A Second Look at T-Unit Analysis: Reconsidering the Sentence

Abstract

I Although originally designed to assess syntactic development in written work of children learning their first language, T unit has been widely used in second language research and is particularly useful for oral production (see, for example, Beebe, 1983; Gaies, 1980; Larsen-Freeman, 1983). A variety of measures of complexity have been based on T unit, e.g., words per T unit, words per error-free T unit, and clauses per T unit (Gaies, 1980; Hunt, 1965; Larsen-Freeman, 1983). Nevertheless, for description of syntactic complexity in writing of adult second language learners, sentence-based analyses may be superior. Hunt (1965) introduced T unit, or minimal terminable unit, to measure development of sentences in writing of grade-school children. Each T unit contains one independent clause and its dependent clauses. Hunt (1970) described T units as the shortest units into which a piece of discourse can be cut without leaving any sentence fragments as residue (p. 189). A sentence has two (or more) T units when independent clauses (with subjects and finite verbs) are conjoined as in Example 1, but a single T unit when one or more clauses are embedded in an independent clause as in Example 2:

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 173 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.