With data ranging from 4th year undergraduates to third year graduate students, we will be able to analyze graduate student writing development, asking questions like: What changes, for example in grammar, lexis, and discourse patterns, do we find in texts as students become more advanced?
Our database will provide a snapshot of assessed genres used at a major US research university, allowing us to explore questions like: What are the characteristics of the various genres that have developed? How similar is graduate student writing to published scholarly writing?
As a complement to the MICASE corpus, MICUSP will make it possible to analyze the similarities and differences between student speech and student writing. We can investigate questions like: What linguistic effects do the different conditions of speech and writing give rise to?
With data from a range of departments at the U of M, we will be able to map disciplinary variation in academic writing across the university, asking questions like: How do writing styles, conventions and technique vary across departments and disciplines?
With a balanced sample of writing by both native and non-native speaker students, we will have an opportunity to ask questions like: What are the differences between native-speaker versus non-native-speaker writing patterns, for example in the use of metadiscourse?
With a matching database of proficient student writing in other languages or language varieties, we will be able to make cross-linguistic and cross-dialectal comparisons. As an example, a similar corpus of British English student writing is being compiled in the UK.