How To Help Ell Students In Math

The Actual Problem With EL Students in Math

Most people think the barrier is language. It is not always language. The barrier is usually that math teachers and ELD teachers exist in separate buildings at most schools, and nobody thought to figure out how they overlap. I ran into this last year with a 9th-grade algebra class that had four EL students across three different proficiency levels. One student could solve linear equations in their native language but froze the moment I wrote "solve for x" on the board. Another student could read the questions but kept subtracting instead of adding because the word problem structure was unfamiliar. A third one was actually bilingual in math—she understood the concepts perfectly—but her English writing was so halting that she lost points on process work anyway. The fix was not to dumb down the math. It was to separate the language demand from the math demand and then slowly merge them back together.

How To Help Ell Students In Math Without Losing the Rigor

The first thing you need to do is audit every math assignment for hidden language tasks. When you write "Explain your reasoning," that is an essay prompt disguised as math. EL students are often doing dual translation in their heads—converting the problem from English to their conceptual understanding, solving it, then converting their answer back to English for the write-up. That is three times the cognitive load. Most of them will default to just guessing the answer or staying silent because the task feels too heavy. I started using sentence frames that were actually useful instead of the generic "First I ___, then I ___" templates. Something like "I know ___ because ___" or "The answer is ___ since the problem says ___." These frames gave them a structure to plug math vocabulary into without having to build a whole sentence from scratch. It cut my grading time down significantly because I stopped seeing ten different ways to say the same correct thing. Visuals matter, but not in the way most teachers use them. I am not talking about putting a picture next to every word problem. That is decorative. I mean building visual models that exist before the words ever enter the room. Bar models for addition and subtraction. Number lines for integer operations. Area models for multiplication and factoring. The visual gives them a non-linguistic anchor to hold onto while their English catches up.

Here is the counter-intuitive part that nobody talks about: sometimes the best move is to let them work in their home language first, then translate. I had a student who could not parse a ratio problem in English at all. I let her explain it to a partner in Spanish. She got the right answer immediately. Then we walked through the English version together. She was frustrated when I first suggested it, but once she saw that the math had not changed, only the words had, her anxiety dropped noticeably. The rule is simple—let them access the math in whatever language gets them there, then scaffold the English on top. Another thing that trips people up is the assumption that vocabulary instruction means a word wall. Word walls are fine for reference, but they do nothing if students do not encounter the words in context. I started doing five-minute warm-ups where we would take one key term—"difference," "quotient," "at least," "no more than"—and write three real examples on the board before anything else. Not definitions. Examples. You learn what "difference" means by seeing 15 - 7, by seeing "the difference between twelve and four," and by seeing it used incorrectly on purpose so students can spot the error. That last part is the one most people skip. Deliberately writing a wrong interpretation and asking the class to find it forces engagement without requiring perfect English production. Assessment is where EL students get burned the hardest. If you give a multiple-choice test with dense word problems, you are testing reading comprehension as much as math ability. I switched to giving the same problems with simplified language options on the side. Same numbers, same questions, just clearer sentences. It kept the math rigorous while removing the language noise. Test scores for my EL students went from averaging 52 percent to 74 percent in one semester. That is not a perfect fix, and it is not fair to other students if you only do it for some, but it is better than letting language be the gatekeeper.

The real bottleneck is time. You cannot slow down the entire class to re-teach vocabulary for three students every day. What works instead is peer pairing. Put an EL student with a supportive peer who can model the language without explaining the math. The peer does not need to be advanced. They just need to be comfortable saying things out loud. I found that having students explain their steps to each other in pairs before sharing with the whole class reduced the pressure dramatically and actually improved comprehension for everyone, EL or not. I also stopped correcting grammar during math class unless it changed the mathematical meaning. "I add the numbers together" is fine. "I adds the numbers together" is fine too, as long as the operation is correct. You are teaching math, not writing. If you correct everything, you slow them down and make them afraid to participate. Save the grammar lessons for English class. The one scenario where all of this fails is when the student is severely below basic English proficiency and has had interrupted formal education. No amount of sentence frames or visual models replaces foundational literacy. Those students need a different intervention, usually pull-out support focused on basic academic English before they can tackle grade-level math with any real access. Acknowledging that limit early saves everyone time.