Creating characters and reconstructing texts: Evaluation in children's oral narrative re-tellings

This research analyzes the use of evaluative features in English language oral narrative re-tellings among a multi-lingual population of ninety-eight 2nd and 4th grade students. The results of the analyses strengthen our understanding of the use of evaluation by child narrators, suggesting that younger narrators reconstruct stories through re-creating the characters, while older children focus more on (precisely) reconstructing the text itself. Parallels with particular approaches to cognitive/psychological development are outlined, as are preliminary ramifications for educational methodology. In the initial rounds of both qualitative and quantitative analyses, it was revealed that the employment of seven evaluative forms cited in earlier research (e.g., Peterson and McCabe 1983, Bamberg 1991, Reilly 1992) as among the most commonly used by the present age group (i.e., causals, compulsion words, emphatic pronunciation, gratuitous terms, hedges, lengthening, and negatives) was unable to account for differences in perceived narrative skill within the sample. The manipulation of these seven features was extremely homogenous across skill, age, and language groups. As a result, a second round of analyses was undertaken. Both qualitative and quantitative findings concurred that the use of two particular evaluative features (i.e., references to mental activity, and character speech ), in addition to the utilization of certain textual devices (i.e., the presentation of mental activity within causal constructions, deference to a third person "other" as the source of the narrative information, careful monitoring and marking of errors), were capable of distinguishing both skill and age groupings within the sample. The manner in which the data from this research reflects the Vygotskian perspective on cognitive/psychological development is discussed. The educational implications of these findings---from assessment paradigms, to the planning of curriculum and instruction---are addressed. One of the major discoveries was that, counter to expectations, the multilingual subjects in this sample did not demonstrate divergent narrative forms based on their differing linguistic/cultural schemas. In fact, the perceived skill scores among the Limited English Proficient subjects appeased to be related to issues of fluency rather than differences in narrative form. These findings indicate that given a rich context in which information is repeatedly co-constructed, most language minority students are highly capable of both interpreting and reproducing information in a culturally/contextually prescribed manner