Linguistic and cognitive development in the writing of middle-grade English language learners

ABSTRACT. This study investigates the writing development exhibited in essays written four months apart on two topics by 55 middle-grade (sixth through eighth), primarily Hispanic English language learners. Five measures of writing development are employed: words per T-unit, clauses per T-unit, lexical diversity D-optimum score, percent of error-free T-units, and number of T-units. Correlations between measures is examined as well as differences related to topic, time of writing, and grade level. Interpretation of the results suggests that cognitive, age-related development must be considered separately from general language development when working with adolescent second language writers. * INTRODUCTION. The following study investigates the development of writing in middle-grade second language (L2) writers from a largely Hispanic background. In the school district where the study was conducted, English as a Second Language (ESL) students at each grade level are classified into three proficiency levels beginning in the fifth grade. To determine a student's appropriate level, teachers administer a standardized oral proficiency exam that asks increasingly difficult questions until the student reaches a level at which they can no longer answer. Although the focus of the ESL program is Language Arts and preparing the students for regular English classes, there is no evaluation of their writing for placement purposes. This is not because the school system is not interested in the writing abilities of their students; they have asked me several times to conduct staff development workshops on ESL writing. Rather, I believe it is because it is not clear what would constitute a valid measure of the development of L2 writing abilities for this age group. Creating a valid index of development for this age group is problematic because the relative importance of linguistic and cognitive dimensions in their writing is not clear. With adult L2 writers, a noun preceded by several descriptive adjectives and followed by a relative clause signals greater linguistic sophistication than a series of short, simple sentences with no subordination or embedding. The cognitive ability to generate descriptive content is not at question. But what if the heavily modified noun phrase comes from a writer who has had six years of formal education with extensive reading in the first language (L1) whereas the series of simple sentences comes from a writer who has never written elaborated prose in any language and who has had limited access to reading materials? Then we are no longer sure whether the limitations are linguistic or cognitive. With adolescent L1 writers or younger L2 learners enrolled in bilingual education programs, the question is moot. Language development is treated as an integral part of cognitive development. For 12- to 14-year-old English language learners with varied schooling backgrounds, however, it is important to consider what aspects of writing are attributable to their age and cognitive development and what aspects reflect the degree to which their language abilities differ from their native-speaking peers. In the following study five measures related to both cognitive and linguistic dimensions of writing are used to gauge changes in the writing of middle-grade English language learners both within and across grade levels and between two topics: T-unit length, syntactic complexity, lexical diversity, morphosyntactic accuracy, and number of T-units. While no substitute for detailed readings of a collection of writing samples exists for understanding the needs of a particular writer, quantitative indices such as these provide insight into macro-level changes that may be occurring within the group. The purpose of this study is to gain insight into how the cognitive and linguistic dimensions of writing relate to each other. 1. MEASURING WRITING DEVELOPMENT. The term WRITING DEVELOPMENT is ambiguous in that it can signify either elaboration of a single text or changes in writing ability over time. …