English Learners in STEM Subjects : Contemporary Views on STEM Subjects and Language With English Learners

Heterogeneity of English Learners

The heterogeneity of ELs in terms of their demographics is commonly known; however, the specific details of this heterogeneity are more nuanced and require some unpacking. For the 2015–2016 school year, ELs constituted 9.5% of public school students, or an estimated 4.8 million students (National Center for Education Statistics [NCES], 2018). Not surprisingly, ELs vary in many ways—in their home countries and languages, their proficiency in their home language, the age at which they enter school and their prior schooling in other contexts, and their prior knowledge about STEM subjects (NASEM, 2018, pp. 27–33).

The EL STEM report then highlights what is not commonly known about ELs: “Against common intuition, the majority of ELs in the country are U.S. born (Zong & Batalova, 2015)” (NASEM, 2018, p. 30). Generally, ELs are students whose native language is a language other than English and whose English proficiency may have consequences for success in classrooms in which the language of instruction is English (NASEM, 2018, p. 1). Long-term ELs are those ELs who have been receiving English language development (ELD) and/or English as a second language (ESL) services in U.S. schools for at least 6 years and have not yet met reclassification criteria for their state. In contrast, newcomers are foreign-born ELs who have recently arrived in the United States. The EL STEM report highlights that all ELs, even newcomers, with sufficient support(s), can interact with children who speak English and participate and contribute in authentic STEM learning contexts. That is, all ELs regardless of their English language proficiency should have the same opportunities in STEM subjects as their non-EL counterparts.

Inconsistency of Educational Policies With English Learners

Compounding the heterogeneity of ELs, educational policies that address ELs are often inconsistent across states. These inconsistencies include the definition of ELs and approaches to classify and reclassify them as well as the variety of program models and variability in the quality of instruction under all program models.

Some educational policies that are fundamental to EL education are not commonly known. First, when examining policies articulating how to define ELs, there is no common definition of ELs across states. Section 8101(20) of the Elementary and Secondary Education Act provides a definition of ELs. However, “because states have different criteria to implement legislation regarding the definition of ELs, whether a student is regarded or not as being an EL depends, at least to some extent, on the state in which a given student lives” (NASEM, 2018, p. 212). Only recently was there an effort to develop processes, recommendations, and tools to build a common definition of ELs across states (Linquanti et al., 2016).

Second, there is no common approach to classification and reclassification of ELs. The classification to enter EL status and reclassification to exit EL status “var[y] considerably across states, and even across districts within states” (NASEM, 2018, p. 39). Across all states, English language proficiency (ELP) assessment—as specified by the state—is a major criterion for classification and reclassification of EL status. However, some states with large EL student populations require additional criteria. These criteria may include proficiency in content area achievement as measured by standardized test scores and/or grades in English language arts and/or mathematics. Such variability in classification and reclassification policies contributes to variations in STEM learning opportunities for ELs (NASEM, 2018, pp. 39–40, 252–256).

Third, the ways in which ELs are accounted for in the accountability system have led to overestimation of achievement gaps in STEM subjects between ELs and non-ELs. The EL subgroup is unlike other accountability subgroups under Title I in that the EL designation is dynamic—a student’s classification as EL changes as the student becomes proficient in English (NASEM, 2018, p. 41). As ELs become proficient in English, they are reclassified and counted as non-ELs. Exclusion of recently English-proficient ELs from the EL accountability group leads to overestimation of achievement gaps in STEM subjects between ELs and non-ELs and underestimation of ELs’ STEM proficiency (NASEM, 2018, pp. 41–44, 212, 219, 253–254). To address this issue, researchers have suggested using “an ever-EL framework by allowing states to count former ELs in their calculations of EL outcomes” (Umansky, Thompson, & Diaz, 2017, p. 78; also see Thompson, 2017). Kieffer and Thompson (2018) reported that including potentially English-proficient ELs in the EL accountability group showed a reduction in the mathematics achievement gap (NASEM, 2018, p. 42).

Lastly, although the broad categorization of language instruction program models is commonly known, the degree to which these program models emphasize and measure academic achievement in STEM subjects is less understood. The various program models are predominantly categorized into (a) primary language instruction for newcomers, (b) bilingual instruction, or (c) English-only instruction (NASEM, 2018, pp. 34–39 and Table 2-1). For all program models, the emphasis is on English language development as required by law, and they are “often not organized in a way that enables ELs to maintain and develop age-appropriate knowledge of STEM subjects” (NASEM, 2018, p. 38). Moreover, given the requirement by law to measure English language development, there is less emphasis on measuring STEM-related outcomes (NASEM, 2018, p. 39).

Contemporary Views on Content Standards and ELP Standards From Policy Perspectives

From policy perspectives, federal legislation has mandated alignment between content standards and ELP standards since No Child Left Behind (NCLB) of 2001 (U.S. Department of Education, 2001, as cited in NASEM, 2018, pp. 10, 34, 210–211, 255, 270–271). Then, the Elementary and Secondary Education Act Flexibility of 2012 slightly revised the original NCLB legislation for alignment (U.S. Department of Education, 2012, p. 1). Currently, the Every Student Succeeds Act of 2015 mandates that “the State has adopted English language proficiency standards that . . . are aligned with the challenging State academic standards” (U.S. Department of Education, 2015, p. 24). The direction of this relationship that “language proficiency standards align to content standards and not the other way around” suggests that “the language to be learned needs to focus on the important STEM content and what is known about how children learn STEM content” (NASEM, 2018, p. 10). Moreover, federal legislation makes clear that “language proficiency is not a prerequisite for content instruction, but an outcome of effective content instruction” (NASEM, 2018, p. 10), which is repeated throughout the EL STEM report (and described in more detail in the following).

Content standards across STEM subjects have been evolving for almost three decades (Chapter 3). The mathematics education community presents a contemporary view of mathematics instruction to promote mathematical proficiency, practices, and discourse. These three aspects of mathematics instruction are evident in reforms initiated by the National Council of Teachers of Mathematics in the 1990s, Adding It Up (NRC, 2001), and the CCSS for mathematics. Likewise, the science education community presents a contemporary view of science instruction to promote three-dimensional science learning by blending science and engineering practices, crosscutting concepts, and disciplinary core ideas. These three dimensions of science instruction are evident in reforms initiated by Science for All Americans (American Association for the Advancement of Science [AAAS], 1989) and Benchmarks for Science Literacy (AAAS, 1993) by Project 2061, National Science Education Standards (NRC, 1996), A Framework for K-12 Science Education (NRC, 2012), and the NGSS (NGSS Lead States, 2013). Engineering education is a relatively recent addition to K–12 education, and A Framework for K-12 Science Education and the NGSS articulated a new vision for three-dimensional engineering learning. Technology education is interpreted in a variety of ways, and currently, there is no coherent conception of technology education, including computer science or computational thinking.

Given that content standards are continuously evolving, ELP standards must also change and evolve so that ELP standards are aligned with content standards (NASEM, 2018, p. 10). Whereas mathematics and science standards have been evolving for almost three decades, ELP standards have a primary shortcoming—there has been a lack of a consensus on what language is and does and how language is learned in EL education (Lee, 2018; Valdés, Kibler, & Walqui, 2014). In the absence of an agreed-on framework for ELP standards, there are multiple sets of ELP standards, each with its own theoretical orientation toward what counts as language and how ELs learn (English) language. This lack of agreement in EL education presents challenges to establishing alignment between STEM content standards and ELP standards, although this alignment has been required by federal legislation since NCLB and has intensified with rigorous academic content standards in recent years.

Contemporary Views on STEM Subjects and Language From Theoretical Perspectives

From theoretical perspectives, there have been parallel shifts in STEM subjects and second-language acquisition. In STEM subjects, contemporary views emphasize that students make sense of phenomena and problems in the classroom community (knowledge-in-use), whereas traditional views have focused on learners’ mastery of discrete elements of content. In second-language acquisition, contemporary views emphasize that language is a set of dynamic meaning-making resources learned through participation in social contexts (language-in-use), whereas traditional views have focused on discrete elements of vocabulary (lexicon) and grammar (syntax) to be internalized by learners. Recognizing these instructional shifts as mutually supportive can promote rigorous STEM learning and rich language use with ELs (NASEM, 2018, pp. 63–66).

As ELs engage in STEM disciplinary practices (e.g., developing models, arguing from evidence, constructing explanations), they use language for the purpose of making meaning of STEM subjects through social interactions with peers and the teacher in the classroom community. The focus is on “what language does” (i.e., functional use of language for a purpose) more than “what language is” (i.e., structural elements of language, including vocabulary and grammar; Grapin et al., 2019). While communicating ideas with peers and the teacher, students use multiple modalities, including both the linguistic modalities (i.e., listening, speaking, reading, writing) and other semiotic modalities (e.g., embodiments, gestures, tables, graphs, diagrammatic or computational models). They draw on a variety of registers, ranging from everyday to specialized language. In addition, students participate in a range of different types of interactions (e.g., one-to-one, one-to-small group, one-to-many, small group-to-many). To communicate the growing sophistication of their ideas over the course of instruction, ELs use multiple modalities more strategically and increasingly more specialized registers to meet the communicative demands of interactions in pair, small-group, and whole-class settings (for an example in a science classroom, see NASEM, 2018, Box 3-1, pp. 64–65).

Because specialized registers afford the precision necessary to communicate STEM disciplinary meaning, they allow ELs to be more precise as their understanding becomes more sophisticated (Grapin et al., 2019). The mathematical practice of attending to precision is “not principally about formal or technical vocabulary” and “should not be interpreted as using the perfect word” (NASEM, 2018, pp. 77–78). Likewise, “precision goes beyond science vocabulary” (NASEM, 2018, p. 65). In both mathematics and science classrooms, “precision privileges disciplinary meaning by focusing on how students use language to engage in the STEM disciplinary practices” (NASEM, 2018, pp. 77–78) regardless of the linguistic features used. In other words, precision privileges disciplinary meaning while engaging in disciplinary practices beyond the precision of linguistic accuracy (see NASEM, 2018, p. 78, for a mathematics example and pp. 64–65 for a science example). ELs can communicate precise disciplinary meaning using less than perfect English. Again, the EL STEM report emphasizes that in the science (or any STEM) classroom, “language is a product of doing science, not a precursor or prerequisite for doing science and ELs need ample opportunities to do science” (NASEM, 2018, p. 65).

Traditional Instructional Approaches

Although the EL STEM report does not provide a specific list of less effective strategies or traditional instructional approaches, it describes prominent ones and contrasts them to contemporary views. First, it is a misconception that disciplinary vocabulary is disciplinary language. Although contemporary views recognize that disciplinary vocabulary is one key feature of disciplinary language when the vocabulary is used in context, STEM subjects and language go far beyond vocabulary. In STEM classrooms, ELs use language to engage in disciplinary practices, learn disciplinary content, and communicate disciplinary meaning. Through this engagement, ELs learn language as a product (NASEM, 2018, pp. 104–105, 117, 183, 298–299).

Second, there are misconceptions that “a certain level of English proficiency is a prerequisite to meaningfully engage in STEM learning” (NASEM, 2018, p. 2), that ELs “must be proficient in English before they can be successful in content area classes” (NASEM, 2018, p. 44), and that “for participation in STEM subjects, ELs first needed to have proficiency in the disciplinary talk—the words, vocabulary, or definitions” (NASEM, 2018, p. 98). These misconceptions have led to preteaching and frontloading of vocabulary (for critique, see NASEM, 2018, Box 6-4, pp. 186–187). Instead, contemporary views highlight a functional use of language in which “language is a product of interaction and learning, not a precursor or prerequisite” (NASEM, 2018, p. 98).

Third, in content-based language teaching, which has been most common until recently, ESL teachers are asked to develop “content objectives” and “language objectives” (for critique, see NASEM, 2018, Box 6-4, pp. 186–187). MacDonald, Miller, and Lord (2017) provided examples of typical language objectives that focus on grammatical forms (past tense) or a particular function out of context (“Compare . . . ”). Then they presented revised language objectives that focus on the science to be learned and the use of language as students develop a model to explain and predict (NASEM, 2018, pp. 96–97). Once again, contemporary views highlight using language while engaging in STEM disciplinary practices and learning language as a product.

Finally, sheltered instruction with ELs often provides “highly simplified content that seldom satisfies grade-level STEM content expectations” (NASEM, 2018, p. 38). This approach fails to meet the goals of content standards that are expected of all students, including ELs. Moreover, simplification of language coupled with simplification of content can have unintended consequences for ELs. For example, as cause and effect is a crosscutting concept across STEM subjects, shortening a sentence by eliminating words that establish a causal relationship (e.g., because, therefore) can actually make it more difficult for ELs to learn disciplinary content (Davison & Kantor, 1982). Contemporary views highlight amplifying language to support and challenge ELs with academically rigorous content.

Contemporary Instructional Approaches

Grounded in contemporary views on STEM subjects and language, ELs learn disciplinary content and communicate disciplinary meaning through engaging in disciplinary practices in the classroom community. As teachers explicitly focus on language in STEM education, they encourage ELs to draw on a full range of linguistic and cultural resources and strategically use multiple modalities and specialized registers to communicate disciplinary meaning across different types of interaction. As opposed to the traditional instructional approaches described previously, the EL STEM report identifies five promising instructional strategies (for details about each strategy, see NASEM, 2018, pp. 99–122):

These instructional strategies highlight two key features of contemporary thinking. First, the strategies begin with STEM subjects and then use language needed to learn STEM subjects. The strategies describe how to support and challenge ELs to engage in STEM disciplinary practices and discourse in interactions with peers and the teacher (the first and second bullets) while supporting them to use multiple modalities, specialized registers, other meaning-making resources, and knowledge of how language works in STEM disciplines (the third, fourth, and fifth bullets). Second, the emphasis on the functional use of language to learn STEM subjects in contemporary instructional approaches differs from the emphasis on vocabulary and grammar as a precursor or prerequisite to learn STEM subjects in traditional instructional approaches.