Building a corpus of spoken Chinese interlanguage and some results of preliminary analyses

The corpus of spoken Chinese interlanguage in this study consists of over one million characters of transcribed student speech from data collected from nearly ten years of study abroad research. The main research method was a comparison with a similar corpus of spoken Chinese by native speakers. Preliminary analyses show that 11 of the top 20 most frequent words in both the learner corpus and native corpus are the same. Learners used some grammatical function words, such as 把 (bǎ), 了 (le), 它 (tā), and 着 (zhe) less than native speakers, while other ones, such as 我 (wǒ) and 的 (de), much more frequently. Possible explanations for these patterns, as well as pedagogical implications and directions for further research are discussed.