The data-driven learning (DDL) approach has been applied to English teaching for over a decade. However, corpus-based pedagogy has not been widely applied to Chinese as a second language (CSL) teaching, probably because CSL teachers lack corpus literacy, such as familiarity with corpus tools, corpus consultation techniques, and corpus-based activities. To cultivate corpus literacy and promote DDL application in classrooms, 73 CSL teachers participated in a 4-week DDL training workshop for 3 h per week. The workshop covered four topics: introduction to Chinese corpora, corpus consultation techniques, practical training in the use of corpus analysis tools (e.g., AntConc), and pedagogical designs or activities for guiding learners to induct or deduce rules from corpus data. After the workshop, a postworkshop questionnaire consisting of 39 questions scored on a 5-point Likert scale and 4 open-ended questions was distributed to participants. Of the 39 quantitative questions, 36 were related to seven categories that reflected the attributes of the participants. Through these questions, the participants provided self-evaluations of their (1) motivation for participating in the workshop, (2) prior knowledge of corpora, (3) skills in using corpora after the workshop, (4) negative perceptions of adopting corpora in teaching, (5) skills in designing and implementing DDL activities, (6) willingness to adopt corpus-based methods in future teaching, and (7) the quality and usefulness of the workshop. The results of Spearman’s rank-order correlation test indicated that, regardless of the level of motivation of the teachers to attend the workshop, they gained corpus-related abilities through lectures and practice. Among the teachers, those with more prior knowledge of corpora and with greater confidence in their skills to use corpora were more likely to implement DDL teaching activities in their classrooms. These results indicate the effectiveness of cultivating teachers’ corpus literacy, and continually offering training courses can increase the implementation of the corpus-based approach in the future.