Self-Assessment and Teacher Assessment of Writing Situated Within Previous and Future Conversations and Writing Events

• In Event: 26.025 - Dialogic Perspectives on Classroom Assessment of Writing

Abstract

Purpose
This paper explores how individual students' self-assessments and their teacher’s assessment of argumentative essays reflect past conversations and point toward future writing events. Previous research on writing assessment in secondary English classrooms has focused on the impact of high-stakes assessment (Ketter & Pool, 2001) on classrooms or the nature of the assessments themselves (Beck & Jeffery, 2007) but few have examined how classroom-based assessment is situated within the arc of writing instruction.

Theoretical Framework
Bakhtin (1981) argues that all speech is dialogic, that an utterance reflects not just the person speaking but all of those to whom the utterance is addressed. If language – whether spoken of written - is seen as a response within a social, cultural, historical context (Bakhtin, 1986), then embedded within it is a history and future.

Data Sources and Methods
The data for this study are from a semester-long extended case study of three students learning to write argumentative essays about literature. Data sources for each student include two events, both on-demand argumentative essays about poetry, and for each event comprised the writing prompts and scoring rubric, student self-reflections on the essays, teacher’s written feedback and holistic scoring, transcripts from audio-recorded individual student writing conferences with the teacher, and related transcripts from video-recorded whole-class instructional conversations. I conducted an intertextual analysis (Prior, 1995) across these data sources. Specifically, for each of the two writing events, I foregrounded the student self-assessment and the teacher’s assessment through written and/or verbal feedback. I then forward and backward mapped (Dixon & Green, 2005) from these assessments across the other data sources to trace connections between previous and future texts and events, creating “item files” (Nuthall, 1993). For example, if a student wrote in a self-assessment about needing to provide better evidence, I identified previous and future instances within the student’s data set in which providing evidence occurred as an issue.

Findings
Analysis showed that what the student and teacher referenced in the assessment process was related to previous whole-class and individual conversations. Although the student self-assessment form was open-ended, students focused on writing elements that the teacher and students discussed in previous whole-class conversations. The teacher also referenced these same writing elements, and in referencing this shared collective memory, indicated that the students were assessed within the context of these whole class conversations. At the same time, individual students’ self and teacher-assessments focused on different writing elements, and these areas of focus impacted each student’s subsequent writing and self-assessment.

Significance
This study highlights the situated nature of assessment, that it not a decontextualized measure of writing but carries traces of past conversational events while also influencing future writing events. It raises questions about the validity of decontextualized assessments, particularly standardized writing assessments that may or may not be measuring the particularities of what students are learning in various writing classrooms.

Author

• Jennifer Lynn VanDerHeide, Michigan State University