About MICUSP

Welcome to the MICUSP website!

MICUSP, the Michigan Corpus of Upper-level Student Papers, is a research project at the English Language Institute at the University of Michigan. Launched in late 2004, the project is still underway, and we are now actively developing the corpus. Our goal is to compile a large selection of writing samples, totalling some 1.6 million words, from writers at different stages of undergraduate and graduate level study, both native and non-native speakers, from across the university.

The MICUSP database will be used to systematically study linguistic features of academic writing. The MICUSP project is the first of its kind in North America and will serve as a valuable resource for teachers of English as a Second Language, testers, and researchers who are studying and teaching academic writing. The specific rationale for our project is that, while there are a number of existing research projects designed to study academic writing at a professional level (based, for example, on published papers) as well as at the undergraduate level, there is a lack of resources pertaining to graduate level study.

MICUSP is an official, IRB-approved, University of Michigan research project. All contributions to the database will be anonymized. The finished corpus will be made accessible to academic researchers and teachers by subscription only.

If you are a student at the University of Michigan and wish to contribute texts to the project, please go to CONTRIBUTE.

Quick facts about MICUSP

• The corpus consists of proficient texts, i.e. A-grade papers or ungraded papers that have been assessed and accepted (such as research proposals), but not published. The samples collected will demonstrate the standards set by instructors at each stage of study.
• The writers are senior undergraduates or graduate students at the University of Michigan, in the following disciplines:
  Philosophy, Linguistics (Humanities)
  Psychology, Sociology, Economics (Social sciences and education)
  Physics, Civil Engineering, IOE (Physical sciences and engineering)
  Nursing, Biology (Biological and health sciences)
• The corpus offers a semi-balanced sample of native and non-native speakers of English.
• The length of the texts ranges from 500 to 10,000 words.