Yabancı Dil Olarak Türkçe Öğrenicilerinin Yazdıkları Metinlerde ve ChatGPT Tarafından Yazılan Metinlerde Karmaşıklık, Doğruluk, Akıcılık ve İşlevsel Yeterlik
Complexity, Accuracy, Fluency, and Functional Adequacy in Texts Written by Learners of Turkish as a Foreign Language and Texts Generated by ChatGPT

Summary/Abstract: The purpose of this study is to compare texts written by learners of Turkish as a foreign language with texts generatedby ChatGPT in terms of complexity, accuracy, fluency, and functional adequacy. The study was conducted using acomparative descriptive research design. The study examined 72 texts written by learners at the A2 and B2 levels onsix different topics and 6 texts written by ChatGPT on the same topics. Quantitative content analysis was applied tothe texts. The results of the study determined that ChatGPT had a higher percentage of complex sentences but did notshow consistency across levels. ChatGPT produced more accurate texts than the learners at the sentence and wordlevels. ChatGPT had higher averages in terms of different word usage, but the learners achieved higher values in termsof meaningful word ratio. In terms of functional adequacy, ChatGPT scored higher. It was also observed that somelearner texts at the A2 level were more qualified than those at the B2 level. The results indicate that ChatGPT has thepotential to serve as a support tool for learners of Turkish as a foreign language in improving sentence and wordaccuracy and ensuring functional adequacy. It is recommended to use ChatGPT during the text planning or reviewstages.Keywords: fluency, ChatGPT, accuracy, functional adequacy, complexity, teaching Turkish as a foreign language

• Details
• Contents