Promoting and Assessing Metalinguistic Awareness for Learning in Science with Multilingual Adolescents

• In Event: Word Learning for All Students: Metalinguistic Awareness and Vocabulary Instruction With Multilingual and Multidialectal Students

Abstract

Purpose and Perspective

This paper reports findings from the first year (Y1) of Words as Tools, a four-year project, to develop and test a program for multilingual adolescents to promote metalinguistic awareness about words that are central to learning in science. Metalinguistic awareness includes knowledge of how to use word meanings, word parts, and word functions (Gombert, 1992; Nagy, 2007). Rather than teaching only meanings of science words, Words as Tools aims to equip students with flexible, generative word knowledge. Two specific areas of metalinguistic awareness were targeted in Y1:
• identifying and analyzing Greek and Latin roots in science to raise awareness of semantic relationships and build networks among morphologically related words, and
• navigating polysemous words in science to promote cognitive flexibility (Colé et al., 2014; Deák, 2004; Scott, 1962) for problem-solving multiple sense words that carry scientific and everyday meanings.
Interventions designed to increase students’ metalinguistic awareness provide preliminary evidence these cognitive processes can be developed through carefully designed instruction and, ultimately, show positive effects on comprehension outcomes (Cartwright et al., 2017; Zipke et al., 2009). As multilingual learners should have opportunities to draw on linguistic and cultural resources, a lens of Culturally Sustaining Pedagogy (Alim & Paris, 2017) informs the design.

This study aimed to 1) examine correlations between researcher-designed measures of metalinguistic awareness and an English proficiency standardized assessment; and 2) compare pre- to post-treatment effects for metalinguistic skills as preliminary promise for the Words as Tools intervention.

Method

Multilingual students (n=45) in grades 6-8 enrolled in six English as a Second Language classes in two partner school districts participated in two 8-lesson mini-inquiries related to life sciences.

Pre and post measures were curriculum-based tasks designed to assess multiple dimensions of metalinguistic skills in the context of science language (Table 2). Standardized scores for WIDA English Language proficiency were collected from the districts.

Results

Table 3 presents correlations between WIDA standardized measure of English proficiency and tasks of metalinguistic awareness. Correlations were strongest with comprehension except for CtS (i.e., polysemy). Academic word and morpheme knowledge had positive and significant correlations with oral language. The more distal measures of AMK and CtS did not have significant correlations with WIDA. Results suggest that metalinguistic skills in science, as assessed in our tasks, have partial overlap with traditional measures of language proficiency.

Table 4 presents pre and posttest results (paired t-tests). We note significant differences for AWK and M, and associated large effect sizes (ESs), a moderate ES for CtS, and a small treatment effect for AMK.

Significance

This study did not include a comparison group thus we do not interpret results as causal. Results provide 1) preliminary evidence that our metalinguistic tasks tap related but not overlapping linguistic skills vis-a-vis a standardized English proficiency measure; and 2) positive evidence for the intervention’s potential to promote word knowledge and metalinguistic skill for learning science words. In future work, we anticipate that these metalinguistic dimensions would boost generative word learning and contribute to development of word knowledge networked around science concepts.

Authors

• Amy C. Crosson, Pennsylvania State University

• Dianna R. Townsend, University of Nevada - Reno

• Pui-Wa Lei, Pennsylvania State University

• Vickie Smith Barrios, University of Nevada - Reno

• Sophia Lopez, Pennsylvania State University