Identifying Leverage Points Beyond Cognates for Spanish-English Bilingual Adolescents: Effects of an Academic Vocabulary and Morphology Intervention

• In Event: 43.076 - Learning Vocabulary in a Second Language: Exploring Challenges and Opportunities Across Levels of Development and Proficiency

Abstract

Objective: We will present findings from an academic vocabulary intervention that was infused with instruction about Latin roots and cross-linguistic connections for multilingual adolescents whose home language is Spanish (n=131) and whose home language is other than Spanish (n=61) in U.S. schools.
Theoretical Framework: The intervention was designed to improve lexical quality of academic words key for comprehension (Perfetti & Stafura, 2014) and to promote a problem-solving orientation using morphological information associated with comprehension growth (Levesque, Kieffer, & Deacon, 2018). We hypothesized that instruction emphasizing morphological analysis of roots integrated into robust academic vocabulary instruction (Beck et al., 2013) would strengthen semantic networks between words and across languages (Kroll & Bialystok, 2013), leading to more robust semantic representations, and in turn, greater depth of knowledge of targeted academic words; and would enhance morphological analysis skills. Previous research has shown that improvements in vocabulary knowledge is associated with multilingual learners’ comprehension outcomes (Lesaux et al., 2010; Snow, Lawrence & White, 2009).
Methods: Participants were middle school students assigned to an intervention (n= 92) or comparison (n=100) condition, from 19 classes in two urban U.S. school districts. The majority spoke Spanish as their home language (L1) with 26 other L1s (e.g., Tongan, Malay, Kizigua) represented. Intervention students participated in a 16-week program targeting 64 high utility target academic words (e.g., confine, innovative). The instruction, provided by classroom teachers, emphasized active processing of word meanings across multiple contexts. For each target academic word, students were taught a corresponding Latin root (fin= limit; nov= new) and cross-linguistic connections between roots and everyday words in Spanish, called “Spanish Friends” (e.g., fin = “end”; nuevo/a = “new”). Students engaged in using roots to problem-solve meanings of new words (e.g., finite, novice). Measures administered at pre and posttest included: a) academic word meanings task; b) roots knowledge task; and c) morphological analysis task. Multilevel models with students nested in classrooms were built for each posttest outcome, controlling for standardized reading comprehension (pretest Lexile), district, and pretest performance on each outcome, with condition and L1 (Spanish or not) as key predictor variables. Interactions between condition and L1 were tested for each model.
Results: Two important findings were observed. First, significant treatment effects of condition on word knowledge (p<.001) and root knowledge (p<.0001) were observed. Moreover, L1 Spanish was a significant predictor for root knowledge (p=.02) but not word knowledge. For root knowledge, the model estimates that L1 Spanish speakers score 3.348 points higher on average than other L1s at posttest, regardless of condition.
Significance: Findings suggest value of the intervention for both L1 Spanish-speakers and other L1 backgrounds for learning word and root meanings, and for learning to use roots for morphological problem-solving of unfamiliar words. Spanish- English bilinguals draw on L1 resources to identify root meanings, even in the absence of intervention. However, we did not find evidence that Spanish-English bilinguals leverage this knowledge for morphological problem-solving more effectively than speakers of other L1s. Instruction, whether or not they participated in the academic vocabulary and morphology intervention.

Authors

• Amy C. Crosson, The Pennsylvania State University

• Margaret G. McKeown, University of Pittsburgh

• Pui-Wa Lei, The Pennsylvania State University

• Xinyue Li, The Pennsylvania State University

• Kelly P. Robbins

• Kathleen J. Brown, University of Utah