<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE ArticleSet PUBLIC "-//NLM//DTD PubMed 2.7//EN" "https://dtd.nlm.nih.gov/ncbi/pubmed/in/PubMed.dtd">
<ArticleSet>
<Article>
<Journal>
				<PublisherName>University of Sistan and Baluchestan</PublisherName>
				<JournalTitle>Iranian Journal of Applied Language Studies</JournalTitle>
				<Issn>2008-5494</Issn>
				<Volume>18</Volume>
				<Issue>1</Issue>
				<PubDate PubStatus="epublish">
					<Year>2026</Year>
					<Month>07</Month>
					<Day>01</Day>
				</PubDate>
			</Journal>
<ArticleTitle>The Impact of Using Grammarly on Complexity, Accuracy, Fluency, and Self-Efficacy in Iranian EFL Learners’ Argumentative Essay Writing</ArticleTitle>
<VernacularTitle></VernacularTitle>
			<FirstPage>43</FirstPage>
			<LastPage>62</LastPage>
			<ELocationID EIdType="pii">9682</ELocationID>
			
<ELocationID EIdType="doi">10.22111/ijals.2026.53481.2563</ELocationID>
			
			<Language>EN</Language>
<AuthorList>
<Author>
					<FirstName>Hamed</FirstName>
					<LastName>Mahsefat</LastName>
<Affiliation>English Language Department, Faculty of Humanities, University of Guilan, Rasht,   Iran</Affiliation>

</Author>
<Author>
					<FirstName>Abdolreza</FirstName>
					<LastName>Tahriri</LastName>
<Affiliation>English Language Department, Faculty of Humanities, University of Guilan, Rasht,   Iran</Affiliation>

</Author>
</AuthorList>
				<PublicationType>Journal Article</PublicationType>
			<History>
				<PubDate PubStatus="received">
					<Year>2025</Year>
					<Month>10</Month>
					<Day>07</Day>
				</PubDate>
			</History>
		<Abstract>Artificial intelligence (AI) is reshaping language education by offering real-time, personalized writing feedback. This study assessed the impact of Grammarly, an AI-powered writing assistant that utilizes advanced natural language processing and generative AI, on the complexity, accuracy, fluency (CAF), and self-efficacy of Iranian EFL students’ argumentative writing. A total of 120 high-intermediate learners from intact classes were screened for proficiency level via the Oxford Quick Placement Test and then randomly divided into an experimental group (Grammarly, n=60) and a control group (conventional teacher feedback, n=60). Over 12 weeks, a period deemed necessary for the development of complex writing skills, the experimental group used Grammarly for automated feedback on grammar, vocabulary, and style during weekly writing tasks, while the control group received conventional teacher feedback. ANCOVA analysis, controlling for pretest results and conducted after confirming key statistical assumptions, showed significant gains for the Grammarly group in complexity (F=1277.009, p &lt;.001), accuracy (F=2118.838, p&lt; .001), fluency (F=49.535, p&lt;.001), and self-efficacy (F=409.237, p&lt;.001), with the largest improvements in accuracy. These results align with the CAF framework, indicating that AI tools excel most in rule-based domains (e.g., grammar) rather than procedural fluency. The findings underscore Grammarly’s value in scaffolding grammatical competence and learner confidence, specifying that the tool is most effective for grammatical scaffolding.</Abstract>
		<ObjectList>
			<Object Type="keyword">
			<Param Name="value">AI-assisted Education</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">Grammarly</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">Complexity</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">Accuracy</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">Fluency</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">self-efficacy</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">Argumentative Writing</Param>
			</Object>
		</ObjectList>
<ArchiveCopySource DocType="pdf">https://ijals.usb.ac.ir/article_9682_f17fdf2e4cc4ea00b0c816e87c9b7588.pdf</ArchiveCopySource>
</Article>
</ArticleSet>
