Date of Award

May 2013

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Department

English

First Advisor

Patricia D. Mayes

Second Advisor

Rachel D. Spilka

Committee Members

Sandra L. Pucci, Fred R. Eckman, Edith A. Moravcsik

Keywords

Bilingual Education, Bilingual Learners, Composition, English as a Second Language, Generation 1.5, Metalinguistic Awareness

Abstract

In spite of growing numbers in high schools and colleges, US-resident adolescent bilingual learners, sometimes termed "English as a second language" (ESL) or "Generation 1.5," are not succeeding academically in proportion to their monolingual English-speaking peers. This achievement gap is evident in their writing as they enter college. Depending on the elementary and secondary schools they have attended, bilingual learners may have received no extra English learning support (often termed "immersion"), ESL support classes, or bilingual education. In addition, depending on school and community resources, bilingual learners have varying knowledge of their first language (L1): some may only speak it, others may have basic L1 literacy, others may have studied their L1 as a school subject, while others may have studied in the medium of their L1, either in their family's home country or in a bilingual education program in the US. The purpose of this study is to determine which kind of English learning support and which kind of L1 education are more likely to prepare bilingual learners to write English successfully at college.

This study uses three sources of data: a survey on language background, a writing sample, and an optional interview. Twenty-nine college undergraduate bilingual learners participated. Their survey responses develop a profile of the varied kinds of English and L1 education they received. Each participant's communication course placement composition, written as she was applying to college, is analyzed with 12 different measures: six for surface features, four for discourse/rhetorical features, and two for coherence. The writing analysis scores are correlated with the survey data and enriched with interview excerpts to discover which forms of English and L1 education correlate with high or low writing analysis scores.

The results for this group of participants show that bilingual education and ESL support correlate most often with highly-rated communication placement compositions. Moreover, formal education in the L1 explains the writing analysis scores more accurately than the kind of language learning education the participants received. Interview data suggests that bilingual education and formal L1 education may assist students' English composition skills by helping them develop metalinguistic awareness.