This is the authors' abstract. We don't add key points for this paper.
Academic writing is a key site for negotiating professional identity under evaluative conditions. Through stance, writers manage epistemic commitment, authorial visibility, and reputational risk while aligning their texts with disciplinary expectations. Although previous studies have documented broad novice–expert differences, less is known about how evaluative resources are distributed across functionally distinct units and whether recurrent distributional patterns reflect different forms of identity positioning. This study adopts a move-sensitive corpus-based approach to examine evaluative stance in abstracts and conclusion sections. The corpus comprises 80 Applied Linguistics texts, including 40 novice-authored theses/dissertations and 40 expert-authored journal articles. Sentence- and sub-sentence-level units were coded by primary rhetorical function, and hedging, boosting, and self-mention were identified and normalized per 100 words. Descriptive and comparative analyses examined the distribution of functional units and stance deployment across writer groups, while k-means clustering of document-level normalized stance rates identified recurring stance profiles. The results show that novice and expert writers differ not only in overall stance density but also in how evaluative resources are distributed across functional units. Expert writing shows denser hedging and slightly greater self-mention, especially in conclusion-oriented discourse, whereas novice writing relies more heavily on boosting, particularly in result- and conclusion-oriented units. Clustering further identifies three recurring profiles (Authorial-balanced, Boosting-oriented, and Hedging-oriented) and their distribution differs significantly across writer groups despite substantial within-group variation. These findings suggest that novice–expert differences are better understood as context-sensitive forms of academic identity positioning than as fixed deficits. The study offers implications for academic writing pedagogy, function-sensitive feedback, and AI-supported writing development.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Cheng et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: