This study explored college student errors, teachers' oral corrective feedback, learner uptake and repair, and learners' corrective feedback preferences in a Chinese language classroom setting. Data collection included the oral interactions in the classrooms, a preference of feedback survey, and focus-group interviews. The present study adopted the coding schemes used in Yoshida (2008) and in Lyster and Ranta (1997) to categorize data. Chi Square tests were run to examine whether there were significant differences in the frequency of students' error types and teachers' feedback types, and whether teachers' feedback types and learners' subsequent responses were related. The results showed that the most frequently made errors were phonological and lexical and that recasts were the most frequently used feedback type. Also, this study found a statistical significance between feedback type and learner repair in the beginner class. In addition, the survey results showed that the majority of the beginners preferred recasts while the advanced participants had preferences more scattered among different feedback types. The interview results showed that the participants' preferences were influenced by their learning styles and beliefs, their proficiency levels, the nature of the Chinese language, and the differences between Chinese and their native language, English.