A ten-year-old slid a worksheet across the kitchen table. “I don’t get what it’s asking,” she said.
Her mother looked at the problem. It was subtraction, the kind her daughter had done a hundred times on scrap paper. Nothing about the numbers was hard. The sentence around them was: “Determine the difference between the two quantities.”
Her mother tried again, in plainer words: “How much more is one than the other?” The girl answered in about three seconds.
That small gap, between a child who can’t do the math and a child who can’t read the question, sits underneath a lot of what gets labeled as falling behind in school.A Kitchen Table Standoff
Most homework standoffs start with the same assumption: the child is capable, the work is grade level, so the problem must be effort.
Photo by Focus harder. Read it again. Stop rushing.
But watch what happens when the question gets rephrased. A child who freezes on “determine the difference” often solves the same problem instantly once it becomes “how much more.” Her arithmetic didn’t change in those three seconds. Only the wording did.
This is worth sitting with, because the two situations look identical from the outside. A blank page and a frustrated kid look the same whether the block is mathematical or linguistic. Parents and teachers can’t tell them apart by looking. They can only tell them apart by asking.
The Hidden Language Barrier
Schools run on a way of speaking that almost nobody uses at dinner.
Photo by Educators call it academic language: the compressed, precise style of textbooks, worksheets, and test questions.
In plain terms, it’s the difference between “why do you think that” and “justify your reasoning.” Same request, different dialect.
A few of its habits show up everywhere:
-
Precise instruction verbs that each mean something specific: analyze, evaluate, summarize, justify
-
Connectors that carry the logic of a sentence: therefore, in contrast to, as a result
-
Compression, where a phrase like “the resulting effect” stands in for a whole cause-and-effect sentence
These words show up in every subject and get explicitly taught in almost none. Kids who hear this register at home pick it up the way they pick up an accent. Kids who don’t have to reverse-engineer it from context, question by question, while also learning the actual content.
Two Ways to Read a Low Score
Here the picture splits into two competing explanations, and both have honest supporters.
Photo by The first reads a low score as a signal about the student: not enough practice, not enough focus, not enough support at home. It’s intuitive, and sometimes it’s right.
The second reads the same score as a signal about the test’s language. A student may reason perfectly well and still misread what a question is asking. The numbers give this view real weight. In 2024, 67% of fourth- and eighth-grade English learners scored below the NAEP Basic level in math1. That’s math, not reading. A gap that large in a subject built on numbers points at least partly at the words surrounding them.
The pattern holds in the classroom too. Federal data from 2023 put average grade-retention rates for English learners in kindergarten through fifth grade at roughly 14% to 19%2. Meanwhile, fewer than 20% of multilingual English learners in the United States are taught through bilingual or dual-language approaches, according to a report on program access3. Most are learning content and the language of that content at the same time, with no bridge between the two.
What the Evidence Suggests
The fuller picture is less dramatic than either camp, and more useful.
Photo by Language isn’t the whole story, but it’s a bigger part of the story than most score reports admit.
A 2023 systematic review and meta-analysis of mother-tongue instruction found measurable benefits, including an effect size of 0.28 standard deviations for letter knowledge4. Those are modest numbers by everyday standards. For a general reader, it means letting kids build early literacy in the language they already think in gives them a real, repeatable head start, not a miracle cure.
UNESCO guidance points the same direction, placing multilingual instruction near the center of current thinking about foundational learning. The logic is simple: a child can’t learn to read while also guessing what the words mean.
And that’s exactly why academic language is teachable. It follows patterns, and patterns can be practiced.
Practices That Build the Bridge
None of this requires a curriculum or an extra hour at night. It just takes a habit.
- Before answering, ask the child to restate the question in their own words. If they can, the block is content. If they can’t, the block is language, and you’ve just found it in ten seconds.
- Let instruction verbs into ordinary conversation. Ask a child to justify choosing one snack over another, or to summarize the movie for someone who missed it.
- When a word stops them, unpack it together instead of supplying the answer. Say the plain version, then the school version, then move on.
Returning to the Table
The same worksheet, weeks later, produces a different conversation. Instead of “why don’t you know this,” the question becomes “what is this one asking you to do.”
That reframing changes who is stuck. The child isn’t failing a test of ability anymore. Parent and child are decoding a sentence together, a task with an obvious method and a clear end.
Equity, at this scale, looks unglamorous. It’s a parent slowing down long enough to notice that the obstacle was five words, not the arithmetic behind them.
The next time a child stalls on an assignment, it can help to hold off on explaining the content. Ask instead what the question is asking, in their own words. The answer tells you which problem you’re actually solving, and it takes about ten seconds to find out.
That worksheet on the kitchen table never changed. The numbers stayed the same, the difficulty stayed the same. What changed was that someone asked “how much more” instead of “determine the difference,” and a stuck child picked up her pencil.
