Genres of high-stakes writing assessments and the construct of writing competence

High-stakes writing assessments currently exert a strong influence on the writing curriculum and instruction in schools across the United States. Under these circumstances it is important to examine the construct of writing competence on which these assessments are based, as well as the extent to which this construct supports the goals of secondary education. In this paper we conduct an exploratory analysis of the genre demands of high-stakes writing assessments from three states – California, Texas, and New York – with the aim of discerning, comparing, and evaluating the role that genre knowledge plays in the construct of writing competence measured by these assessments. Our method of inquiry includes both task analysis of the prompts and genre analysis of high-scoring benchmark papers written in response to these prompts. For the analysis of benchmark papers we employed both structural analysis and quantitative counts of key linguistic features to characterize the genres represented in these assessment tasks. Our results suggest a lack of alignment between the genres of the benchmark papers designated as exemplary and the genre demands of the prompts to which they were written. Exceptions to this pattern were most common on the New York assessments, which contextualize writing tasks in tests of subject–matter knowledge. Findings from our exploratory analysis lead us to argue for greater consistency and clarity of expectations in the design of high-stakes writing exams, and for the design of writing tasks that adequately represent the demands of discipline-specific forms of written discourse.