دراسات، مثل المراجع [15،21،23،24]، حول الدردشات الآلية (مثل ChatGPT) الدور المحوري لكل من PE و EE في التنبؤ بقبول الطلاب واستخدامهم الفعلي لهذه التكنولوجيا المبتكرة. تتأثر هذه المحددات الحاسمة بعدة متغيرات تتعلق بفائدة الدردشة الآلية المدركة، وسهولة الاستخدام، والبساطة، والتوافق مع احتياجات المستخدمين [17،18]. بالإضافة إلى ذلك، تم تحديد عدة جوانب مثل إنجاز المهام، وإحساس الإنجاز، وزيادة التفاعل كعوامل رئيسية تساهم في PE [19]. في سياق التعليم، جادل مينون وشيلبا [15] بأن PE يعتبر عاملاً بارزاً يؤثر على تصورات الطلاب لاستخدام ChatGPT لأغراض تعليمية. علاوة على ذلك، كشفت الأبحاث الحديثة [1] أن فهم EE تجاه ChatGPT، أي سهولة الاستخدام، والبساطة، والتبني، والاندماج في الروتين اليومي، من المرجح أن يؤدي إلى زيادة الاستخدام الفعلي وزيادة قيمة ChatGPT في السياقات التعليمية. فيما يتعلق بـ SI تجاه ChatGPT في السياقات التعليمية، أظهرت الدراسات الحديثة، مثل المراجع [ لقد لوحظ أن المواقف الإيجابية، والتأييدات، أو الاستخدام النشط للذكاء الاصطناعي القائم على الدردشة ضمن الشبكات الاجتماعية (مثل الزملاء) من المحتمل أن تؤثر بشكل إيجابي على الاستخدام الفعلي للأشخاص لـ ChatGPT. بالإضافة إلى ذلك، فإن التوصيات، والتشجيع، أو التجارب الإيجابية المشتركة من الآخرين لها تأثير مهم على قرار المستخدمين بدمج ChatGPT بنشاط في مهامهم وروتينهم. . الأبحاث السابقة، أي، المراجع. الذي قيّم قابلية اعتماد ChatGPT في التعليم قد أبرز التأثير الكبير للظروف المواتية على اعتماده واستخدامه الفعلي. علاوة على ذلك، أظهرت الأبحاث التي أجراها مينون وشيلبا [15] أن الظروف المواتية، أي جهاز محمول، اتصال إنترنت مستمر، والدعم الفني، هي عوامل حاسمة يمكن أن تؤثر بشكل عميق على اعتماد ChatGPT. ومن ثم، يمكننا أن نفترض أن
H5: للـ PE تأثير إيجابي كبير على استخدام ChatGPT.
H6: للتعليم الإلكتروني تأثير إيجابي كبير على استخدام ChatGPT.
H7: لدى SI تأثير إيجابي كبير على استخدام ChatGPT.
H8: لدى مراكز الاتصال تأثير إيجابي كبير على استخدام ChatGPT.

2.3. الذكاء التجاري واستخدام ChatGPT

ترتبط العلاقة بين الدافع لاستخدام ChatGPT والاستخدام الفعلي بجانب حاسم من جوانب اعتماد التكنولوجيا واستخدامها. عرّف فينكاتيش وآخرون [13] الدافع بأنه استعداد الأفراد ورغبتهم في التفاعل مع تكنولوجيا معينة. وغالبًا ما يُعتبر هذا الدافع مقدمة مباشرة لاستخدام التكنولوجيا [22،32]. العديد من الدراسات، أي المراجع. ] أظهرت أن عندما يعبر الطلاب عن توجه إيجابي نحو ChatGPT، مما يدل على استعدادهم لدمجه في روتينهم أو مهامهم، فإن هناك احتمالًا متزايدًا لتحويل هذه النية إلى استخدام فعلي. علاوة على ذلك، كلما كانت النية لاستخدام ChatGPT أكبر، زادت احتمالية انخراط الطلاب في تفاعل نشط مع المنصة، والبحث عن مساعدتها، ودمجها في أنشطتهم الأكاديمية [13،29]. لذلك، استنادًا إلى هذه المناقشات، نحن مدعوون لافتراض ما يلي:
H9: تؤثر الذكاء التجاري بشكل إيجابي على استخدام ChatGPT.
لا شك أن استعداد الطلاب لاستخدام ChatGPT يلعب دورًا محوريًا في العلاقة بين قبولهم لاستخدام هذه التكنولوجيا المبتكرة واستخدامها الفعلي في التعليم العالي. ومع ذلك، من الجدير بالذكر أن هناك فجوة بحثية تتعلق بالأثر الوسيط للذكاء التجاري على العلاقة بين قبول الطلاب واستخدامهم لـ ChatGPT في التعليم العالي. تحاول هذه الدراسة معالجة هذه الفجوة البحثية وتستند إلى نموذج UTAUT لافتراض أن
H10: تلعب أنظمة المعلومات دور الوسيط بين التجربة الشخصية واستخدام ChatGPT.
H11: تلعب أنظمة المعلومات التجارية دور الوسيط بين الانخراط في العمل واستخدام ChatGPT.
H12: الوسائط التجارية تربط بين الذكاء الاصطناعي واستخدام ChatGPT.
H13: تلعب أنظمة المعلومات التجارية دور الوسيط بين التواصل الفعال واستخدام ChatGPT.
يتم تقديم ملخص لجميع العلاقات في الإطار المفاهيمي للبحث أدناه (الشكل 1).
الشكل 1. النموذج المفاهيمي.

3. الطرق

3.1. قياس وتطوير المقياس

تتكون الاستبانة المعتمدة في هذا البحث من ثلاثة أقسام. يوضح القسم الأول أهداف البحث والإرشادات اللازمة لإكمال الاستبانة. يتم جمع معلومات عن ملف المشاركين في القسم الثاني. أما القسم الثالث من الدراسة فقد شمل التحقيق في جوانب مختلفة، حيث يمثل التقييم 1 “موافق بشدة” ويمثل التقييم 5 “موافق بشدة”. تم اعتماد مقياس ChatGPT UTAUT من Venkatesh وآخرون [13] ويشمل أربعة متغيرات: PE وEE وSI وFC. بالإضافة إلى هذه المتغيرات، يتضمن مقياس ChatGPT TPB أيضًا متغير نية سلوكية تم اعتماده من Ajzen وFishbein [36] بالإضافة إلى متغير نية الاستخدام الذي تم تطويره من دراسات سابقة [22]. للتحقق من اتساق وسهولة استخدام الاستبانة، تم تقييم الاستبانة من قبل أساتذة وقادة التعليم. من أجل الحفاظ على صلاحية محتوى الاستبانة، تم تعديل صياغة بعض العبارات وترتيبها بناءً على ردود المشاركين والعلماء [37].

3.2. عينة البحث وطريقة جمع البيانات

ركزت الأبحاث الحالية على الطلاب الذين يستخدمون ChatGPT في أنشطتهم التعليمية داخل الجامعات العامة في المملكة العربية السعودية. لجمع البيانات اللازمة، تم استخدام طريقة العينة الملائمة، مع استهداف محدد لطلاب كلية الأعمال في جامعة الملك فيصل. بدأت عملية جمع الردود في بداية الفصل الدراسي الأول للعام الدراسي 2023-2024، والذي يتوافق مع سبتمبر 2023، واستمرت لمدة شهر واحد. من بين 600 استبيان تم توزيعه، تم إكمال 520 بنجاح، مما أسفر عن معدل استجابة ملحوظ قدره ، ومن الجدير بالذكر أنه لم تكن هناك بيانات مفقودة. كان حجم العينة 520 استجابة صالحة يتماشى مع معايير نونالي [36] للحفاظ على نسبة 1:10 بين العناصر والعينة. في هذا السياق، نظرًا لوجود 21 عنصرًا في الاستبيان، اعتُبر حجم عينة لا يقل عن 210 مشارك مناسبًا. من بين 520 استجابة صالحة، لوحظ أن من الطلاب الذين تم تحديدهم كذكور، ليصل العدد الإجمالي إلى 285 طالبًا، بينما من الطلاب الذين تم تحديدهم كإناث، بلغ إجماليهم 235 طالبة. وكانت الفئة العمرية التي تشكل غالبية الطلاب، تمثل من المستجيبين، كان من 20 إلى 25. ضمن هذه الفئة العمرية، تم تصنيف الطلاب على أنهم طلاب في السنة الثالثة والرابعة.

3.3. تحليل البيانات

تم تحليل البيانات باستخدام PLS-SEM، وهو خوارزمية قائمة على التباين تُستخدم لتحليل المسارات. يوفر PLS-SEM بديلاً غير تقليدي لنموذج المعادلات الهيكلية القائم على التباين (CB-SEM) [38]. ومعترف به لملاءمته في الأبحاث التي تركز على التنبؤ والاستكشاف، زاد PLS-SEM من أهميته [39]. يعمل دون قيود فرضيات الطبيعية في توزيع العينة وقد أظهر فعالية مع كل من أحجام العينات الصغيرة والكبيرة [38،40]. كانت تفضيلات البحث الاستكشافي والطبيعة القابلة للتكيف للطريقة في التعامل مع أحجام العينات المتنوعة هي العوامل الدافعة وراء اختيار هذا النهج لدراستنا. تم تنفيذ تحليل PLS باستخدام SmartPLS 4 [41]. شمل تقدير النموذج تقنية إعادة العينة (bootstrapping) مع 5000 إعادة عينة باستخدام الوضع A (الوضع الانعكاسي) [40]. علاوة على ذلك، لتخفيف تباين الطريقة الشائعة (CMV)، تم إجراء تحليل باستخدام اختبار عامل هارمان الواحد وفقًا لـ Podsakoff وآخرون [42]. تم إخضاع جميع العناصر الـ 25 لتحليل العوامل الاستكشافية (EFA)، مما أظهر أن المتغير الأول أوضح فقط من إجمالي التباين. وهذا يشير إلى أن CMV لم يكن مصدر قلق شائع في الدراسة الحالية. بالإضافة إلى ذلك، كانت جميع درجات عامل تضخم التباين (VIF) أقل من 0.5، مما يدل على غياب مشاكل التعدد الخطي (الجدول 1).
الجدول 1. ملفات تعريف الطلاب ).
ملف شخصي التردد %
جنس ذكر ٢٨٥ ٥٤.٨
أنثى 235 ٤٥.٢
أقل من 20 عامًا 241 ٤٦.٣
عمر من 20 إلى 25 سنة 267 ٥١.٣
من 26 إلى 30 عامًا 12 ٢.٤
سنة أولى ١١٦ ٢٢.٢
السنة الثانية 123 ٢٣.٧
مستوى الدراسة السنة الثالثة 147 ٢٨.٣
السنة الرابعة ١٣٤ ٢٥.٨
لا 113 21.7

4. نتائج الدراسة

تضمنت تقييم النموذج الخارجي (نموذج القياس) حساب الخصائص النفسية لمقاييس متنوعة باستخدام معايير مثل كرونباخ. موثوقيات التركيب (CRs) ومتوسط التباين المستخرج (AVE). أظهرت جميع عناصر المقياس تحميلات معيارية تبلغ 0.7 أو أعلى، مما يدل على صلاحية تقارب قوية. علاوة على ذلك، كانت جميع قيم كرونباخ وتجاوزت قيم CR الحد الأدنى البالغ 0.7، مما يدل على الاتساق الداخلي للعناصر والبناءات (الجدول 2). كما أن نتائج AVE لجميع البناءات تجاوزت الحد الموصى به البالغ 0.5 الذي اقترحه فورنيل ولارك [43]. وبالتالي، اعتُبرت الصلاحية التقاربية مرضية، حيث وصلت جميع قيم AVE إلى 0.5 وما فوق.
قمنا بتأسيس صلاحية التمييز من خلال نهج فورنيل ولارك [43]، مما يضمن أن الجذر التربيعي لـ AVE لكل بناء يتجاوز العلاقة بين ذلك البناء وجميع الآخرين (انظر الجدول 3). بالإضافة إلى ذلك، قمنا بتقييم صلاحية التمييز باستخدام نسبة الارتباط بين الخصائص المتباينة (HTMT)، وهي طريقة تم التأكد من أنها أكثر قوة من نهج فورنيل ولارك [43]. عندما تتجاوز قيم HTMT 0.9، تظهر مخاوف بشأن صلاحية التمييز. ومع ذلك، ظلت جميع النسب أقل من القيمة المحددة 0.9، مما يؤكد صلاحية التمييز (انظر الجدول 3).
قمنا بتأسيس صلاحية التمييز من خلال نهج فورنيل ولارك [43]، مما يضمن أن الجذر التربيعي للاستخراج المتوسط للتباين (AVE) لكل بناء تجاوز الارتباطات بين ذلك البناء وجميع الآخرين (الجدول 3). بالإضافة إلى ذلك، قمنا بتقييم صلاحية التمييز باستخدام نسبة الارتباطات بين الخصائص المختلفة والخصائص المتشابهة (HTMT)، وهي طريقة تعتبر أكثر قوة من نهج فورنيل ولارك [43]. عندما يكون HTMT
تتجاوز القيم 0.9، مما يثير القلق بشأن صلاحية التمييز. ومع ذلك، كما هو موضح في الجدول 3، تظل جميع النسب أقل من القيمة المحددة 0.9، مما يؤكد صلاحية التمييز.
الجدول 2. أبعاد وخصائص المتغيرات النفسية القياسية.
متغيرات ومكونات المقياس التحميلات VIF
PE: ( )
PE1 تشات جي بي تي أداة قيمة لمشاريعي الأكاديمية 0.861 1.613
PE2 استخدام ChatGPT يزيد من احتمال تحقيق الأهداف المهمة في مساعيك الأكاديمية 0.791 1.710
PE3 “تشات جي بي تي يعزز الإنتاجية في الدراسات الأكاديمية من خلال تسريع إنجاز المهام والمشاريع” 0.789 1.523
PE4 استخدام ChatGPT يمكن أن يرفع من أدائي الأكاديمي 0.734 1.613
EE: ( )
EE1 أجد أنه من السهل تعلم كيفية استخدام ChatGPT 0.910 ٤.٠٣٣
EE2 التواصل مع ChatGPT شفاف وسهل الفهم 0.952 ٤.٨٣٢
EE3 “تشات جي بي تي سهل الاستخدام وبديهي” 0.924 ٤.٢٥٦
EE4 أجد أنه من السهل اكتساب الخبرة في استخدام ChatGPT 0.913 3.679
SI: ( )
SI1 “الأشخاص الذين يلعبون دورًا حاسمًا في حياتي يرون أنه يجب علي استخدام ChatGPT” 0.898 1.806
SI2 “الأشخاص الذين يشكلون سلوكي يوصون باستخدام ChatGPT” 0.830 ٢.٢٨٧
SI3 “أولئك الذين أكن لهم تقديراً عالياً يقترحون أن أستفيد من ChatGPT” 0.885 2.835
FC:
FC1 أنا مجهز بشكل كافٍ بالموارد اللازمة لاستخدام ChatGPT 0.864 3.984
FC2 أنا متمكن في استخدام ChatGPT بفضل المعرفة المكتسبة 0.855 ٤.٠٣٥
FC3 “تشات جي بي تي مناسب للتقنيات التي أستخدمها” 0.957 ٤.١٥٧
FC4 عند مواجهة صعوبات مع ChatGPT، من الممكن الحصول على الدعم والمساعدة من مصادر خارجية 0.953 ٣.٩٨٤
بي 1 لقد قررت الاستمرار في استخدام ChatGPT في الأوقات القادمة 0.900 ٢.٣٦٦
بي 2 أنا ملتزم باستخدام ChatGPT كأداة لدراستي 0.906 ٢.٣٤٣
بي 3 أهدف إلى الاستمرار في استخدام ChatGPT بشكل متكرر 0.783 1.572
الاستخدام الفعلي (AU) ( )
AU1 أعتزم استخدام المعرفة والمهارات التي اكتسبتها من ChatGPT في أنشطتي التعليمية 0.785 1.578
AU2 المعرفة والمهارات التي اكتسبتها من ChatGPT ستكون مفيدة لي في الصف 0.947 ٤.١٣٤
AU3 استخدام ChatGPT ساعد في تحسين أدائي الأكاديمي 0.894 ٤.١٠٩
الجدول 3. الصلاحية التمييزية استنادًا إلى طريقتي فورنيل ولارك وHTMT.
ذكاء الأعمال EE نادي كرة القدم التربية البدنية نعم الاستخدام
ذكاء الأعمال 0.864
EE 0.058 [0.73] 0.925
نادي كرة القدم -0.171 [0.154] 0.450 [0.523] 0.908
التربية البدنية 0.297 [0.354] 0.166 [0.188] -0.027 [0.066] 0.795
نعم 0.319 [0.361] -0.136 [0.159] -0.046 [0.045] -0.293 [0.354] 0.872
الاستخدام 0.801 [0.195] 0.017 [0.044] -0.160 [0.141] 0.348 [0.417] 0.286 [0.312] 0.878
تظهر الأرقام الغامقة الجذر التربيعي لـ AVE؛ وتظهر نسب HTMT بين قوسين.
تشير قيم R 2 المعتمدة على طريقة البوتستراب إلى أن وFC شكلت معًا التباين في BI. علاوة على ذلك، ساهمت PE و EE و SI و FC و BI بشكل جماعي في من التباين في استخدام ChatGPT.
عند تحليل معاملات المسار المعززة (انظر الجدول 4)، لوحظ أن BI أثرت بشكل كبير وإيجابي على PE. ) ، ، ) و SI ( )، مما يوفر الدعم لـ H1 و H2 و H3. من المدهش أن FC أظهر تأثيرًا سلبيًا كبيرًا على BI ( ، )، مما أدى إلى رفض H4. علاوة على ذلك، تشير النتائج من تحليل PLS-SEM إلى وجود ارتباط إيجابي كبير بين PE واستخدام ChatGPT ( )، مما يؤكد H5. بالإضافة إلى ذلك، أشارت النتائج إلى تأثير إيجابي معتدل ذو دلالة على استخدام ChatGPT من قبل SI ( ، )، مما يوفر الدعم لـ H7. ومع ذلك، فإن توقع الجهد ( ، ) وتسهيل الظروف ( لم تُظهر تأثيرًا مباشرًا وملحوظًا على استخدام ChatGPT، مما أدى إلى رفض H6 و H8. أخيرًا، أكدت النتائج على تأثير إيجابي قوي وملحوظ لـ BI على استخدام ChatGPT. وبذلك تدعم H9.
الجدول 4. معامل المسار والمتعلق به و القيم.
طرق معامل المسار الإحصائيات القيم النتائج
PE -> BI [H1]. 0.398 6.346 0.000 مقبول
EE -> BI [H2]. 0.144 ٢.٥٩٦ 0.009 مقبول
SI -> BI [H3]. 0.445 ٧.٠٩٥ 0.000 مقبول
FC -> BI [H4]. -0.204 ٤.٦٣٥ 0.000 مرفوض
استخدام ChatGPT [H5]. 0.141 ٤.٤٨٩ 0.000 مقبول
EE -> استخدام ChatGPT [H6]. -0.042 1.610 0.107 مرفوض
استخدام ChatGPT [H7]. 0.070 ٢.٦٠٣ 0.009 مقبول
FC -> استخدام ChatGPT [H8]. 0.001 0.026 0.979 مرفوض
استخدام BI -> ChatGPT [H9]. 0.789 ٢٧.٣٦٦ 0.000 مقبول
مسارات غير مباشرة محددة
PE -> BI -> استخدام ChatGPT [H10]. 0.314 6.076 0.000 مقبول
EE -> BI -> استخدام ChatGPT [H11]. 0.114 ٢.٥٧٥ 0.010 مقبول
استخدام SI -> BI -> ChatGPT [H12]. 0.352 6.708 0.000 مقبول
FC -> BI -> استخدام ChatGPT [H13]. -0.161 ٤.٥٤٠ 0.000 مقبول
تحليل البيانات المعاد تشكيلها ( أظهرت التأثيرات غير المباشرة المحددة أن BIs تتوسط العلاقة بين PE واستخدام ChatGPT )، استخدام EE و ChatGPT ( )، استخدام SI و ChatGPT ( )، وبين استخدام FC و ChatGPT ( داعمًا H10 وH11 وH12 وH13 (انظر الجدول 4 والشكل 2).
الشكل 2. نموذج البحث الذي تم فحصه.

5. المناقشة والآثار

هدفت هذه الدراسة إلى فهم قبول واستخدام طلاب التعليم العالي لـ ChatGPT لأسباب أكاديمية، مع التركيز بشكل خاص على الطلاب في الجامعات العامة في السعودية. أظهرت نتائج استبيان تم اختباره مسبقًا والذي يعتمد على إطار عمل UTAUT نتائج مثيرة للاهتمام بشأن العوامل الرئيسية لاستخدام ChatGPT لأسباب تعليمية. أكدت النتائج أن أفضل عامل لتحديد استخدام ChatGPT لأسباب أكاديمية بين الطلاب السعوديين هو SI. وذلك لأن الطلاب في السعودية يُصنفون على أنهم ينتمون إلى ثقافة جماعية، مما يعني أن نوايا الأفراد السلوكية تتشكل بشكل كبير من قبل أقرانهم وأفراد عائلتهم وزملائهم. أظهرت النتائج أن الطلاب لديهم نية لاستخدام ChatGPT لأن زملاءهم يرون أنه ينبغي عليهم تجربة هذه الأداة ويوصون باستخدام ChatGPT لأسباب أكاديمية. تتماشى هذه النتيجة مع افتراض UTAUT بأن SI له تأثير كبير على نوايا قبول التكنولوجيا. تتفق هذه النتائج مع الدراسات السابقة. الذي كشف عن تأثير إيجابي كبير للذكاء الاجتماعي على استخدام ChatGPT في التعليم العالي. تؤكد هذه النتيجة أنه عندما ينشر الأقران والزملاء تعليقات مناسبة حول الذكاء الاصطناعي وفوائده لأسباب تعليمية، فإن ذلك يؤثر بشكل إيجابي على نية الطلاب الآخرين في اعتماد ChatGPT للتعلم.
كشفت النتائج أن أفضل مؤشر ثانٍ لنية استخدام ChatGPT لأغراض أكاديمية بين طلاب التعليم العالي في السعودية هو كفاءتهم الشخصية. وجد الطلاب أن ChatGPT أداة قيمة لمتابعة دراستهم الأكاديمية حيث يسمح لهم بالحصول على معلومات مهمة ومطلوبة، ويعزز إنتاجيتهم في الدراسات الأكاديمية، ويرتقي بأدائهم الأكاديمي. تدعم هذه النتيجة نتائج سترزيلكي [4]، الذي وجد أيضًا أن الكفاءة الشخصية هي ثاني أهم محدد لنية الطلاب لاستخدام ChatGPT في التعليم. تدعم هذه النتيجة دراسات سابقة أخرى، مثل المرجع [21]، التي وجدت أيضًا ارتباطًا إيجابيًا كبيرًا بين الكفاءة الشخصية ونية استخدام ChatGPT في التعليم. ظهرت سهولة الاستخدام كالمحدد الثالث لنية استخدام ChatGPT لأغراض أكاديمية. وهذا يعني أنه عندما يجد الطلاب ChatGPT سهل الفهم، وسهل الاستخدام، وسهل التواصل معه، ولا يتطلب جهدًا كبيرًا للاستخدام، فإنهم يظهرون نية إيجابية لاستخدام ChatGPT لأسباب أكاديمية. تدعم هذه النتيجة عمل سترزيلكي [4] ومانون وشيلبا [15]، الذين ذكروا أنه عندما يدرك الطلاب أن ChatGPT سهل الاستخدام، وسهل الفهم، ومتكامل بسلاسة في الأنشطة اليومية، فإن لديهم نية إيجابية لاستخدام ChatGPT في تعليمهم. بالإضافة إلى ذلك، فشلت سهولة الاستخدام في التأثير المباشر الكبير على الاستخدام الفعلي لـ ChatGPT في التعليم. وهذا يعني أنه عندما يدرك الطلاب ChatGPT كأداة سهلة الاستخدام، وسهلة التواصل، ولا تتطلب جهدًا كبيرًا، فإنهم أقل احتمالًا لاستخدام ChatGPT بشكل مباشر، على الرغم من وجود نية إيجابية لاستخدامه (وهو ما يؤثر بدوره على استخدامهم الفعلي لـ ChatGPT في التعلم).
على عكس افتراض UTAUT [13] ونتائج الأبحاث السابقة، مثل المرجع [4]، التي أظهرت عدم وجود تأثير كبير لـ FC على BI لاستخدام ChatGPT في التعليم، أظهرت الدراسة الحالية تأثيرًا سلبيًا كبيرًا لـ FC على BI لاستخدام ChatGPT في التعليم بين الطلاب في السعودية. وذلك لأن الطلاب في الجامعات في السعودية لم يكونوا مزودين بالموارد اللازمة للاستفادة من ChatGPT، ولم يجدوا أنفسهم متمكنين في استخدام ChatGPT، ولم يتلقوا الدعم من جامعتهم فيما يتعلق بذلك. ومع ذلك، عندما حصل الطلاب على الدعم المناسب من مؤسساتهم (فيما يتعلق باستخدام ChatGPT في تعليمهم) أظهروا BI إيجابي لاستخدامه. وبالمثل، وُجد أن FC لم يكن له تأثير كبير على الاستخدام الفعلي لـ ChatGPT. وهذا يتعارض مع إطار عمل UTAUT [13] والنتائج السابقة [4]. تؤكد الدراسة الحالية أنه عندما لم يتلق الطلاب الدعم المناسب من معلميهم ومؤسستهم بشأن ChatGPT، أظهروا BI سلبي لاستخدام ChatGPT ولم يستخدموا ChatGPT في تعليمهم.
فيما يتعلق بالمسارات غير المباشرة، وبما يتماشى مع نموذج UTAUT [13] ونتائج سترزليكي [4]، وُجد أن BI كانت وسيطًا في العلاقة بين PR وSI وEE وFC واستخدام ChatGPT الفعلي في التعليم العالي لأسباب أكاديمية. بشكل أكثر تحديدًا، وُجد أن BI كانت وسيطًا جزئيًا في العلاقة بين PE وSE واستخدام ChatGPT في التعليم. ومع ذلك، كان لها تأثير وساطة كامل على العلاقة بين EE واستخدام ChatGPT لأسباب أكاديمية. علاوة على ذلك، كان لـ BI تأثير وساطة سالب كبير على العلاقة.
بين FC والاستخدام الفعلي لـ ChatGPT في التعليم. مرة أخرى، أدى نقص الموارد المتاحة والدعم المقدم للطلاب إلى تطويرهم لمشاعر سلبية تجاه BI وكان له تأثير غير ملحوظ على استخدامهم لـ ChatGPT في التعليم.
تتضمن نتائج هذه الدراسة العديد من الآثار النظرية والعملية. من الناحية النظرية، أكدت الدراسة أن SI هو أفضل مؤشر لقبول الطلاب واستخدامهم لـ ChatGPT لأغراض أكاديمية. ويرجع ذلك أساسًا إلى التأثيرات الثقافية على الطلاب في السعودية، الذين يتشكل سلوكهم من قبل عائلاتهم وزملائهم. ومع ذلك، أظهر زملاؤهم سلوكًا إيجابيًا تجاه استخدام ChatGPT لأغراض أكاديمية. وشملت المؤشرات الأخرى PE وEE على BI للطلاب لاستخدام ChatGPT في التعلم. ومن المثير للاهتمام، أن FC وُجد أنه يؤثر سلبًا على BI للطلاب لاستخدام ChatGPT لأغراض أكاديمية ولم يكن له تأثير كبير على الاستخدام الفعلي للطلاب لـ ChatGPT لأغراض أكاديمية. هذه النتائج لا تدعم إطار UTAUT ولا الأبحاث السابقة حول استخدام ChatGPT لأغراض أكاديمية [4]. ومع ذلك، تدعو هذه النتائج مؤسسات التعليم العالي إلى تجهيز طلابها بشكل أفضل بالموارد والدعم المطلوب إذا كانوا يرغبون في دمج ChatGPT كأداة لدعم التعلم في التعليم العالي. يجب على مؤسسات التعليم العالي تطوير آلية تشجع الاستخدام المسؤول لأدوات الذكاء الاصطناعي، بما في ذلك ChatGPT، لأغراض أكاديمية بين طلابها. وبالتالي، يجب توفير الموارد والدعم الضروريين، مثل وحدة دعم متخصصة، للطلاب. من المهم أيضًا أن يطلق قادة مؤسسات التعليم العالي حملات توعية لرفع وعي طلابهم حول أوجه القصور في أدوات الذكاء الاصطناعي بشكل عام، بما في ذلك ChatGPT. يجب أن تشمل هذه الحملات كل من الآثار الإيجابية والسلبية لاستخدام أدوات الذكاء الاصطناعي لأغراض أكاديمية. يحتاجون إلى رفع مستوى المساءلة لدى طلابهم فيما يتعلق بالاستخدام المسؤول والأخلاقي لهذه الأدوات [45].

6. القيود واتجاهات البحث المستقبلية

تم إجراء هذه الدراسة على عينة من طلاب الإدارة في مؤسسة تعليم عالي عامة في السعودية. أظهرت النتائج بعض الاكتشافات المثيرة للاهتمام، التي دعمت وتناقضت مع إطار UTAUT. ومع ذلك، لم تشمل عينة الدراسة مشاركين من مؤسسات مختلفة (بما في ذلك مؤسسات القطاع الخاص)، الذين قد تكون لديهم تصورات مختلفة حول دمج أدوات الذكاء الاصطناعي مثل ChatGPT في التعليم. لذلك، يمكن أن تتناول الأبحاث المستقبلية هذه الفرصة البحثية. لم تدرس هذه البحث التأثير المعتدل لخصائص المشاركين مثل الجنس، العمر، التخصص، سنة الدراسة، والخبرة على اعتماد التكنولوجيا في التعليم. يمكن أن تكون هذه مجالًا آخر للتحقيق في الدراسات المستقبلية. يمكن أيضًا إجراء مقارنة بين الطلاب الجامعيين وطلاب الدراسات العليا في الدراسات المستقبلية في دول مختلفة.

7. الاستنتاجات

تستجيب هذه الدراسة للحاجة إلى فحص قبول الطلاب واستخدامهم لـ ChatGPT لأسباب أكاديمية نظرًا للاهتمام المتزايد بهذه التكنولوجيا بين الطلاب منذ إطلاقها في عام 2022. تعزز هذه الدراسة فهمنا للمتغيرات التي تؤثر على قبول الطلاب واستخدامهم لـ ChatGPT لأغراض أكاديمية. تدعم النتائج جزئيًا إطار UTAUT والأبحاث السابقة فيما يتعلق بالتأثير المباشر الكبير لـ PE وSI وEE على BI لاستخدام ChatGPT والتأثير المباشر الكبير لـ PE وSI وBI على الاستخدام الفعلي لـ ChatGPT. ومع ذلك، لم تدعم النتائج الدراسات السابقة فيما يتعلق بالتأثير المباشر لـ FC على كل من BI والاستخدام الفعلي لـ ChatGPT، والذي وُجد أنه سلبي في العلاقة الأولى وغير ذي دلالة في العلاقة الثانية. لذلك، كان هناك وساطة جزئية لـ BI في الرابط بين PE وSI وFC والاستخدام الفعلي لـ ChatGPT في التعليم. ومع ذلك، كانت هناك وساطة كاملة في الرابط بين BI وEE والاستخدام الفعلي لـ ChatGPT في التعليم. تؤكد نتائج الدراسة على الحاجة لدعم وتجهيز الطلاب بالموارد اللازمة لاستخدام ChatGPT بشكل أكثر فعالية لأغراض أكاديمية. من المهم أن تضمن مؤسسات التعليم العالي أن يتعرف طلابها على أوجه القصور في أدوات الذكاء الاصطناعي وآثارها طويلة الأمد
on تعلم الطلاب للحفاظ على نتائج تعلم مستدامة. وبالتالي، يجب عليهم تشجيع الاستخدام المسؤول والأخلاقي لـ ChatGPT لأغراض أكاديمية.
مساهمات المؤلفين: التصور، A.E.E.S.، I.A.E. و A.M.H.; المنهجية، I.A.E.، A.E.E.S. و A.M.H.; البرمجيات، I.A.E.; التحقق، A.E.E.S. و I.A.E.; التحليل الرسمي، I.A.E. و A.E.E.S.; التحقيق A.E.E.S.، A.M.H. و I.A.E.; الموارد، A.E.E.S.; تنسيق البيانات، A.E.E.S. و I.A.E.; الكتابة – إعداد المسودة الأصلية، I.A.E.، A.E.E.S. و A.M.H.; الكتابة – المراجعة والتحرير، A.E.E.S. و I.A.E.; التصور، A.M.H. و I.A.E.; الإشراف، I.A.E.; إدارة المشروع، A.E.E.S.، I.A.E. و A.M.H.; الحصول على التمويل، A.E.E.S. جميع المؤلفين قرأوا ووافقوا على النسخة المنشورة من المخطوطة.
التمويل: تم تمويل هذه البحث من قبل عمادة البحث العلمي، نائب رئيس الدراسات العليا والبحث العلمي، جامعة الملك فيصل، المملكة العربية السعودية، رقم المنحة GRANT5451.
بيان مجلس المراجعة المؤسسية: تم إجراء الدراسة وفقًا لإرشادات إعلان هلسنكي وتمت الموافقة عليها من قبل لجنة الأخلاقيات في عمادة البحث العلمي، جامعة الملك فيصل (رقم المشروع: (GRANT5451، تاريخ الموافقة: 1 مايو 2023).
بيان الموافقة المستنيرة: تم الحصول على الموافقة المستنيرة من جميع المشاركين المعنيين في الدراسة.
بيان توفر البيانات: البيانات متاحة عند الطلب من الباحثين الذين يستوفون معايير الأهلية. يرجى الاتصال بالمؤلف الأول بشكل خاص عبر البريد الإلكتروني.
تعارض المصالح: يعلن المؤلفون عدم وجود أي تعارضات في المصالح.

References

  1. Hasanein, A.M.; Sobaih, A.E.E. Drivers and Consequences of ChatGPT Use in Higher Education: Key Stakeholder Perspectives. Eur. J. Investig. Health Psychol. Educ. 2023, 13, 2599-2614. [CrossRef]
  2. Benuyenah, V. Commentary: ChatGPT use in higher education assessment: Prospects and epistemic threats. J. Res. Innov. Teach. Learn. 2023, 16, 134-135. [CrossRef]
  3. Rasul, T.; Nair, S.; Kalendra, D.; Robin, M.; Santini, F.d.O.; Ladeira, W.J.; Sun, M.; Day, I.; Rather, R.A.; Heathcote, L. The role of ChatGPT in higher education: Benefits, challenges, and future research directions. J. Appl. Learn. Teach. 2023, 6, 41-56.
  4. Strzelecki, A. To use or not to use ChatGPT in higher education? A study of students’ acceptance and use of technology. Interact. Learn. Environ. 2023, 1-14. [CrossRef]
  5. Obaid, O.I.; Ali, A.H.; Yaseen, M.G. Impact of Chat GPT on Scientific Research: Opportunities, Risks, Limitations, and Ethical Issues. Iraqi J. Comput. Sci. Math. 2023, 4, 13-17. [CrossRef]
  6. Rahman, M.M.; Watanobe, Y. ChatGPT for education and research: Opportunities, threats, and strategies. Appl. Sci. 2023, 13, 5783. [CrossRef]
  7. King, M.R. ChatGPT. A conversation on artificial intelligence, chatbots, and plagiarism in higher education. Cell Mol. Bioeng. 2023, 16, 1-2. [CrossRef] [PubMed]
  8. Lund, B.D.; Wang, T.; Mannuru, N.R.; Nie, B.; Shimray, S.; Wang, Z. ChatGPT and a new academic reality: Artificial Intelligencewritten research papers and the ethics of the large language models in scholarly publishing. J. Assoc. Inf. Sci. Technol. 2023, 74, 570-581. [CrossRef]
  9. Neumann, M.; Rauschenberger, M.; Schön, E.M. “We Need to Talk About ChatGPT”: The Future of AI and Higher Education. In Proceedings of the 2023 IEEE/ACM 5th International Workshop on Software Engineering Education for the Next Generation (SEENG), Melbourne, Australia, 16 May 2023.
  10. Alotaibi, N.S.; Alshehri, A.H. Prospers and obstacles in using artificial intelligence in Saudi Arabia higher education institutions-The potential of AI-based learning outcomes. Sustainability 2023, 15, 10723. [CrossRef]
  11. Magsamen-Conrad, K.; Upadhyaya, S.; Joa, C.Y.; Dowd, J. Bridging the divide: Using UTAUT to predict multigenerational tablet adoption practices. Comput. Hum. Behav. 2015, 50, 186-196. [CrossRef] [PubMed]
  12. Venkatesh, V. Adoption and use of AI tools: A research agenda grounded in UTAUT. Ann. Oper. Res. 2022, 308, 641-652. [CrossRef]
  13. Venkatesh, V.; Morris, M.G.; Davis, G.B.; Davis, F.D. User acceptance of information technology: Toward a unified view. MIS Q. 2003, 27, 425-478. [CrossRef]
  14. Shahsavar, Y.; Choudhury, A. User Intentions to Use ChatGPT for Self-Diagnosis and Health-Related Purposes: Cross-sectional Survey Study. JMIR Hum. Factors 2023, 10, e47564. [CrossRef] [PubMed]
  15. Menon, D.; Shilpa, K. “Chatting with ChatGPT”: Analyzing the factors influencing users’ intention to Use the Open AI’s ChatGPT using the UTAUT model. Heliyon 2023, 9, e20962. [CrossRef]
  16. Fagan, M.; Kilmon, C.; Pandey, V. Exploring the adoption of a virtual reality simulation: The role of perceived ease of use, perceived usefulness and personal innovativeness. Campus-Wide Inf. Syst. 2012, 29, 117-127. [CrossRef]
  17. Kasilingam, D.L. Understanding the attitude and intention to use smartphone chatbots for shopping. Technol. Soc. 2020, 62, 101280. [CrossRef]
  18. Brachten, F.; Kissmer, T.; Stieglitz, S. The acceptance of chatbots in an enterprise context—A survey study. Int. J. Inf. Manag. 2021, 60, 102375. [CrossRef]
  19. Mogaji, E.; Balakrishnan, J.; Nwoba, A.C.; Nguyen, N.P. Emerging-market consumers’ interactions with banking chatbots. Telemat. Inform. 2021, 65, 101711. [CrossRef]
  20. Bin-Nashwan, S.A.; Sadallah, M.; Bouteraa, M. Use of ChatGPT in academia: Academic integrity hangs in the balance. Technol. Soc. 2023, 75, 102370. [CrossRef]
  21. Al-Emran, M.; AlQudah, A.A.; Abbasi, G.A.; Al-Sharafi, M.A.; Iranmanesh, M. Determinants of using AI-based chatbots for knowledge sharing: Evidence from PLS-SEM and fuzzy sets (fsQCA). IEEE Trans. Eng. Manag. 2023, 71, 4985-4999. [CrossRef]
  22. Venkatesh, V.; Thong, J.Y.; Xu, X. Consumer acceptance and use of information technology: Extending the unified theory of acceptance and use of technology. MIS Q. 2012, 36, 157-178. [CrossRef]
  23. Foroughi, B.; Senali, M.G.; Iranmanesh, M.; Khanfar, A.; Ghobakhloo, M.; Annamalai, N.; Naghmeh-Abbaspour, B. Determinants of Intention to Use ChatGPT for Educational Purposes: Findings from PLS-SEM and fsQCA. Int. J. Human-Computer Interact. 2023, 1-20. [CrossRef]
  24. Shaengchart, Y.; Bhumpenpein, N.; Kongnakorn, K.; Khwannu, P.; Tiwtakul, A.; Detmee, S. Factors influencing the acceptance of ChatGPT usage among higher education students in Bangkok, Thailand. Adv. Knowl. Exec. 2023, 2, 1-14.
  25. Tiwari, C.K.; Bhat, M.A.; Khan, S.T.; Subramaniam, R.; Khan, M.A.I. What drives students toward ChatGPT? An investigation of the factors influencing adoption and usage of ChatGPT. Interact. Technol. Smart Educ. 2023, ahead-of-print.
  26. Jo, H. Understanding AI tool engagement: A study of ChatGPT usage and word-of-mouth among university students and office workers. Telemat. Inform. 2023, 85, 102067. [CrossRef]
  27. Ma, X.; Huo, Y. Are users willing to embrace ChatGPT? Exploring the factors on the acceptance of chatbots from the perspective of AIDUA framework. Technol. Soc. 2023, 75, 102362. [CrossRef]
  28. Oye, N.D.; A.Iahad, N.; Ab.Rahim, N. The history of UTAUT model and its impact on ICT acceptance and usage by academicians. Educ. Inf. Technol. 2014, 19, 251-270. [CrossRef]
  29. Ivanov, S.; Soliman, M. Game of algorithms: ChatGPT implications for the future of tourism education and research. J. Tour. Futur. 2023, 9, 214-221. [CrossRef]
  30. Chan, C.K.Y.; Zhou, W. Deconstructing Student Perceptions of Generative AI (GenAI) through an Expectancy Value Theory (EVT)-based Instrument. arXiv 2023, arXiv:2305.01186.
  31. Duong, C.D.; Vu, T.N.; Ngo, T.V.N. Applying a modified technology acceptance model to explain higher education students’ usage of ChatGPT: A serial multiple mediation model with knowledge sharing as a moderator. Int. J. Manag. Educ. 2023, 21, 100883. [CrossRef]
  32. Terblanche, N.; Kidd, M. Adoption factors and moderating effects of age and gender that influence the intention to use a non-directive reflective coaching chatbot. SAGE Open 2022, 12, 21582440221096136. [CrossRef]
  33. Anayat, S.; Rasool, G.; Pathania, A. Examining the context-specific reasons and adoption of artificial intelligence-based voice assistants: A behavioural reasoning theory approach. Int. J. Consum. Stud. 2023, 47, 1885-1910. [CrossRef]
  34. Ge, H.; Wu, Y. An Empirical Study of Adoption of ChatGPT for Bug Fixing among Professional Developers. Innov. Technol. Adv. 2023, 1, 21-29. [CrossRef]
  35. Cai, Q.; Lin, Y.; Yu, Z. Factors Influencing Learner Attitudes Towards ChatGPT-Assisted Language Learning in Higher Education. Int. J. Hum.-Comput. Interact. 2023, 1-15. [CrossRef]
  36. Ajzen, I.; Fishbein, M. Attitudes and normative beliefs as factors influencing behavioral intentions. J. Pers. Soc. Psychol. 1972, 21, 1. [CrossRef]
  37. Nunnally, J.C.; Bernstein, I.H. Psychometric Theory, 3rd ed.; McGraw-Hill Series in Psychology; McGraw-Hill: New York, NY, USA, 1994.
  38. Hair, J.F., Jr.; Matthews, L.M.; Matthews, R.L.; Sarstedt, M. PLS-SEM or CB-SEM: Updated Guidelines on Which Method to Use. Int. J. Multivar. Data Anal. 2017, 1, 107. [CrossRef]
  39. Henseler, J.; Ringle, C.M.; Sinkovics, R.R. The Use of Partial Least Squares Path Modeling in International Marketing. In Advances in International Marketing; Sinkovics, R.R., Ghauri, P.N., Eds.; Emerald Group Publishing Limited: Bingley, UK, 2009; Volume 20, pp. 277-319, ISBN 978-1-84855-468-9.
  40. Do Valle, P.O.; Assaker, G. Using Partial Least Squares Structural Equation Modeling in Tourism Research: A Review of Past Research and Recommendations for Future Applications. J. Travel Res. 2016, 55, 695-708. [CrossRef]
  41. Ringle, C.M.; Sarstedt, M.; Mitchell, R.; Gudergan, S.P. Partial Least Squares Structural Equation Modeling in HRM Research. Int. J. Hum. Resour. Manag. 2020, 31, 1617-1643. [CrossRef]
  42. Podsakoff, P.M.; MacKenzie, S.B.; Lee, J.-Y.; Podsakoff, N.P. Common Method Biases in Behavioral Research: A Critical Review of the Literature and Recommended Remedies. J. Appl. Psychol. 2003, 88, 879. [CrossRef]
  43. Fornell, C.; Larcker, D.F. Structural Equation Models with Unobservable Variables and Measurement Error: Algebra and Statistics. J. Mark. Res. 1981, 18, 382-388. [CrossRef]
  44. Sobaih, A.E.E.; Algezawy, M.; Elshaer, I.A. Adopting an Extended Theory of Planned Behaviour to Examine Buying Intention and Behaviour of Nutrition-Labelled Menu for Healthy Food Choices in Quick Service Restaurants: Does the Culture of Consumers Really Matter? Int. J. Environ. Res. Public Health 2023, 20, 4498. [CrossRef]
  45. Sobaih, A.E. Ethical concerns for using artificial intelligence chatbots in research and publication: Evidences from Saudi Arabia. J. Appl. Learn. Teach. 2024, 7, 1-11. [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Journal: European Journal of Investigation in Health Psychology and Education, Volume: 14, Issue: 3
DOI: https://doi.org/10.3390/ejihpe14030047
PMID: https://pubmed.ncbi.nlm.nih.gov/38534908
Publication Date: 2024-03-17

Examining Students’ Acceptance and Use of ChatGPT in Saudi Arabian Higher Education

Abu Elnasr E. Sobaih , Ibrahim A. Elshaer (D) and Ahmed M. Hasanein (D)1 Management Department, College of Business Administration, King Faisal University, Al-Hassa 31982, Saudi Arabia; ielshaer@kfu.edu.sa (I.A.E.); aabdelrazek@kfu.edu.sa (A.M.H.)2 Faculty of Tourism and Hotel Management, Helwan University, Cairo 12612, Egypt Faculty of Tourism and Hotel Management, Suez Canal University, Ismailia 41522, Egypt* Correspondence: asobaih@kfu.edu.sa

Citation: Sobaih, A.E.E.; Elshaer, I.A.; Hasanein, A.M. Examining Students’ Acceptance and Use of ChatGPT in Saudi Arabian Higher Education. Eur. J. Investig. Health Psychol. Educ. 2024, 14, 709-721. https://doi.org/ 10.3390/ejihpe14030047
Academic Editors: María del Mar Simón Márquez and Hyunsu Lee
Received: 3 January 2024
Revised: 10 March 2024
Accepted: 15 March 2024
Published: 17 March 2024
Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Abstract

This study examines students’ acceptance and use of ChatGPT in Saudi Arabian (SA) higher education, where there is growing interest in the use of this tool since its inauguration in 2022. Quantitative research data, through a self-reporting survey drawing on the “Unified Theory of Acceptance and Use of Technology” (UTAUT2), were collected from 520 students in one of the public universities in SA at the start of the first semester of the study year 2023-2024. The findings of structural equation modeling partially supported the UTAUT and previous research in relation to the significant direct effect of performance expectancy (PE), social influence (SI), and effort expectancy (EE) on behavioral intention (BI) on the use of ChatGPT and the significant direct effect of PE, SI, and BI on actual use of ChatGPT. Nonetheless, the results did not support earlier research in relation to the direct relationship between facilitating conditions (FCs) and both BI and actual use of ChatGPT, which was found to be negative in the first relationship and insignificant in the second one. These findings were because of the absence of resources, support, and aid from external sources in relation to the use of ChatGPT. The results showed partial mediation of BI in the link between PE, SI, and FC and actual use of ChatGPT in education and a full mediation in the link of BI between EE and actual use of ChatGPT in education. The findings provide numerous implications for scholars and higher education institutions in SA, which are also of interest to other institutions in similar contexts.

Keywords: ChatGPT; Chatbot; artificial intelligence; technology acceptance; technology use; Saudi Arabian education

1. Introduction

Since the inauguration of Chat Generative Pre-trained Transformer (ChatGPT) in 2022, there has been growing interest among students in using ChatGPT in education [1]. ChatGPT is a chatbot generated by the OpenAI, which assists users in creating human-like text by processing commands and conversations [2]. Studies on the use of ChatGPT in the context of higher education, see for example, refs. [1-4], have shown several benefits to using ChatGPT in education, such as receiving quick and prompt service, and academic and nonacademic support; developing language skills; and using editing services. Previous research has also shown that ChatGPT assists students in their assignments, projects, homework, and research tasks. However, there are some concerns among scholars and higher education leadership about the incorporation of ChatGPT in higher education and its use for supporting teaching and learning [5-8]. Other concerns include generation of biased information, fake citations, and reliability and intellectual property concerns, which raise ethical alarms about content provided by students. Hasanein and Sobaih [1] commented on the destructive influence of ChatGPT on academic integrity and the misuse of information provided by ChatGPT. Furthermore, there are other concerns among many scholars, e.g., refs. [1,9], about the integration of ChatGPT into education; said concerns relate to learning outcomes and students’ skills, particularly critical thinking, which are
expected to be negatively affected. Therefore, it has been argued that caution should be taken by academic institutions to prevent over-dependence by students on using ChatGPT for academic purposes [1].
There have been some attempts by scholars, e.g., refs. [2,6], to understand the adoption of AI, e.g., ChatGPT, in education. Nonetheless, there have been few attempts (despite growing numbers of studies) to understand the variables that drive students’ use of ChatGPT in education [1,4,10]. The study of Alotaibi and Alshehri [10] delved into the potential advantages and obstacles associated with integrating AI-driven learning outcomes within Saudi Arabia’s higher education system. Their findings underscored that AI is still in an early developmental phase in the realm of education, yet its significance for higher education institutions is undeniable. Embracing this transformative technology is essential for addressing forthcoming learning challenges. Equally crucial is ensuring that all Saudi Arabian students develop the requisite technical competencies to engage with and contribute to the field of artificial intelligence in the future.
Strzelecki [4] identified a research gap regarding students’ acceptance and use of ChatGPT in education. Hasanein and Sobaih [1] conducted in-depth interviews to explore key stakeholders’ perceptions, including students, about the drivers of using ChatGPT in higher education, and they identified 12 factors affecting students’ usage of ChatGPT for academic-related reasons. These factors are ease of use, homework assistance, quick response, language editing, problem solving, exam preparation, data analysis, research support, adaptive learning, defining concepts, learning resources, and assessment activities. Strzelecki [4] adopted “Unified Theory of Acceptance and Use of Technology (UTAUT2)” to examine the variables that influence higher education students’ acceptance and use of ChatGPT in Poland, albeit calling for future research studies in other contexts from the students’ perspective. This study attempts to fill this research gap in the Saudi higher education context and explore the factors that affect students’ use of ChatGPT for learning.
This research examines students’ acceptance and use of ChatGPT in Saudi Arabian higher education institutions, where students have become interested in the use of this tool for academic purposes. Understanding the variables that affect students’ acceptance and use of ChatGPT in education will enable leaders of universities and scholars to gain better understanding of these factors and carry out superior management of students’ incorporation of AI tools for academic purposes. This research adopted the UTAUT framework for better understanding these factors among Saudi students in higher education. For this reason, the next part of the article reviews the UTAUT framework and discusses the theoretical foundation of the study. It then explains the methods adopted in the study and presents and discusses the results. Final remarks are then presented and guidelines for further studies are suggested.

2. Literature Review

2.1. Students’ Acceptance of ChatGPT and Behavioral Intentions

UTAUT serves as a comprehensive background for investigating the acceptance and utilization of technological innovations in various contexts [11-13]. Applying the UTAUT framework to integrate ChatGPT in educational settings provides a structured approach to understanding students’ behavioral intentions (BIs) regarding this advanced language model [14,15]. According to Venkatesh et al. [13], the key determinants influencing users’ accepting and using behaviors encompass performance expectancy (PE), effort expectancy (EE), social influence (SI), and facilitating conditions (FCs). PE is related to individuals’ perceptions of the model’s capabilities in enhancing their learning experience [13]. Students’ perceptions of ChatGPT’s capabilities and its potential to contribute positively to their learning experiences directly influence their BIs [15-20]. Furthermore, Al-Emran et al. [21] revealed that when students believe that ChatGPT can effectively assist them in learning complex topics, developing creative ideas, or boosting overall academic achievement, they are more likely to show a positive BI to utilize ChatGPT for various educational purposes. With regard to EE, it refers to users’ perspectives concerning the degree of simplicity
or complexity of using the technology [13,22]. It plays a key part in shaping people’s willingness and intention to engage with the language model [12]. Menon and Shilpa [15] argued that students’ perception of ChatGPT as user-friendly, easily understandable, and seamlessly integrated into daily activities positively influences their BIs. Moreover, recent studies, e.g., refs. [ ] have revealed that students’ perception has been influenced by several elements such as the model’s ease of interaction, user interface design, and overall user experience. When students anticipate minimal effort in using ChatGPT, this increases their likelihood of integrating ChatGPT into their daily educational routine [ ].
SI considers the impact of peers and instructors in shaping students’ attitudes towards ChatGPT . It manifests when students observe the attitudes, opinions, and behaviors of those around them in relation to the incorporation of ChatGPT into education [12,14]. Several studies, e.g., refs. [15,26,27], have argued that colleagues who spread positive information about utilizing ChatGPT for educational activities are likely to positively affect the BI of other students. Regarding FCs, this factor assesses the availability of necessary resources for its effective use [13]. It encompasses elements like the accessibility of the technology, the provision of necessary tools and training, and the overall institutional support for its integration . This should have a significant influence on students’ BIs to use ChatGPT when they recognize that the educational setting offers the necessary resources and support for its effective utilization [1,29]. The study of Al-Emran et al. [21] argued that FCs go beyond simply necessary technological resources; they entail the creation of a supportive infrastructure that enables users to navigate and harness ChatGPT effectively, thereby bolstering students’ BIs to engage with the technology. Therefore, based on these discussions, we hypothesize that
H1: PE has a significant positive effect on BI to use ChatGPT in education.
H2: EE has a significant positive effect on BI to use ChatGPT in education.
H3: SI has a significant positive effect on BI to use ChatGPT in education.
H4: FC has a significant positive effect on BI to use ChatGPT in education.

2.2. Students’ Acceptance and Actual Use of ChatGPT

Several studies, e.g., refs. [15,21,23,24], on chatbots (e.g., ChatGPT) have consistently highlighted the pivotal role of PE and EE in predicting student acceptance and actual use of such innovative technology. These crucial determinants are affected by various variables related to the chatbot’s perceived usefulness, ease of use, simplicity, and compatibility with user needs [17,18]. Additionally, several aspects such as task achievement, sense of achievement, and heightened engagement have been identified as key contributors to PE [19]. In terms of educational context, Menon & Shilpa [15] argued that PE is considered a predominant factor affecting students’ perceptions of using ChatGPT for educational purposes. Moreover, recent research [1] has revealed that understanding EE toward ChatGPT, i.e., user-friendliness, simplicity, adoption, and integration into daily routines, is more likely, leading to increased actual usage and the greater value of ChatGPT in educational contexts. Regarding SI toward ChatGPT in educational contexts, recent studies, e.g., refs. [ ], have observed that positive attitudes, endorsements, or active usage of chatbot-based AI within social networks (e.g., colleagues) are likely to positively influence people’s actual use of ChatGPT. Additionally, recommendations, encouragement, or shared positive experiences from others have an important impact on users’ decision to actively incorporate ChatGPT into their tasks and routines . Previous research, i.e., refs. , that has measured the adoptability of ChatGPT in education has underscored the substantial impact of FCs on its adoption and real usage. Furthermore, the research conducted by Menon and Shilpa [15] has shown that FCs, i.e., a mobile device, consistent internet connection, and technical support, are pivotal factors that can profoundly affect the adoption of ChatGPT. Hence, we could postulate that
H5: PE has a significant positive effect on Use of ChatGPT.
H6: EE has a significant positive effect on Use of ChatGPT.
H7: SI has a significant positive effect on Use of ChatGPT.
H8: FCs have a significant positive effect on Use of ChatGPT.

2.3. BI and Use of ChatGPT

The association between BI to use ChatGPT and actual usage is a critical aspect of the technology’s adoption and utilization. Venkatesh et al. [13] defined BI as the willingness and readiness of individuals to engage with a particular technology. This intention is often considered a direct precursor to use of the technology [22,32]. Numerous studies, i.e., refs. [ ] showed that when students convey a positive BI towards ChatGPT, signifying their readiness to include it in their routines or tasks, there is a heightened probability this intention being translated into actual use. Furthermore, the more substantial the intention to use ChatGPT, the more likely students are to engage in active interaction with the platform, seek its assistance, and integrate it into their academic activities [13,29]. Therefore, based on these discussions, we are prompted to hypothesize the following:
H9: BI positively influences use of ChatGPT.
Undoubtedly, students’ BI to use ChatGPT has a pivotal role in the relationship between their acceptance of using this innovative technology and its actual usage in higher education [ ]. However, it is noteworthy that there is a research gap concerning the mediating impact of BI on the link between students’ acceptance and usage of ChatGPT in higher education. This research attempts to address this gap in research and draws on the UTAUT to hypothesize that
H10: BIs mediate the link between PE and Use of ChatGPT.
H11: BIs mediate the link between EE and Use of ChatGPT.
H12: BIs mediate the link between SI and Use of ChatGPT.
H13: BIs mediate the link between FC and Use of ChatGPT.
A summary of all the relationships in the research conceptual framework is presented below (Figure 1).
Figure 1. The conceptual model.

3. Methods

3.1. Measures and Scale Development

The survey adopted in this research has three sections. The first section outlines the research objectives and guidelines needed for completing the survey. Throughout the second section, information about participants’ profile is gathered. The third section of the study involved investigating various aspects, with a rating of 1 representing “strongly disagree” and a rating of 5 indicating “strongly agree”. The ChatGPT UTAUT scale was adopted from Venkatesh et al. [13] and includes four variables: PE, EE, SI, and FC. In addition to these variables, the ChatGPT TPB scale also includes a behavioral intention variable adopted from Ajzen and Fishbein [36] as well as an intent-to-use variable developed from previous studies [22]. To verify the consistency and ease of use of the survey, the survey was subjected to evaluation by professors and education leaders. In order to maintain the validity of the survey’s content, certain statements’ wording was modified and arranged based upon the responses of participants and scholars [37].

3.2. Research Sample and Data Collection Method

The current research focused on students who utilize ChatGPT in their educational activities within public universities in Saudi Arabia. To gather the necessary data, a convenience sampling method was used, specifically targeting students from the School of Business at King Faisal University. The gathering of responses commenced at the start of first semester for study year 2023-2024, which corresponds to September 2023, and continued for a duration of one month. Among 600 surveys distributed, a total of 520 were successfully completed, resulting in a remarkable response rate of , and notably, there were no missing data. The sample size of 520 valid responses adhered to Nunnally’s criteria [36] for maintaining a 1:10 item-to-sample ratio. In this context, since there were 21 items in the survey, a sample size of at least 210 participants was considered appropriate. Among the 520 valid responses, it was observed that of the students identified as male, amounting to a total of 285 students, while of the students identified as female, totaling 235 students. The age group that constituted the majority of students, accounting for of the respondents, was from 20 to 25 . Within this age group, of the students were classified as junior and senior students.

3.3. Data Analysis

The data were subjected to analysis utilizing PLS-SEM, a variance-based algorithm employed for analyzing paths. PLS-SEM offers an unconventional alternative to the classical covariance-based SEM (CB-SEM) [38]. Acknowledged for its applicability in prediction-focused and exploratory research, PLS-SEM has risen in prominence [39]. It operates without the constraint of normality assumptions in the sample spreading and has demonstrated efficacy with both small and large sample sizes [38,40]. The preference for exploratory research and the adaptable nature of the method in handling diverse sample sizes were the driving factors behind choosing this approach for our study. The PLS analysis was executed utilizing SmartPLS 4 [41]. Model estimation involved a bootstrapping technique with 5000 resamples using Mode A (reflective mode) [40]. Furthermore, to mitigate common-method variance (CMV), following Podsakoff et al. [42], an analysis was performed using Harman’s one-factor test. All 25 items were subjected to an exploratory factor analysis (EFA), demonstrating that the first variable clarified only of the total variance. This suggests that CMV was not a prevalent concern in the present study. Additionally, all variance inflation factor (VIF) scores were below 0.5 , indicating the absence of multicollinearity issues (Table 1).
Table 1. Students’ profiles ( ).
Profile Freq. %
Gender Male 285 54.8
Female 235 45.2
Less than 20 years 241 46.3
Age 20 to 25 years 267 51.3
26 to 30 years 12 2.4
Freshman (year one) 116 22.2
Sophomore (year two) 123 23.7
Study level Junior (year three) 147 28.3
Senior (year four) 134 25.8
No 113 21.7

4. Results of the Study

The assessment of the outer model (measurement model) included calculation of the psychometric properties of diverse scales using criteria such as Cronbach’s , composite reliabilities (CRs), and average variance extracted (AVE). All scale items demonstrated standardized loadings of 0.7 and higher, signifying robust convergent validity. Furthermore, all values of Cronbach’s and CRs exceeded the minimum threshold of 0.7 , indicating the internal consistency of items and constructs (Table 2). The AVE results for all constructs surpassed the recommended threshold of 0.5 proposed by Fornell & Larcker [43]. Consequently, convergent validity was considered satisfactory, with all AVEs reaching 0.5 and beyond.
We established discriminant validity through Fornell and Larcker’s [43] approach, ensuring that the square root of the AVE for each construct exceeded the association between that construct and all others (see Table 3). Additionally, we evaluated discriminant validity using the Heterotrait-Monotrait (HTMT) ratio of association, a method confirmed to be more robust than Fornell and Larcker’s [43]. When HTMT values exceed 0.9, concerns about discriminant validity arise. However, all ratios remained below the indicated value of 0.9 , confirming discriminant validity (see Table 3).
We established discriminant validity through Fornell and Larcker’s [43] approach, ensuring that the square root of the Average Variance Extracted (AVE) for each construct exceeded the correlations between that construct and all others (Table 3). Additionally, we evaluated discriminant validity using the Heterotrait-Monotrait (HTMT) ratio of correlations, a method considered more robust than Fornell and Larcker’s [43]. When HTMT
values exceed 0.9, concerns about discriminant validity arise. However, as shown in Table 3, all ratios remain below the specified value of 0.9 , confirming discriminant validity.
Table 2. Dimensions and variables’ psychometric properties.
Scale Variables and Items Loadings VIF
PE: ( )
PE1 “ChatGPT is a valuable tool for my academic pursuits” 0.861 1.613
PE2 “Utilizing ChatGPT improves the probability of attaining important objectives in your academic pursuits” 0.791 1.710
PE3 “ChatGPT enhances productivity in academic studies by expediting the completion of tasks and projects” 0.789 1.523
PE4 “Using ChatGPT can elevate my academic performance” 0.734 1.613
EE: ( )
EE1 “I find it easy to learn how to use ChatGPT” 0.910 4.033
EE2 “Communication with ChatGPT is transparent and easy to comprehend” 0.952 4.832
EE3 “ChatGPT is user-friendly and intuitive” 0.924 4.256
EE4 “I find it effortless to acquire expertise in using ChatGPT” 0.913 3.679
SI: ( )
SI1 “People who play a crucial role in my life are of the opinion that I should utilize ChatGPT” 0.898 1.806
SI2 “People who shape my behavior recommend the utilization of ChatGPT” 0.830 2.287
SI3 “Those whose opinions I hold in high esteem suggest that I make use of ChatGPT” 0.885 2.835
FC:
FC1 “I am adequately equipped with the necessary resources to make use of ChatGPT” 0.864 3.984
FC2 “I am proficient in utilizing ChatGPT due to acquired knowledge” 0.855 4.035
FC3 “ChatGPT is suitable for the technologies I utilize” 0.957 4.157
FC4 “When facing difficulties with ChatGPT, it is possible to receive support and aid from external sources” 0.953 3.984
BI1 “I have decided to continue using ChatGPT in the times ahead” 0.900 2.366
BI2 “I am dedicated to utilizing ChatGPT as a tool for my studies” 0.906 2.343
BI3 “I aim to continue using ChatGPT on a frequent basis” 0.783 1.572
Actual Use (AU) ( )
AU1 “I intend to use the knowledge and skills I acquired from the ChatGPT in my educational activities” 0.785 1.578
AU2 “The knowledge and skills I acquired from the ChatGPT will be useful to me in class” 0.947 4.134
AU3 “Using ChatGPT has helped to improve my academic performance” 0.894 4.109
Table 3. Discriminant validity based on the Fornell and Larcker and HTMT methods.
BI EE FC PE SI Usage
BI 0.864
EE 0.058 [0.73] 0.925
FC -0.171 [0.154] 0.450 [0.523] 0.908
PE 0.297 [0.354] 0.166 [0.188] -0.027 [0.066] 0.795
SI 0.319 [0.361] -0.136 [0.159] -0.046 [0.045] -0.293 [0.354] 0.872
Usage 0.801 [0.195] 0.017 [0.044] -0.160 [0.141] 0.348 [0.417] 0.286 [0.312] 0.878
Bold figures show the square root of AVE,; HTMT ratios are shown in brackets.
The bootstrapped R 2 values indicated that and FC collectively accounted for of the variance in BI. Moreover, PE, EE, SI, FC, and BI collectively contributed to of the variance in use of ChatGPT.
Analyzing the bootstrapped path coefficients (refer to Table 4), it was observed that BI significantly and positively affected PE ( ), , ) and SI ( ), thereby providing support for H1, H2, and H3. Surprisingly, FC exhibited a significant negative impact on BI ( , ), leading to the rejection of H4. Moreover, the outcomes from the PLS-SEM analysis suggest a significant positive association between PE and the utilization of ChatGPT ( ), thus confirming H5. Additionally, the results indicated a modest significant positive effect of SI on ChatGPT usage ( , ), providing support for H7. Nevertheless, effort expectancy ( , ) and facilitating conditions ( ) did not demonstrate a direct and significant impact on ChatGPT usage, resulting in the rejection of H6 and H8. Finally, the findings underscored a robust positive and significant effect of BI on ChatGPT usage ( ), thereby supporting H9.
Table 4. Path coefficient and related and values.
Paths Path Coefficient Statistics Values Results
PE -> BI [H1]. 0.398 6.346 0.000 Accepted
EE -> BI [H2]. 0.144 2.596 0.009 Accepted
SI -> BI [H3]. 0.445 7.095 0.000 Accepted
FC -> BI [H4]. -0.204 4.635 0.000 Rejected
PE -> ChatGPT usage [H5]. 0.141 4.489 0.000 Accepted
EE -> ChatGPT usage [H6]. -0.042 1.610 0.107 Rejected
SI -> ChatGPT usage [H7]. 0.070 2.603 0.009 Accepted
FC -> ChatGPT usage [H8]. 0.001 0.026 0.979 Rejected
BI -> ChatGPT usage [H9]. 0.789 27.366 0.000 Accepted
Specific indirect paths
PE -> BI -> ChatGPT usage [H10]. 0.314 6.076 0.000 Accepted
EE -> BI -> ChatGPT usage [H11]. 0.114 2.575 0.010 Accepted
SI -> BI -> ChatGPT usage [H12]. 0.352 6.708 0.000 Accepted
FC -> BI -> ChatGPT usage [H13]. -0.161 4.540 0.000 Accepted
Analyzing the bootstrapped ( ) specific indirect effects showed that BIs mediate the link between PE and ChatGPT usage ( ), EE and ChatGPT usage ( ), SI and ChatGPT usage ( ), and between FC and ChatGPT usage ( ), supporting H10, H11,H12, and H13 (See Table 4 and Figure 2).
Figure 2. The examined research model.

5. Discussion and Implications

This study aimed to understand higher education students’ acceptance and usage of ChatGPT for academic reasons, with a particular focus on students in public universities in SA. The results of a pre-tested survey drawing on the UTAUT framework showed interesting findings regarding the key determinants of ChatGPT usage for educational reasons. The results confirmed that the best determinant of ChatGPT use for academic reasons among SA students is SI. This is because students in SA are categorized as having a collective culture, which implies that the behavioral intentions of people are strongly shaped by their peers, family members, and colleagues [44]. The results showed that students have the intent to use ChatGPT because their colleagues are of the opinion that they should try this tool and recommend that they utilize ChatGPT for academic reasons. This finding is in line with the UTAUT’s assumption that SI has a significant impact on the BIs of technology acceptance [13]. These results are in agreement with previous studies that revealed a positive significant influence of SI on ChatGPT use in higher education. This finding confirms that when peers and colleagues spread appropriate feedback about AI and its benefits for educational reasons, this positively affects other students’ BI to adopt ChatGPT for learning.
The findings revealed that the second-best predictor of BI to use ChatGPT for academic purposes among higher education students in SA is their PE. Students found ChatGPT a valuable tool for their academic pursuits as it allows them attain important and required information, enhances their productivity in academic studies, and elevates their academic performance. This outcome supports the results of Strzelecki [4], who also found that PE is the second-top determinant of students’ BI to use ChatGPT in education. This result is supported by other earlier studies, e.g., ref. [21], that also found a significant positive association between PE and BI to use ChatGPT in education. EE emerged as the third determinant of BI to use ChatGPT for academic purposes. This means that when students found ChatGPT easy to comprehend, easy to use, and easy to communicate with and it did not require much effort to use, they exhibited positive BI to use ChatGPT for academic reasons. This finding supports the work of Strzelecki [4] and Menon and Shilpa [15], who stated that when students perceive ChatGPT to be user-friendly, easily understandable, and seamlessly integrated into daily activities, they have a positive BI to use ChatGPT in their education. In addition, EE failed to have a direct significant influence on actual usage of ChatGPT in education. This means that when students perceive ChatGPT as user-friendly, easy to communicate with, and not requiring much effort, they are less likely to use ChatGPT directly, despite having positive BI to use it (which in turn affects their actual usage of ChatGPT for learning).
Unlike the UTAUT assumption [13] and previous research findings, e.g., ref. [4], which showed no significant influence of FC on BI to use ChatGPT in education, the current study showed negative significant influence of FC on BI to use ChatGPT in education among SA students. This is because students in universities in SA were not equipped with the necessary resources to make use of ChatGPT, they did not find themselves proficient in utilizing ChatGPT, and they did not receive support from their university in relation to it. However, when students had proper support from their institutions (as related to ChatGPT use in their educations) they exhibited positive BI to use it. Similarly, FC was found to have no significant influence on the actual use of ChatGPT. This contradicts the UTAUT framework [13] and previous findings [4]. The current study confirms that when students did not receive proper support from their tutors and institution regarding ChatGPT, they exhibited negative BI to use ChatGPT and would not use ChatGPT in their education.
With regard to indirect paths, aligning with UTAUT model [13] and the results of Strzelecki [4], BI was found to mediate the link between PR, SI, EE, FC and actual ChatGPT usage in higher education for academic reasons. More specifically, BI was found to partially mediate the relationship between PE, SE, and ChatGPT usage in education. Nonetheless, it had a full mediating effect on the link between EE and the use of ChatGPT for academic reasons. Furthermore, BI had a significant negative mediating effect on the relationship
between FC and actual use of ChatGPT in education. Again, lack of available resources and support given to students made them develop negative BI and had an insignificant effect on their ChatGPT usage in education.
The findings of this study have several theoretical and practical implications. Theoretically, the study confirmed that SI is best predictor of students’ acceptance and use of ChatGPT for academic purposes. This is mainly due to the cultural impacts on students in SA, whose behavior is shaped by their families and colleagues. Notwithstanding this, their colleagues showed positive behavior towards the use of ChatGPT for academic purposes. Other predictors included PE and EE on the BI of students to use ChatGPT for learning. Interestingly, FC was found to negatively affect students’ BI to use ChatGPT for academic purposes and had no significant effect on students’ actual use of ChatGPT for academic purposes. These results support neither the UTAUT framework nor previous research on the use of ChatGPT for academic purposes [4]. However, these results call for higher education institutions to better equip their students with the required resources and support if they would like to integrate ChatGPT as a tool for supporting learning in higher education. Higher education institutions should develop a mechanism that encourages the responsible use of AI tools, including ChatGPT, for academic purposes among their students. Hence, necessary resources and support, such as a specialized support unit, should be made available to students. It is also important that leaders of higher education institutions launch awareness campaigns to raise the awareness of their students about the shortcomings of AI tools in general, including ChatGPT. This should include both the positive and negative impacts of using AI tools for academic purposes. They need to raise the accountability of their students in relation to the responsible and ethical use of these tools [45].

6. Limitations and Future Research Directions

This study was undertaken on a sample of management students in a public higher education institution in SA. The results showed some interesting findings, which supported and contradicted the UTAUT framework. However, the sample of study did not include participants from different institutions (including private sector institutions), who may have different perceptions of integrating AI tools like ChatGPT into education. Therefore, future research could address this research opportunity. This research did not study the moderating influence of participants’ characteristics such as gender, age, specialization, study year, and experience on the adoption of technology in education. This could be another venue for investigation in future studies. A comparison between undergraduate and postgraduate students could also be undertaken in future studies in different countries.

7. Conclusions

This study responds to the need to examine students’ acceptance and usage of ChatGPT for academic reasons due to the growing interest in this technology among students since its inauguration in 2022. This study advances our understanding on the variables that affect students’ acceptance and usage of ChatGPT for academic purposes. The results partially support the UTAUT framework and previous research in relation to the significant direct impact of PE, SI and EE on BI to use ChatGPT and the significant direct effect of PE, SI and BI on actual use of ChatGPT. Nonetheless, the results did not support previous studies in relation to the direct effect of FC on both BI and actual use of ChatGPT, which was found to be negative in the first relationship and insignificant in the second relationship. Therefore, there was partial mediation of BI in the link between PE, SI, and FC and actual use of ChatGPT in education. However, there was full mediation in the link between BI and EE and actual use of ChatGPT in education. The results of study confirm a need to support and equip students with the necessary resources to make more effective use of ChatGPT for academic purposes. It is important that higher education institutions ensure that their students recognize the shortcomings of AI tools and their long-term impacts
on students’ learning to maintain sustainable learning outcomes. Hence, they should encourage responsible and ethical use of ChatGPT for academic purposes.
Author Contributions: Conceptualization, A.E.E.S., I.A.E. and A.M.H.; methodology, I.A.E., A.E.E.S. and A.M.H.; software, I.A.E.; validation, A.E.E.S. and I.A.E.; formal analysis, I.A.E. and A.E.E.S.; investigation A.E.E.S., A.M.H. and I.A.E.; resources, A.E.E.S.; data curation, A.E.E.S. and I.A.E.; writing—original draft preparation, I.A.E., A.E.E.S. and A.M.H.; writing—review and editing, A.E.E.S. and I.A.E.; visualization, A.M.H. and I.A.E.; supervision, I.A.E.; project administration, A.E.E.S., I.A.E. and A.M.H.; funding acquisition, A.E.E.S. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Deanship of Scientific Research, Vice Presidency for Graduate Studies and Scientific Research, King Faisal University, Saudi Arabia, grant number GRANT5451.
Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Deanship of Scientific Research Ethical Committee, King Faisal University (project number: (GRANT5451, date of approval: 1 May 2023).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: Data are available upon request from researchers who meet the eligibility criteria. Kindly contact the first author privately through e-mail.
Conflicts of Interest: The authors declare no conflicts of interest.

References

  1. Hasanein, A.M.; Sobaih, A.E.E. Drivers and Consequences of ChatGPT Use in Higher Education: Key Stakeholder Perspectives. Eur. J. Investig. Health Psychol. Educ. 2023, 13, 2599-2614. [CrossRef]
  2. Benuyenah, V. Commentary: ChatGPT use in higher education assessment: Prospects and epistemic threats. J. Res. Innov. Teach. Learn. 2023, 16, 134-135. [CrossRef]
  3. Rasul, T.; Nair, S.; Kalendra, D.; Robin, M.; Santini, F.d.O.; Ladeira, W.J.; Sun, M.; Day, I.; Rather, R.A.; Heathcote, L. The role of ChatGPT in higher education: Benefits, challenges, and future research directions. J. Appl. Learn. Teach. 2023, 6, 41-56.
  4. Strzelecki, A. To use or not to use ChatGPT in higher education? A study of students’ acceptance and use of technology. Interact. Learn. Environ. 2023, 1-14. [CrossRef]
  5. Obaid, O.I.; Ali, A.H.; Yaseen, M.G. Impact of Chat GPT on Scientific Research: Opportunities, Risks, Limitations, and Ethical Issues. Iraqi J. Comput. Sci. Math. 2023, 4, 13-17. [CrossRef]
  6. Rahman, M.M.; Watanobe, Y. ChatGPT for education and research: Opportunities, threats, and strategies. Appl. Sci. 2023, 13, 5783. [CrossRef]
  7. King, M.R. ChatGPT. A conversation on artificial intelligence, chatbots, and plagiarism in higher education. Cell Mol. Bioeng. 2023, 16, 1-2. [CrossRef] [PubMed]
  8. Lund, B.D.; Wang, T.; Mannuru, N.R.; Nie, B.; Shimray, S.; Wang, Z. ChatGPT and a new academic reality: Artificial Intelligencewritten research papers and the ethics of the large language models in scholarly publishing. J. Assoc. Inf. Sci. Technol. 2023, 74, 570-581. [CrossRef]
  9. Neumann, M.; Rauschenberger, M.; Schön, E.M. “We Need to Talk About ChatGPT”: The Future of AI and Higher Education. In Proceedings of the 2023 IEEE/ACM 5th International Workshop on Software Engineering Education for the Next Generation (SEENG), Melbourne, Australia, 16 May 2023.
  10. Alotaibi, N.S.; Alshehri, A.H. Prospers and obstacles in using artificial intelligence in Saudi Arabia higher education institutions-The potential of AI-based learning outcomes. Sustainability 2023, 15, 10723. [CrossRef]
  11. Magsamen-Conrad, K.; Upadhyaya, S.; Joa, C.Y.; Dowd, J. Bridging the divide: Using UTAUT to predict multigenerational tablet adoption practices. Comput. Hum. Behav. 2015, 50, 186-196. [CrossRef] [PubMed]
  12. Venkatesh, V. Adoption and use of AI tools: A research agenda grounded in UTAUT. Ann. Oper. Res. 2022, 308, 641-652. [CrossRef]
  13. Venkatesh, V.; Morris, M.G.; Davis, G.B.; Davis, F.D. User acceptance of information technology: Toward a unified view. MIS Q. 2003, 27, 425-478. [CrossRef]
  14. Shahsavar, Y.; Choudhury, A. User Intentions to Use ChatGPT for Self-Diagnosis and Health-Related Purposes: Cross-sectional Survey Study. JMIR Hum. Factors 2023, 10, e47564. [CrossRef] [PubMed]
  15. Menon, D.; Shilpa, K. “Chatting with ChatGPT”: Analyzing the factors influencing users’ intention to Use the Open AI’s ChatGPT using the UTAUT model. Heliyon 2023, 9, e20962. [CrossRef]
  16. Fagan, M.; Kilmon, C.; Pandey, V. Exploring the adoption of a virtual reality simulation: The role of perceived ease of use, perceived usefulness and personal innovativeness. Campus-Wide Inf. Syst. 2012, 29, 117-127. [CrossRef]
  17. Kasilingam, D.L. Understanding the attitude and intention to use smartphone chatbots for shopping. Technol. Soc. 2020, 62, 101280. [CrossRef]
  18. Brachten, F.; Kissmer, T.; Stieglitz, S. The acceptance of chatbots in an enterprise context—A survey study. Int. J. Inf. Manag. 2021, 60, 102375. [CrossRef]
  19. Mogaji, E.; Balakrishnan, J.; Nwoba, A.C.; Nguyen, N.P. Emerging-market consumers’ interactions with banking chatbots. Telemat. Inform. 2021, 65, 101711. [CrossRef]
  20. Bin-Nashwan, S.A.; Sadallah, M.; Bouteraa, M. Use of ChatGPT in academia: Academic integrity hangs in the balance. Technol. Soc. 2023, 75, 102370. [CrossRef]
  21. Al-Emran, M.; AlQudah, A.A.; Abbasi, G.A.; Al-Sharafi, M.A.; Iranmanesh, M. Determinants of using AI-based chatbots for knowledge sharing: Evidence from PLS-SEM and fuzzy sets (fsQCA). IEEE Trans. Eng. Manag. 2023, 71, 4985-4999. [CrossRef]
  22. Venkatesh, V.; Thong, J.Y.; Xu, X. Consumer acceptance and use of information technology: Extending the unified theory of acceptance and use of technology. MIS Q. 2012, 36, 157-178. [CrossRef]
  23. Foroughi, B.; Senali, M.G.; Iranmanesh, M.; Khanfar, A.; Ghobakhloo, M.; Annamalai, N.; Naghmeh-Abbaspour, B. Determinants of Intention to Use ChatGPT for Educational Purposes: Findings from PLS-SEM and fsQCA. Int. J. Human-Computer Interact. 2023, 1-20. [CrossRef]
  24. Shaengchart, Y.; Bhumpenpein, N.; Kongnakorn, K.; Khwannu, P.; Tiwtakul, A.; Detmee, S. Factors influencing the acceptance of ChatGPT usage among higher education students in Bangkok, Thailand. Adv. Knowl. Exec. 2023, 2, 1-14.
  25. Tiwari, C.K.; Bhat, M.A.; Khan, S.T.; Subramaniam, R.; Khan, M.A.I. What drives students toward ChatGPT? An investigation of the factors influencing adoption and usage of ChatGPT. Interact. Technol. Smart Educ. 2023, ahead-of-print.
  26. Jo, H. Understanding AI tool engagement: A study of ChatGPT usage and word-of-mouth among university students and office workers. Telemat. Inform. 2023, 85, 102067. [CrossRef]
  27. Ma, X.; Huo, Y. Are users willing to embrace ChatGPT? Exploring the factors on the acceptance of chatbots from the perspective of AIDUA framework. Technol. Soc. 2023, 75, 102362. [CrossRef]
  28. Oye, N.D.; A.Iahad, N.; Ab.Rahim, N. The history of UTAUT model and its impact on ICT acceptance and usage by academicians. Educ. Inf. Technol. 2014, 19, 251-270. [CrossRef]
  29. Ivanov, S.; Soliman, M. Game of algorithms: ChatGPT implications for the future of tourism education and research. J. Tour. Futur. 2023, 9, 214-221. [CrossRef]
  30. Chan, C.K.Y.; Zhou, W. Deconstructing Student Perceptions of Generative AI (GenAI) through an Expectancy Value Theory (EVT)-based Instrument. arXiv 2023, arXiv:2305.01186.
  31. Duong, C.D.; Vu, T.N.; Ngo, T.V.N. Applying a modified technology acceptance model to explain higher education students’ usage of ChatGPT: A serial multiple mediation model with knowledge sharing as a moderator. Int. J. Manag. Educ. 2023, 21, 100883. [CrossRef]
  32. Terblanche, N.; Kidd, M. Adoption factors and moderating effects of age and gender that influence the intention to use a non-directive reflective coaching chatbot. SAGE Open 2022, 12, 21582440221096136. [CrossRef]
  33. Anayat, S.; Rasool, G.; Pathania, A. Examining the context-specific reasons and adoption of artificial intelligence-based voice assistants: A behavioural reasoning theory approach. Int. J. Consum. Stud. 2023, 47, 1885-1910. [CrossRef]
  34. Ge, H.; Wu, Y. An Empirical Study of Adoption of ChatGPT for Bug Fixing among Professional Developers. Innov. Technol. Adv. 2023, 1, 21-29. [CrossRef]
  35. Cai, Q.; Lin, Y.; Yu, Z. Factors Influencing Learner Attitudes Towards ChatGPT-Assisted Language Learning in Higher Education. Int. J. Hum.-Comput. Interact. 2023, 1-15. [CrossRef]
  36. Ajzen, I.; Fishbein, M. Attitudes and normative beliefs as factors influencing behavioral intentions. J. Pers. Soc. Psychol. 1972, 21, 1. [CrossRef]
  37. Nunnally, J.C.; Bernstein, I.H. Psychometric Theory, 3rd ed.; McGraw-Hill Series in Psychology; McGraw-Hill: New York, NY, USA, 1994.
  38. Hair, J.F., Jr.; Matthews, L.M.; Matthews, R.L.; Sarstedt, M. PLS-SEM or CB-SEM: Updated Guidelines on Which Method to Use. Int. J. Multivar. Data Anal. 2017, 1, 107. [CrossRef]
  39. Henseler, J.; Ringle, C.M.; Sinkovics, R.R. The Use of Partial Least Squares Path Modeling in International Marketing. In Advances in International Marketing; Sinkovics, R.R., Ghauri, P.N., Eds.; Emerald Group Publishing Limited: Bingley, UK, 2009; Volume 20, pp. 277-319, ISBN 978-1-84855-468-9.
  40. Do Valle, P.O.; Assaker, G. Using Partial Least Squares Structural Equation Modeling in Tourism Research: A Review of Past Research and Recommendations for Future Applications. J. Travel Res. 2016, 55, 695-708. [CrossRef]
  41. Ringle, C.M.; Sarstedt, M.; Mitchell, R.; Gudergan, S.P. Partial Least Squares Structural Equation Modeling in HRM Research. Int. J. Hum. Resour. Manag. 2020, 31, 1617-1643. [CrossRef]
  42. Podsakoff, P.M.; MacKenzie, S.B.; Lee, J.-Y.; Podsakoff, N.P. Common Method Biases in Behavioral Research: A Critical Review of the Literature and Recommended Remedies. J. Appl. Psychol. 2003, 88, 879. [CrossRef]
  43. Fornell, C.; Larcker, D.F. Structural Equation Models with Unobservable Variables and Measurement Error: Algebra and Statistics. J. Mark. Res. 1981, 18, 382-388. [CrossRef]
  44. Sobaih, A.E.E.; Algezawy, M.; Elshaer, I.A. Adopting an Extended Theory of Planned Behaviour to Examine Buying Intention and Behaviour of Nutrition-Labelled Menu for Healthy Food Choices in Quick Service Restaurants: Does the Culture of Consumers Really Matter? Int. J. Environ. Res. Public Health 2023, 20, 4498. [CrossRef]
  45. Sobaih, A.E. Ethical concerns for using artificial intelligence chatbots in research and publication: Evidences from Saudi Arabia. J. Appl. Learn. Teach. 2024, 7, 1-11. [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.