Mainland Chinese Students’ Social Identity and Cantonese Learning in Hong Kong : Differences between Undergraduate and Graduate Students

Identity is “one of the unifying themes of social science” (Jenkins, 2004: 8). Among the various perspectives of research on identity, identity and second language learning began to attract more attention in the 1970s (Norton, 2006). The 1990s witnessed much progress of research in this area. For example, Peirce’s (1995) later widely cited work on social identity, investment, and the right to speak in language learning. Other studies examined identity from different perspectives (Duff & Uchida, 1997; Leung, Harris & Rampton, 1997; Morgan,1997; Schecter & Bayley,1997; Thesen, 1997;). The first decade of the 21st century saw diversity in both perspectives and methodologies of research (Cervatiuc, 2009). Concepts such as human agency, the imagined community and multicompetence were newly involved in this area. Four major themes were identified, namely, identity and invest, identity and imagined communities, identity categories and educational change, and identity and literacy (Norton, 2006). Multilingualism became a focus of research related to identity (e.g. Chen, 2010; Dagenais, 2003; Harklau , 2000; Kanno, 2003; King & Ganuza , 2005; Lee & Norton, 2009; Liang, 2006; Miller, 2000; Mucherah, 2008; Rajadurai, 2010; Shin, 2010). Although research went beyond studying English as a second language in immigration contexts in this phase, most studies in this field were mainly focused on English as a second language. Besides, no previous study has been done on learning Cantonese as a second language. This study aims at exploring the issues related to social identity and Mainland Chinese students’ Cantonese learning in Hong Kong. It was focused on the differences between students in undergraduate programs and those in taught postgraduate and research programs. Qualitative and quantitative investigations were conducted. Data were collected through interviews (focus group ones and in-depth ones) and a questionnaire designed based on the interview data. One hundred and seventy Mainland Chinese students in Hong Kong participated in the survey study. Preliminary analysis of the data identified 156 valid responses to be included in the dataset. The reliability test of the scale in the questionnaire revealed that the scale is reliable to test perceptions of identity and Cantonese learning related issues that are focused on in this study (α=0.741). An independent sample t-test was done on the basis of descriptive analysis and exploratory factor analysis, with the program of study as the dividing variable. The original variable of Program was recoded into its dummy variable which was based on the dichotomy between undergraduate programs and non-undergraduate programs (taught postgraduate programs and research programs).Results showed that the two groups differed significantly in the sense of connectedness to Cantonese (p=.001