In recent years, there has been an increased emphasis on integrated writing assessment. Both
large-scale tests, such as the Test of English as a Foreign Language (TOEFL), and classroom
assessments use integrated assessment to evaluate test-takers’ writing abilities. The present study
aimed at investigating the complexity, accuracy and fluency (CAF) in EFL (English as a Foreign
Language) integrated writing assessment in online versus traditional mode. To this end, a total of
44 EFL undergraduate female students who majored in English language and literature at Persian
Gulf University (PGU) in Bushehr, Iran participated in this study by convenience sampling.
They were all at intermediate proficiency level in English. They received integrated writing
instruction for the whole semester (16 weeks). A quasi-experimental approach was adopted in
this study. Data was collected by giving the participants a reading prompt on a specific topic in
both online and traditional groups and asking them to write an integrated essay using the content
of the source text. Online participants used LMS (Learning Management System) as a virtual
classroom for uploading integrated essays produced through Microsoft Word and students in the
face-to-face group did the task using pencils and papers in a classroom. Among CAF triad,
accuracy and fluency were divided into branches. Lexical, morphological and syntactic accuracy
were the sub-components of accuracy and fluency was divided into word-count and sentence-
count fluency. CAF triad was assessed based on the global measurements of writing and
integrated writing essays were analyzed holistically using the TOEFL integrated rubric. The
results of analysis using independent samples T-tests revealed that the mean scores of
complexity, morphological accuracy, word-count fluency and integrated writing in the online
group was more than that of the traditional mode. On the other hand, the traditional group was
more competent in lexical accuracy, syntac