Multilingual learners enter classrooms with language histories, cultural knowledge, literacy experiences, and subject-matter understanding that may be distributed across several languages. Their educational success depends on how schools develop the language of instruction while preserving access to mathematics, science, literature, history, and other intellectually demanding work. Effective instruction therefore integrates content learning and language development, evaluates knowledge using suitable evidence, and builds opportunities for students to participate in meaningful academic communication. The central research question is how teachers can make rigorous curricular knowledge accessible while students acquire additional linguistic resources. National Academies of Sciences, Engineering, and Medicine, 2017; Baker et al., 2014.
Who Are Multilingual Learners?
A multilingual learner is a student who learns, knows, or uses more than one language, with proficiency that can vary by language, setting, task, and stage of development. The term describes an educational and linguistic profile rather than a level of intelligence, a fixed cognitive type, or a clinical classification. Students may have learned languages simultaneously from early childhood, acquired an additional language after beginning school, or moved among schools with different instructional languages. The relevant classroom variables include the languages used at home and in previous schooling, current language exposure, literacy history, access to schooling, and the linguistic requirements of the curriculum. National Academies, 2017.
In the United States, English learner (EL) is a specific administrative and educational designation related to English proficiency and school services. Dual language learner (DLL) commonly describes young children developing two or more languages, especially before entry into K–12 school. Emergent bilingual and multilingual learner are broader educational descriptions, and the labels are not interchangeable with the official classifications used in federal or state reporting. A student may be multilingual and highly proficient in English, or may be classified as an English learner while already possessing sophisticated academic knowledge in another language. National Academies, 2017; U.S. Department of Education, 2016.
A classroom can contain children who speak two languages fluently, students with interrupted formal schooling, newcomers with extensive prior subject knowledge, and U.S.-born students who have long histories of using two languages in distinct settings. Language history, educational opportunity, socioeconomic conditions, migration-related disruptions, disability status, and instructional quality can all affect outcomes. Population-level findings should be interpreted with this heterogeneity in mind. National Academies, 2017.
The Core Educational Problem: Language Demand and Conceptual Understanding
Classroom achievement depends partly on what students know and partly on the language through which they are asked to demonstrate knowledge. A student may understand fractions, ecosystems, or historical causation yet struggle to parse a lengthy English prompt, interpret a metaphorical expression, or produce the precise disciplinary explanation expected by a rubric. The task then measures language proficiency together with the intended academic construct. Researchers and educators need to specify which demands are integral to the learning objective and which introduce incidental barriers. National Academies, 2017; Baker et al., 2014.
Language is also part of the content itself. Understanding a scientific claim requires recognizing words, sentence relations, definitions, and patterns of argument. In history, causal connectives, hedging, chronology, and evidence statements shape the interpretation of sources. In mathematics, comparative structures such as 'at least,' 'inversely,' and 'per unit' can alter the problem. Teaching content without its linguistic forms leaves a central part of disciplinary knowledge implicit. A productive classroom makes those forms visible during substantive inquiry. National Academies, 2017; Baker et al., 2014.
Conversational fluency and academic language
Students can become effective in familiar social conversations while still learning to interpret dense informational texts, generate evidence-based arguments, or compare competing explanations. The distinction between conversational and academic language describes differences in discourse demands, vocabulary, context, and schooling opportunities. It should not be treated as a fixed two-stage developmental law with a universal number of years attached to each stage. Academic language is acquired through guided participation in subject-specific practices; teachers must examine the particular texts, explanations, and assessment formats students encounter. Cummins, 1979; National Academies, 2017.
Fluency varies across tasks and languages
Multilingual proficiency is multidimensional: listening, discussion, narrative comprehension, disciplinary reading, formal writing, and word retrieval develop through different experiences. A learner may explain a mechanism aloud but need substantial support writing a laboratory report; another may read accurately without fully integrating the argument in an unfamiliar academic text. Teachers can plan from task-specific evidence instead of reducing a student to one global language label. National Academies, 2017.
How Languages Interact in Learning
Shared knowledge and cross-language relationships
Prior knowledge and literacy experiences can support learning through a new language. Familiar scientific concepts, text structures, and problem-solving strategies may be applied when the student learns the relevant English terminology. In reading research, relationships between first- and second-language skills are stronger for some foundational literacy processes than for broad oral language abilities. A meta-analysis of cross-linguistic associations found moderate-to-large relations for phonological awareness and decoding, with much weaker average relationships for oral-language skills and substantial variation across samples and measures. Because these are correlational findings, they do not prove that a specific instructional transfer intervention caused later achievement. Melby-Lervåg and Lervåg, 2011.
Cummins' linguistic-interdependence account helped establish the idea that development in one language can support development in another through shared knowledge and literacy-related capacities. Its historical role is substantial, while individual threshold formulations and predictions of generalized cognitive benefit remain theoretical propositions requiring separate tests. In contemporary instructional planning, the useful implication is to investigate the student's existing conceptual and literacy resources instead of assuming that every unfamiliar English word represents an unfamiliar idea. Cummins, 1979; Melby-Lervåg and Lervåg, 2011.
Distributed vocabulary and concept knowledge
A child may know an idea in one language and use another language for related but distinct experiences. This distribution can depress scores obtained from a single-language vocabulary test, even when the student has substantial conceptual knowledge. In a study of 247 Spanish–English bilingual children, Anaya, Peña, and Bedore compared monolingual and conceptual scoring approaches. Across-test conceptual scoring improved classification relative to monolingual scoring, but none of the scoring methods met a conventional 80-percent threshold for both sensitivity and specificity. The result supports multi-source assessment and cautions against making a diagnosis from isolated word-naming scores. Anaya et al., 2018.
Distributed knowledge is not a reason to disregard English development. A student must ultimately acquire the English vocabulary needed to understand assignments, participate in discussion, and write subject-specific explanations in an English-medium school. The instructional sequence can begin with an already-understood concept, make the English label and disciplinary relations explicit, and provide repeated occasions to use the new language. Baker et al., 2014; National Academies, 2017.
Code-switching and translanguaging
Code-switching describes alternation between languages or language varieties in discourse; translanguaging is used in educational research for pedagogical practices that mobilize a learner's linguistic repertoire during meaning-making. Classroom examples include brainstorming in a familiar language, comparing cognates or explanations across languages, collaborating before producing an English response, and consulting bilingual sources. These practices may increase access to concepts and participation, but their effects depend on implementation, population, outcome, and comparator. Huang and Chalmers, 2023.
A 2023 systematic review of pedagogical translanguaging in English-as-a-foreign-language settings included ten studies. Five favored translanguaging on reported outcomes; four of these five had a high risk of bias, and other studies yielded null or narrowly specific findings. The review supports cautious experimentation with clear learning objectives and outcome measurement. Its EFL samples should not be treated as direct proof of effectiveness for every U.S. K–12 multilingual classroom, where language exposure, immigration histories, and curriculum structures may differ. Evidence for universal superiority over English-only instruction is limited. Huang and Chalmers, 2023.
What Cognitive Science Contributes
Attention, working memory, and language processing
Comprehending instructional language consumes attention and working-memory resources. When vocabulary, syntax, and disciplinary concepts are all unfamiliar, students must coordinate more elements while solving the academic problem. Educators can reduce avoidable processing demands by presenting essential terms in context, using diagrams with clear verbal explanations, rehearsing conceptual relationships, and organizing tasks into manageable steps. These are principled instructional applications, and their effectiveness for a particular multilingual subgroup requires outcome-specific evaluation. National Academies, 2017; Baker et al., 2014.
Supports should maintain intellectual demand. Translating every sentence mechanically, replacing complex questions with trivial ones, or preventing students from hearing subject-specific English can narrow the opportunity to acquire disciplinary language. A carefully scaffolded task preserves the reasoning goal while supplying a bridge to relevant language forms. For example, a science teacher may model the causal frame 'As temperature increased, the rate ... because ...' and then require students to justify a real observation using the frame as a temporary resource. Baker et al., 2014.
The contested bilingual executive-function advantage
A claim that multilingualism automatically produces superior attention, inhibition, working memory, or general intelligence exceeds the evidence. Lowe and colleagues synthesized 1,194 effect sizes drawn from 10,937 bilingual and 12,477 monolingual children. The unadjusted overall executive-function effect was small (g = 0.08), and the estimated advantage was indistinguishable from zero after adjustment for publication bias (g = −0.04). This finding does not negate the value of multilingual competence; it locates educational benefits in demonstrable language, literacy, participation, and academic outcomes rather than in an assumed domain-general cognitive upgrade. The broader debate remains sensitive to task selection, sampling, socioeconomic factors, and analytic choices. Lowe et al., 2021.
The classroom implication is methodological. Evaluate language learning and curricular achievement directly rather than invoking an uncertain executive-function advantage as a rationale for particular programs. Lowe et al., 2021; National Academies, 2017.
Evidence-Based Principles of Classroom Instruction
1. Teach academic vocabulary intensively and in context
Academic words are powerful when their meanings, relations, morphology, and uses become available for repeated retrieval. The What Works Clearinghouse (WWC) 2014 practice guide gives a strong-evidence recommendation to teach a focused set of academic vocabulary words intensively over several days, using varied activities. A word such as 'estimate' should be connected to examples and nonexamples, short explanations, written uses, and the tasks in which estimation matters. Teachers should select terms with explanatory value instead of assigning isolated word lists unrelated to the current lesson. Baker et al., 2014.
Vocabulary instruction can include student-friendly definitions, visuals, morphology, semantic relationships, spoken rehearsal, reading in context, and short writing tasks. The aim is flexible use, not recognition of one definition. In a geography unit, 'migration,' 'population,' 'resource,' and 'distribution' become more memorable when students use them to interpret maps and justify competing explanations. Repeated exposure should include new contexts so that knowledge transfers beyond the original worksheet. Baker et al., 2014.
2. Integrate language instruction with subject teaching
The WWC gives a strong-evidence recommendation to integrate oral and written English-language instruction into content-area teaching. A science lesson can combine investigation with explicit teaching of comparison and causal structures; a history lesson can teach how evidence supports a claim; a mathematics lesson can teach how quantitative relationships are described. Language objectives should identify the actual discourse function learners need, such as explaining a procedure, interpreting evidence, or challenging an argument, while content objectives retain disciplinary rigor. Baker et al., 2014.
An illustrative sequence begins with an authentic problem, invites prior knowledge in any available mode, introduces core vocabulary and text features, models one disciplinary explanation, provides structured peer exchange, and culminates in independently assessed reasoning. This sequence operationalizes evidence-based principles; it is a teaching design rather than a single intervention whose complete bundle has one established universal effect size. National Academies, 2017; Baker et al., 2014.
3. Make disciplinary discussion extended and purposeful
Learners need more than brief responses to closed questions. Extended explanations, negotiated meanings, and discussion around texts give them opportunities to connect language with concepts. Educators can use roles, evidence prompts, think time, pairs, and small-group tasks to distribute participation. The language target should arise from reasoning: 'Which claim does the evidence support?' provides a different communicative demand from 'What word appears in line five?' National Academies, 2017.
Participation structures require monitoring. A teacher may notice that linguistically confident students dominate the discussion while newcomers produce only one-word contributions. Rehearsal time, visual references, purposeful turns, and follow-up questions can make the activity more equitable. Whether a particular discussion protocol improves long-term standardized achievement depends on the specific evidence, age, curriculum, and quality of implementation. National Academies, 2017.
4. Teach writing through models, planning, and revision
Academic writing joins conceptual knowledge with linguistic choices, genre conventions, organization, and revision. A multilingual writer may need examples of how a claim is introduced, evidence is embedded, and conclusions are limited. Teachers can analyze a model paragraph, collaboratively compose a response, provide a planning organizer, and gradually withdraw support. The WWC practice guide recommends regular structured opportunities to develop writing but rates the direct causal evidence for that recommendation as minimal in its 2014 evidence framework. This is a meaningful distinction between a plausible, educationally important practice and a strongly verified intervention effect. Baker et al., 2014.
Students should have chances to revise ideas as well as grammar. If feedback addresses only surface correctness, it may obscure errors in evidence, reasoning, and organization. An effective task specifies separate criteria for disciplinary accuracy and language control, while making the language appropriate to the stage of learning. National Academies, 2017.
5. Provide timely reading interventions when indicated
Multilingual students with reading difficulties should have access to rigorous, monitored instruction rather than waiting for fluent conversational English. The WWC assigns moderate evidence to small-group intervention for learners struggling with literacy and English development. A 2019 meta-analysis of 26 studies reported average effects of reading interventions on accuracy (d = 1.221), fluency (d = 0.802), and comprehension (d = 0.499). These outcome-specific pooled effects reflect different tasks and study contexts; they should not be interpreted as a guaranteed benefit of any one program for any one student. Baker et al., 2014; Ludwig et al., 2019.
Instructional response must be interpreted in context. A child who has not had the opportunity to learn a writing system is in a different position from a child who has received sustained high-quality instruction but continues to struggle across languages. Progress monitoring helps a teacher determine whether instruction is becoming accessible, and multidisciplinary assessment is appropriate when disability concerns persist. National Academies, 2017.
Classroom Scaffolding: A Worked Example
Consider a seventh-grade lesson about why plants in different environments develop different traits. The academic objective is to evaluate an explanation using observations. The language demands include 'adaptation,' 'environment,' 'variation,' 'survival,' and conditional and causal sentences. The teacher first elicits examples and prior ideas through images and discussion, including familiar-language explanations where feasible. Next, a compact reading or diagram introduces key concepts alongside English terminology. Students then compare two cases in pairs, explain the evidence orally, and produce an individual written account of the mechanism. Baker et al., 2014.
A student can demonstrate some conceptual understanding by accurately arranging evidence, selecting a diagram, or explaining the idea in a familiar language while continuing to develop the required English. This formative information helps the teacher choose the next language support. At the final assessment, the teacher makes clear whether the target is scientific reasoning alone, a combination of reasoning and academic English, or a specific language convention. Explicit construct definition makes both grading and feedback more interpretable. National Academies, 2017.
Scaffolding should change as learners become more independent. A frame can start as 'The evidence suggests ___ because ___' and later become an open invitation to evaluate competing interpretations. Teachers can monitor whether students transfer the reasoning and linguistic forms to unfamiliar phenomena; success on a practiced sentence alone is insufficient evidence of transfer. The related concept of generalizing learned knowledge is examined in Transfer of Learning: Why Knowledge Does or Does Not Generalize.
Bilingual Programs, Dual-Language Immersion, and English-Medium Approaches
Bilingual education encompasses many arrangements: transitional bilingual education, dual-language immersion, heritage-language maintenance, and content teaching through more than one language. Labels alone reveal little about teacher expertise, curriculum coherence, proportions of instructional time, or the opportunities students receive. Evaluating programs requires identifying the actual comparison group and the academic, linguistic, and longer-term outcomes. National Academies, 2017.
Slavin and Cheung's systematic best-evidence synthesis compared bilingual and English-only reading programs. Of 13 elementary studies involving Spanish-dominant students, nine favored bilingual approaches on English reading measures and four found no differences; the authors reported a sample-size-weighted effect of approximately 0.33. Study quality, local contexts, and program types matter in interpreting this older synthesis. Its conclusions cannot simply be extended to every language pair, secondary-school subject, or educational system. Slavin and Cheung, 2005.
A stronger causal design appears in Steele and colleagues' study of seven cohorts of dual-language immersion lottery applicants in one urban U.S. school district. Lottery winners showed positive intent-to-treat reading effects of about 0.13–0.22 standard deviations in fifth and eighth grades, with little evidence of a comparable benefit or harm in mathematics and science. The probability of remaining classified as an English learner in later grades was modestly lower for lottery winners. Lottery assignment strengthens causal interpretation within that district, while the particular program implementation and applicant population constrain generalization. Steele et al., 2017.
The combined evidence favors serious consideration of well-designed bilingual and dual-language programs. It does not establish that every multilingual program is effective merely because multiple languages are present. Program quality, continuity, staffing, language allocation, student selection, and outcome measurement remain central. National Academies, 2017; Steele et al., 2017.
Teaching Reading, Mathematics, Science, and History
Reading comprehension
Reading involves decoding, vocabulary, syntactic comprehension, background knowledge, and the integration of ideas across sentences. Some foundational reading skills show cross-language associations, but students also need explicit access to the words, structures, and discourse conventions of English texts. A lesson can prepare critical terms, activate relevant knowledge without preempting the text, model how to interpret one difficult passage, and ask students to justify an interpretation from textual evidence. Outcome evaluation should separate word-reading progress from comprehension of increasingly demanding material. Melby-Lervåg and Lervåg, 2011; Baker et al., 2014.
Mathematics
Mathematical reasoning and language comprehension interact when students parse word problems, compare quantities, and justify procedures. Visual models, symbolic notation, and worked examples can make some relations available before English terminology is fully established; teachers should subsequently connect those representations to the required mathematical language. The same student might solve a symbolic equation correctly and misunderstand a problem phrased with 'fewer than.' Teachers can use this contrast diagnostically to adapt instruction rather than equating one incorrect answer with absent mathematical knowledge. National Academies, 2017.
Science
Scientific learning asks students to name phenomena, make predictions, interpret data, and defend causal explanations. The language of uncertainty, comparisons, and evidence can be taught while students investigate meaningful content. In a quasi-experimental science-and-literacy intervention involving fifth-grade English learners, Lara-Alecio and colleagues reported improved performance on several science and reading measures. The multicomponent design combined professional development, inquiry, explicit vocabulary, literacy integration, and enrichment; its effects cannot be assigned to any one component without additional comparisons. Lara-Alecio et al., 2012.
History and social studies
Historical reasoning demands the interpretation of source perspective, temporal sequence, causal explanations, and differences between evidence and interpretation. A multilingual classroom can make these discourse functions explicit through paired sources, timelines, vocabulary for uncertainty, and structured discussion. Students' cultural and linguistic experiences can enrich interpretations, provided the curriculum invites evidence-based inquiry rather than placing individual students under pressure to represent an entire culture. The direct experimental literature on each specific history teaching routine is narrower than the overall evidence for integrating language and subject-matter instruction. National Academies, 2017.
Assessment: What Does a Test Actually Measure?
Assessment is strongest when teachers define the construct and record the language load. In a geography task, the intended construct may be spatial reasoning; in an English-language development task, precision of English expression is itself an outcome. Students can supply drawings, worked calculations, short discussions, bilingual notes, or written arguments as appropriate to that construct. Such options are assessment-design choices whose validity depends on comparability and the purpose of the score. National Academies, 2017.
Multiple languages, multiple measures
A single English vocabulary score can underestimate language knowledge distributed across languages. Anaya and colleagues' classification study demonstrates both the added information available from conceptual scoring and the limitations of treating it as a diagnostic shortcut. Assessors should consider educational history, caregiver and teacher observations, language exposure, performance across languages, progress over time, and the psychometric suitability of each instrument. Translated test items do not automatically retain the difficulty or validity of the original assessment. Anaya et al., 2018; National Academies, 2017.
Growth, classification, and long-term outcomes
Language proficiency scores, grade-level test performance, classroom participation, and graduation or course completion measure related but different outcomes. Improvement in one does not guarantee improvement in another. Reclassification out of EL services may reflect actual proficiency development, policy thresholds, or testing practices; it should be interpreted alongside subsequent academic progress. Longitudinal designs, stable measures, and attention to students moving between schools improve interpretation. National Academies, 2017; Steele et al., 2017.
Multilingual Development and Learning-Disability Identification
A need for additional English instruction is an educational language-development need. Dyslexia, developmental language disorder, intellectual disability, and other diagnoses require their own validated criteria and specialist evaluation. A teacher who sees slow English reading growth should examine opportunity to learn, prior literacy instruction, language exposure, reading processes, and progress under appropriate teaching before making inferences about a disorder. At the same time, language-learning status should never be a reason to defer indicated educational support indefinitely. National Academies, 2017, chapter 10.
The National Academies reviewed the risks of both misidentifying language differences as disabilities and failing to identify genuine support needs among English learners. Assessment requires converging evidence and a culturally and linguistically informed team. Neither speaking multiple languages nor switching between them constitutes a symptom of a psychiatric or neurodevelopmental disorder. Diagnosis-specific pathways remain the responsibility of their respective assessment and clinical specialists. National Academies, 2017; Anaya et al., 2018.
Universal classroom access includes predictable routines, clear task goals, varied demonstrations of understanding, and responsive feedback. These approaches support equitable learning and can coexist with individualized accommodations when indicated. The efficacy of a universal strategy and the validity of a diagnosis remain separate questions. National Academies, 2017.
Participation, Belonging, and Family Partnerships
Belonging in the classroom affects who participates, seeks clarification, volunteers explanations, and gains opportunities to practice academic language. Respect for pronunciation variation, names, family linguistic practices, and alternative ways of participating can improve the learning environment. Multilingual students should be able to take on demanding intellectual roles instead of being positioned repeatedly as translators or passive recipients of simplified assignments. National Academies, 2017.
Families can share information that standard school records miss: previous schooling, literacy experiences, valued language practices, historical interruptions, and subject strengths. Two-way communication and qualified interpretation facilitate a more accurate understanding of students' needs. Family participation should be designed around realistic access to communication rather than presuming all caregivers are fluent in English or available at identical times. National Academies, 2017; U.S. Department of Education, 2016.
Cultural responsiveness requires more than decorative representation. Teachers can choose substantive texts with multiple perspectives, examine how assumptions influence classroom interactions, and invite students to connect prior knowledge to the curriculum while maintaining evidence-based evaluation of claims. The present article addresses language access and multilingual cognition; a full evaluation of culturally responsive teaching as a broader educational intervention has its own independent scope. National Academies, 2017.
Technology, Translation, and AI-Assisted Learning
Bilingual glossaries, captioning, speech-to-text, text-to-speech, dictionaries, and accessible digital texts can reduce some immediate participation barriers. Their value should be judged by the actual task: a translation may help a learner understand a scientific concept, while an inaccurate translation of a technical term can introduce an error. Teachers should choose technologies in ways that foster increasing independent command of the target academic language and subject knowledge. National Academies, 2017.
Generative AI can produce multilingual explanations, examples, or draft feedback, but its outputs may mistranslate terminology, invent citations, collapse culturally specific meanings, or conceal a learner's misunderstanding. In a classroom, a generated answer is an object for checking against a trusted source and a clear rubric, not evidence of what a student independently knows. Schools should protect students' identifiable work, language profiles, and assessment data, especially when external services store or reuse submissions. These are design and governance requirements; claims that AI-supported translation or tutoring improves long-term multilingual achievement need intervention-specific evidence. The broader interdisciplinary questions of educational AI are covered in Education in the Age of AI: Learning, Motivation, Assessment, and Cognitive Development.
For homework, technology can lower the language barrier to an assigned concept while simultaneously creating opportunities for answer substitution. Educators can distinguish an initial translated explanation, a student's own intermediate reasoning, and a final independently produced response. The grading policy should state whether linguistic independence, scientific understanding, or both are being evaluated. This is a practical inference from construct-valid assessment principles, not a claim that any particular AI product has demonstrated learning gains. National Academies, 2017.
Implementing a Schoolwide Multilingual-Learning Approach
A school can begin by mapping the language requirements of core subjects and comparing those requirements with what students are taught explicitly. Teachers can review representative text passages, typical oral explanations, writing assignments, and assessment prompts. An effective plan identifies the language forms necessary for participating in genuine disciplinary work, supplies professional learning for teachers, and monitors both language growth and academic progress. Baker et al., 2014; National Academies, 2017.
Instructional coordination matters because students may move between designated language-development classes and regular subject classrooms. A common instructional vocabulary and a shared understanding of the content-language relationship can reduce fragmented expectations. The U.S. Department of Education's English Learner Toolkit addresses agency responsibilities and the provision of appropriate educational services. Jurisdiction-specific legal requirements should be checked against applicable current rules rather than inferred from a generic pedagogical guide. U.S. Department of Education, 2016.
A useful improvement cycle defines the target outcome, records a reasonable baseline, implements an instructional approach with attention to fidelity, examines language and subject outcomes separately, and revises teaching from the evidence. Schools should look for subgroup differences by grade, prior literacy, initial proficiency, and instructional context. Average improvement may conceal weak outcomes for newcomers, students with interrupted schooling, or learners requiring additional reading support. National Academies, 2017.
What the Evidence Establishes and Where It Remains Limited
Established educational guidance supports purposeful academic vocabulary instruction, integration of oral and written language into content learning, and timely targeted reading support. These recommendations are supported by WWC evidence reviews that specify different evidence levels across practices. Individual trial results and meta-analyses provide additional support but apply to particular ages, languages, comparators, and outcome measures. The strength of the inference is greatest when research identifies a relevant comparison and measures actual learning instead of only satisfaction or immediate task completion. Baker et al., 2014; Ludwig et al., 2019.
Evidence is promising but context-dependent for well-implemented dual-language immersion. Lottery-based research improves causal identification, yet results from one district are neither a universal estimate nor a substitute for evaluating local staffing and curriculum. Broader bilingual program syntheses support considering first-language resources while showing differences among interventions and populations. Steele et al., 2017; Slavin and Cheung, 2005.
Evidence is limited or contested for claims that any translanguaging strategy automatically outperforms alternatives, that all bilingual students gain a general executive-function advantage, or that digital tools produce durable transfer merely by generating explanations. Risk of bias, publication bias, inconsistent definitions, small samples, and a narrow range of languages can change conclusions. A transparent evidence review asks who was studied, what comparison was used, how long outcomes were followed, which language and subject skills were measured, and whether results replicated in regular classrooms. Huang and Chalmers, 2023; Lowe et al., 2021.
The overarching standard of success is sustained access to curriculum, demonstrable learning, increasing communicative agency, and educational opportunity across time. Narrow English test performance, although informative for its intended purpose, cannot alone summarize all those outcomes. National Academies, 2017.
Frequently Asked Questions
Should a student use a home language while learning English at school?
Home-language resources can help students interpret unfamiliar concepts, preserve access to existing literacy and knowledge, and participate in problem-solving. Teachers should also plan explicit opportunities to understand and use the English language forms required by the curriculum. Experimental evidence on individual translanguaging routines is mixed and context-specific, so practice should be evaluated by actual subject and language learning. Huang and Chalmers, 2023; National Academies, 2017.
Does speaking several languages make a student more intelligent?
Multilingual competence is a substantial linguistic and educational resource, while a generalized advantage in executive function has not been consistently demonstrated. A large meta-analysis found that a small unadjusted average advantage disappeared after publication-bias adjustment. The appropriate educational outcomes are language competence, academic achievement, well-being, and access to opportunity. Lowe et al., 2021.
How can teachers tell whether a student understands the lesson?
Ask learners to demonstrate the target idea in multiple appropriate modes, compare performance across language-heavy and less language-heavy tasks, discuss thinking processes, and examine work over time. Interpret the result in light of the assessment's intended construct. A low English vocabulary score does not necessarily imply a corresponding lack of conceptual knowledge. Anaya et al., 2018; National Academies, 2017.
Should reading intervention wait until a student becomes fluent in English?
Reading support should respond to observed needs and available evidence. The WWC recommends targeted small-group interventions, and a 2019 meta-analysis found positive average outcomes across reading domains. Progress should be monitored, and concerns about a possible learning disorder should be evaluated with appropriate multilingual and educational evidence. Baker et al., 2014; Ludwig et al., 2019.
Is bilingual instruction always better than English-only instruction?
The program label cannot answer that question. Some bilingual-reading syntheses and a lottery-based dual-language immersion study report favorable reading results, but magnitudes and other outcomes vary by program and population. A sound comparison examines instructional quality, grade, language pair, subject achievement, and follow-up outcomes. Slavin and Cheung, 2005; Steele et al., 2017.
Conclusion
Multilingual learning succeeds when academic content and language growth are planned together. Students bring knowledge and linguistic resources that extend beyond any one standardized English score. Educators can develop academic vocabulary, teach disciplinary discourse during meaningful tasks, use multilingual prior knowledge, provide timely literacy interventions, and assess the intended construct with appropriate measures. The strongest evidence supports deliberate instructional design and responsive assessment; broader claims about cognitive superiority or universally effective language policies require narrower, better specified conclusions. National Academies, 2017; Baker et al., 2014.
Related Articles
Cluster Learning and Educational Psychology — The published parent scientific chapter situating multilingual instruction within learning theory, inclusion, teaching, and assessment.
How to Study Effectively: Evidence-Based Learning Strategies — General evidence on independent study techniques that can complement language-aware instruction.
Learning Styles: Evidence on Matching Teaching to Learner Preferences — A critical review of learner-preference matching claims, distinct from multilingual accessibility.
Transfer of Learning: Why Knowledge Does or Does Not Generalize — The research on carrying learned concepts and strategies into new tasks and contexts.
Education in the Age of AI: Learning, Motivation, Assessment, and Cognitive Development — Interdisciplinary analysis of AI, student learning, and the interpretation of educational outcomes.
Supporting Neurodivergent Learners: Universal Teaching Strategies — Review inclusive classroom strategies and access while reserving diagnosis-specific care for specialist clusters.
Universal Design for Learning: Framework, Evidence, and Limitations — Explain UDL principles, inclusive access, implementation, and the strength and limits of outcomes research.
Culturally Responsive Teaching: Evidence and Educational Equity — Explore culturally responsive teaching, evaluation, educational equity, context, and measurement challenges.
References
Anaya, J. B., Peña, E. D., and Bedore, L. M. (2018). Conceptual scoring and classification accuracy of vocabulary testing in bilingual children. Language, Speech, and Hearing Services in Schools, 49(1), 85–97. https://doi.org/10.1044/2017_LSHSS-16-0081.
Baker, S., Lesaux, N., Jayanthi, M., Dimino, J., Proctor, C. P., Morris, J., Gersten, R., Haymond, K., Kieffer, M. J., Linan-Thompson, S., and Newman-Gonchar, R. (2014). Teaching academic content and literacy to English learners in elementary and middle school. What Works Clearinghouse, Institute of Education Sciences. https://ies.ed.gov/ncee/wwc/PracticeGuide/19.
Cummins, J. (1979). Linguistic interdependence and the educational development of bilingual children. Review of Educational Research, 49(2), 222–251. https://doi.org/10.3102/00346543049002222.
Huang, X., and Chalmers, H. (2023). Implementation and effects of pedagogical translanguaging in EFL classrooms: A systematic review. Languages, 8(3), 194. https://doi.org/10.3390/languages8030194.
Lara-Alecio, R., Tong, F., Irby, B. J., Guerrero, C., Huerta, M., and Fan, Y. (2012). The effect of an instructional intervention on middle school English learners' science and English reading achievement. Journal of Research in Science Teaching, 49, 987–1011. https://doi.org/10.1002/tea.21031.
Lowe, C. J., Cho, I., Goldsmith, S. F., and Morton, J. B. (2021). The bilingual advantage in children's executive functioning is not related to language status: A meta-analytic review. Psychological Science, 32(7), 1115–1146. https://doi.org/10.1177/0956797621993108.
Ludwig, C., Guo, K., and Georgiou, G. K. (2019). Are reading interventions for English language learners effective? A meta-analysis. Journal of Learning Disabilities. https://doi.org/10.1177/0022219419825855.
Melby-Lervåg, M., and Lervåg, A. (2011). Cross-linguistic transfer of oral language, decoding, phonological awareness and reading comprehension: A meta-analysis of the correlational evidence. Journal of Research in Reading, 34(1), 114–135. https://doi.org/10.1111/j.1467-9817.2010.01477.x.
National Academies of Sciences, Engineering, and Medicine. (2017). Promoting the educational success of children and youth learning English: Promising futures. National Academies Press. https://doi.org/10.17226/24677.
Slavin, R. E., and Cheung, A. (2005). A synthesis of research on language of reading instruction for English language learners. Review of Educational Research, 75(2), 247–284. https://doi.org/10.3102/00346543075002247.
Steele, J. L., Slater, R. O., Zamarro, G., Miller, T., Li, J., Burkhauser, S., and Bacon, M. (2017). Effects of dual-language immersion programs on student achievement: Evidence from lottery data. American Educational Research Journal, 54(1_suppl), 282S–306S. https://doi.org/10.3102/0002831216634463.
U.S. Department of Education, Office of English Language Acquisition. (2016; resource page reviewed 2025). English Learner Toolkit for State and Local Education Agencies. https://www.ed.gov/teaching-and-administration/lead-and-manage-my-school/state-support-network/ssn-resources/english-learner-toolkit-for-state-and-local-education-agencies-seas-and-leas.