First graders developing emergent writing: A cross -linguistic study

The purpose of the present study is to investigate emergent writing development among 16 monolingual American and 16 bilingual Chinese first grade children. The study uses a mixed two-factor 2 x 2 within subject design with repeated measures on each dependent variable. There are two levels of instruction (Chinese plus English, English only) and two levels of data collection time (Fall 1999, Spring 2000). The between subject factor is the instructional program; the within-subject factor is time. The data are four writing samples (list of known words, spelling lists, dictated story, handwritten story) created at each of the data collection sessions (total data, 256 pieces of writing). The dependent variables were: (a) indices of cohesive and non-cohesive harmony (CHI, NCHI), written language features (WFOI, WFBI), metacognitive utterances (PMI, DMI, TMI) and (b) scores for correctly spelled words under two presentation conditions (auditory [ASS] and visual memory [VMS]. The variables were submitted to MANOVA, ANOVA, and t-tests to determine statistically significant effects. In addition, the writing samples, a teacher interview, and student interviews regarding spelling strategies provided qualitative data. Finally, the lists of known words and the handwritten stories were examined for evidence and the situated occurrence of code-switching with the constant comparison method. A frequency count was also made of occurrences. The study yielded several important findings for the bilingual and monolingual children. At the beginning of the study the two groups showed: (a) no statistically significant differences on any measure, and (b) very similar patterns and knowledge of emergent writing principles and forms/invented spellings. At the end of the study, the bilingual children surpassed their monolingual peers with statistical significance on their use of cohesion (CHI), metacognition (PMI), and spelling strategies (ASS, VMS). In addition, the bilingual children\u27s code-switching proved to be rule-governed, demonstrating a quite sophisticated reliance on register (i.e., situational) factors for its use. Overall, the study suggests that the bilingual children were quite capable of developing two distinctly different writing systems simultaneously, when provided adequate access to both. Implications for general education, bilingual education, and further research are discussed