Moving beyond genre: Mapping student writing in a multidimensional situational space

While research on academic writing has long recognised the importance of text types, consistent and principled classification has remained problematic. Existing taxonomies often prove difficult to apply to new corpora, and key terms (e.g., exposition, narrative) are inconsistently defined. Taxonomies have also been criticised for failing to capture the diversity and dynamism of academic writing and for encouraging formulaic teaching practices. Building on work by Biber, Egbert, and Keller (2020), this paper develops an alternative approach that may provide a more nuanced and dynamic representation by viewing text type variation as continuous and multidimensional, rather than categorical. We develop a detailed set of situational parameters to describe variation in student writing and apply them to a balanced sample of texts from the British Written Academic English (BAWE) corpus. Using factor analysis, we identify four key dimensions of variation: (1) explicit library research, (2) focus on quantitative data vs. individuals, (3) reporting a research or design process, and (4) personal informal writing. These dimensions are proposed as a flexible heuristic for understanding how student texts vary within and across traditional categories.By mapping corpus texts on our dimensions, we empirically characterize traditional genres as cultural constructs, highlighting their core features, overlaps, and disciplinary differences. We also show how individual texts can be positioned in relation to these dimensions and genres. We argue that this multidimensional approach offers a more coherent and adaptable framework for analysing variation in academic writing, with important implications for English for Academic Purposes (EAP) pedagogy and research.