A Corpus-based Study of Spoken and Written Registers

Aim : to provide a relatively comprehensive linguistic description of the range of university registers, surveying the distinctive linguistic characteristics of each register

Question: How is language used across a wide range of university registers? (spoken and written, formal and informal, embracing the major disciplines (humanities, natural and social sciences), the main academic levels (undergraduate and postgraduate), and the typical situations in which students might find themselves (lectures, seminars, tutorials, interaction with administrative staff))

Corpus : T2K-SWAL corpus. The T2K-SWAL corpus consists of 2.7 million words captured at four US universities. Unusually for an academic corpus, almost 1.7 million of those words are from recorded spoken sources, while only 1 million originated in written material. Within these categories, most of the spoken data were obtained from class sessions (1.2 million words), while only 50,000 were from office hours.

Findings : Despite a certain superficial similarity of pragmatic function, textbooks employ vastly more nouns and fewer verbs than are used in classroom teaching. Textbooks also contain more relative clauses but fewer complement and adverbial clauses than classroom teaching, and that the complement clauses used are less likely to be "that-" or "wh-" clauses. Classroom language on the other hand abounds in adverbs of certainty and likelihood, both of which are relatively uncommon in textbooks, and uses many more modal verbs of all kinds. Classroom teaching situates firmly towards the "spoken" pole of the continuum from spoken to written, putting paid to any notion that a class is somehow equivalent to a textbook: the objective may be the same, but the way in which it is achieved is linguistically quite different.

Source: http://tesl-ej.org/ej42/r5.html