"Cognitive and Linguistic Features of Adolescent Argumentative Writing: Do Connectives Signal More Complex Reasoning?"
PhD student Karen Taylor is first author of an article in Reading and Writing: "Cognitive and Linguistic Features of Adolescent Argumentative Writing: Do Connectives Signal More Complex Reasoning?" Co-authors are Joshua Lawrence (University of Olso), Carol M. Connor, and Catherine E. Snow (Harvard University).
The Common Core State Standards (Common Core State Standards Initiative, 2010) feature argumentative writing across the curriculum in grades 4 through 12, yet little is known about how young adolescents develop the challenging advanced language and literacy skills needed for these tasks. This study explored productive academic language use in the persuasive writing of a sample of 40 middle school students (grades 6–8) by examining the use of (1) argumentative moves that display various levels of sophistication and (2) major classes of connectives (additive, adversative, causal, and temporal) that signal different cohesive functions within a text. Essays in our analytical sample (n = 158) were produced in the context of an academic vocabulary curriculum, Word Generation, and were transcribed, coded, and analyzed for types of arguments by researchers and undergraduate research assistants. Subsequently, connectives were calculated by the Tool for the Automated Analysis of Cohesion (TAACO; Crossley, Kyle, & McNamara, 2016). Descriptive analyses reveal that the sixth–eighth grade students in our sample deployed complex reasoning in their essays; at least one dual perspective argument was present in 50% of the essays, and at least one integrative perspective argument was present in 42% of the essays. Multivariate regression analyses (with adjusted standard errors) reveal that adversative connectives (e.g., although, however) were related to the most complex arguments (integrative perspective), controlling for essay length and topic type (β = 20.13, p = .006), as well as to overall argument sophistication (β = 17.25, p = .02). The results show the value of brief, curriculum-based essays for assessing students’ argumentation skills.