Raising Orthographic Awareness of Teachers of Chinese
THE STUDY
Think-Aloud Protocols
Data were collected from think-aloud protocols, class observations, and follow-up interviews. The think-aloud activity was conducted at the end of fall semester 2001. Think-aloud analysis has been used as an effective means to observe the cognitive processes of participants in a number of fields such as psychology, education, and cognitive science (Ericsson et al., 1993). It allows the participants to verbalize the information they attend to with their thought sequencing being uninterrupted. In this study, the researcher could, by using this method, effectively explore the participants’ level of orthographic knowledge and their pedagogical decision-making in character writing.
The think-aloud activity was conducted individually and self-audio taped. It consisted of two parts: Part A and B. While Part A elicited comments on the pedagogy course, Part B focused on error analysis and strategy articulation. In Part A, the participants were instructed to answer three questions: (1) Which part of the pedagogy course do you think is most helpful to you? (2) What knowledge has helped you most in your follow-up teaching practice? (3) Taking character writing for example, how did you treat your students’ errors before and how do you treat them now? In Part B, the participants were asked to identify and correct errors from a list of fifty sample characters, which were either erroneous or non-characters. While correcting, the participant was instructed to (1) tell everything in his/her mind, and (2) answer questions to him/herself such as “How should I help the students correct this error? What should I do in the future to help prevent such errors?”
The fifty sample characters were selected from learners’ homework, quizzes, and examinations, which reflected the fourteen common error types in character writing identified by Shen et al. (2000). Based on their orthographic features, these fourteen error types were further classified into three categories: phonological, graphemic, and semantic. Drawing data from the writing samples of one hundred and one CFL students at three instructional levels, Xiao (2002) found that, out of the three error categories, graphemic errors were the highest, phonological errors the second, and semantic errors the lowest. To reflect CFL students’ actual learning, the fifty sample characters used in the think-aloud protocols did not have an even number of errors for each category; instead, out of them, twenty-eight (56%) were graphemic errors, five (10%) phonological errors, and three (6%) semantic errors. In addition, there were twelve (24%) combined phonological/graphemic errors and two (4%) phonological/semantic errors.
All the six former trainees participated in the think-aloud activity. While the two English-speaking participants used English to describe their thoughts, the four Chinese native speakers chose Chinese. The average time used for this activity was half an hour, ranging from 10 minutes to an hour each. There were noticeable differences among the participants in terms of level of elaboration. Some of the participants could merely name the error types, the others gave in-depth analysis of the cause of errors, suggested corrective and preventive strategies, and recalled their own learning and teaching experiences.
