My bilingual student isn't making progress in reading: dyslexia or second language?
What the research actually allows a teacher to observe, and what to write down before anyone utters the word dyslexia.
The Bilingual Mind: Neuroscience of Literacy, by Andrés Marín-Palomar (teacher, speech-language pathologist and neuropsychologist). Now available in paperback and Kindle.
Also in Spanish: Mente bilingüe. Neurociencia y lectoescritura.
Outside the US? Search The Bilingual Mind or Mente bilingüe on your local Amazon.
The question every dual language teacher asks in March
It is the second week of March in a 90:10 dual language classroom in Texas. Camila is in second grade. She arrived in kindergarten speaking Spanish at home and almost no English. Her teacher has done everything right: explicit phonics in Spanish, daily decodable reading, small-group work. And still, Camila reads syllable by syllable, stops in the middle of long words, and when she is asked to read a list of made-up words she freezes. Her English, meanwhile, is coming along fine in conversation.
The teacher writes the question on a sticky note and leaves it on her desk for weeks:
That question is the most common one I hear from dual language teachers, and it is also, in its current form, impossible to answer. Not because the science is unclear, but because the question is built on a false alternative. This post explains why, what the research actually allows a teacher to observe, and what to write down before anyone utters the word dyslexia.
Why "dyslexia or second language" is the wrong question
The two options are not mutually exclusive. A child can be an emergent bilingual and have dyslexia; the two conditions do not cancel each other. Bilingualism does not cause dyslexia, and it does not protect against it. The base rate of dyslexia among bilingual children is, as far as the evidence shows, the same as among monolinguals.
The real question is different, and it is the one that opens the last chapter of Mente bilingüe:
Is this something that has not yet been taught, or something that is not working in the student?
Pedagogical deprivation, or disorder. The first is a gap in instruction: the child has not received enough explicit, systematic teaching of the code, or has received it in a language she barely understood, or changed schools three times. The second is a difference in how the brain processes the sounds of language, and it does not go away with more of the same instruction.
Those two things look identical from the outside in October. They stop looking identical when you know what to measure.
A child who reads badly because nobody taught her the code properly and a child who reads badly because her brain handles speech sounds differently make the same mistakes at first. The difference is not in the mistakes. It is in what happens when you teach them well: the first child catches up, the second one keeps struggling.
Clue one: decoding is the skill that second language learners do not lose
Here is the finding that should reorganize how teachers think about this. When researchers compare children learning to read in a second language with their native-speaking peers, the second language learners are far behind in oral comprehension and vocabulary, moderately behind in reading comprehension, and almost not behind at all in decoding.
A meta-analysis by Melby-Lervåg and Lervåg (2014), pooling dozens of studies, put numbers on it. The gap in listening comprehension between first and second language learners was large (d = −1.12). The gap in reading comprehension was medium (d = −0.62). The gap in decoding was small, close to negligible (d = −0.12).
Read that again from Camila's point of view. If her problem were "learning two languages", the pattern we would expect is this: she decodes reasonably well but does not understand what she reads, because the words are not in her vocabulary yet. That is the classic second language profile, and it is not a reading disorder; it is a language-exposure profile that improves with time and oral language teaching.
Camila's profile is the opposite. She understands stories read aloud to her in Spanish, her dominant language. What she cannot do is turn letters into sounds quickly and accurately, in Spanish, the language in which she has received the most instruction and has the most vocabulary. A decoding problem in the child's strongest language is not explained by bilingualism. It has to be explained by something else.
It seems that learning in two languages "confuses" a child's decoding and slows it down. The evidence says otherwise: decoding is the component of reading that second language status affects least. If a bilingual child's weakest point is decoding in her dominant language, the second language is not the explanation; it is the alibi.
Clue two: in Spanish, dyslexia is slowness, not errors
Most of what teachers in Texas have read about dyslexia was written about English. English has an opaque orthography: the same letters can sound many ways, and children with dyslexia make many errors reading words aloud. Screeners and checklists built in English look for those errors.
Spanish is a transparent orthography. Almost every letter has one sound. A Spanish-speaking child with dyslexia, after a year or two of instruction, can read most words correctly. What she cannot do is read them fast. Serrano and Defior (2008) showed this in Spanish children with dyslexia: their accuracy on words and pseudowords was close to normal, while their reading speed was dramatically slower than that of their peers. The deficit had not disappeared; it had changed its face.
This is not a Spanish peculiarity. Paulesu and colleagues (2001) compared adults with dyslexia in English, French and Italian. The Italian readers, with their transparent orthography, made far fewer errors than the English ones and in some tasks looked almost unimpaired. Yet all three groups were equally slow, equally weak in phonological tasks, and showed the same reduced activity in the same left-hemisphere reading region when scanned. The authors' title says it all: cultural diversity, biological unity. Same brain difference, different surface symptoms depending on the writing system.
The practical consequence for a dual language teacher is enormous.
She will score "within range" on accuracy and the file will say she is fine. The only way to see the deficit in Spanish is to time her: words per minute, syllables per minute, pseudowords per minute, compared with the norms for her grade and her language.
The idea that the "depth" of a writing system changes how reading difficulties show up is called the orthographic depth hypothesis. Ziegler and colleagues (2010) tested children in five European languages and found that phonological awareness predicts reading everywhere, but its weight is greatest in opaque orthographies like English and smallest in transparent ones like Finnish or Spanish, where speed measures carry more of the information. Chapter 7 of Mente bilingüe builds the whole Spanish phonological-awareness sequence on this finding, and Chapter 12 applies it to assessment.
Clue three: the core is phonological, and it travels across languages
The best-supported account of dyslexia today holds that its cognitive root is a difficulty in representing and manipulating the sounds of speech (Snowling, Hulme and Nation, 2020). Those representations are built from spoken language long before reading, and they are not language-specific in the sense that matters here: a child whose phonological representations are blurred in Spanish will carry that blur into English.
This gives the dual language teacher a tool that monolingual teachers do not have: two windows onto the same brain. If the difficulty is a phonological deficit, it should be visible in both languages, adjusted for how much instruction the child has had in each. If it appears only in English and Spanish is clean, the most likely explanation is exposure, not disorder. If it appears in Spanish, the child's strongest language, then the second language cannot be what is causing it.
That is why the book insists on assessment in both languages, with instruments built for each orthography, with time measured and not just errors. A test in one language leaves half the student outside the room.
Before you say the word "dyslexia", collect these five things, in writing:
- 1Timed reading in Spanish (the dominant language): words per minute and pseudowords per minute, against grade norms. Slowness with near-normal accuracy is the Spanish signature.
- 2Phonological awareness in Spanish, at the syllable and phoneme level. Spanish children normally master syllables early; persistent trouble segmenting and blending syllables in second grade is a red flag.
- 3The same measures in English, interpreted in the light of how many months of English literacy instruction the child has actually had.
- 4Listening comprehension in Spanish. If she understands text read aloud at grade level but cannot decode it herself, the problem is in the code, not in the language.
- 5Response to instruction over 8 to 12 weeks of explicit, systematic, small-group phonics in Spanish, with progress monitored every two weeks. Describe the instruction (what, how often, by whom) and the trajectory, not just the end point.
Deprivation or disorder: the criterion that decides
International classifications are explicit on one condition: a specific learning disorder cannot be diagnosed if the difficulty is better explained by lack of adequate instruction. That is not a loophole; it is the heart of the matter. Before anyone can say that the student's brain is the explanation, the school must be able to show that the teaching was not.
This is where the "response to instruction" criterion comes in, and it is good news for teachers. Hall and colleagues (2023) reviewed forty years of reading intervention research for elementary students with or at risk for dyslexia and found that well-designed, explicit interventions produce reliable gains. Most struggling readers respond. The child who receives that instruction, delivered well, for a sustained period, and still does not move at the expected rate is sending a signal that a single screening score cannot send.
Camila's teacher, then, is not stuck waiting for a specialist. She has twelve weeks of work ahead of her that will produce either a child who caught up or the most valuable document a referral can contain: a dated, specific record of what was taught, how, and what happened.
You cannot say a child has dyslexia until you have shown that good teaching was not enough. The teacher is the only person who can show that. The specialist can test the child; only the teacher can test the teaching.
What goes in the referral
The book devotes its final section to the teacher's referral protocol, and the central idea is that a referral is a functional record, not an opinion. The specialist does not need the teacher to say "I think it's dyslexia". The specialist needs to know:
- ✓What the child does when she reads, in each language, with examples (the actual words she stumbled on, the actual errors, the actual rate).
- ✓What has been tried, for how long, how often, in what group size, with what materials.
- ✓What changed and what did not, with dates.
- ✓Whether there is a family history of reading difficulties (one of the strongest known risk factors).
- ✓What the child understands when text is read to her, so that language comprehension and decoding are separated on paper.
A referral written like this moves a child through the system in weeks. A referral that says "struggles with reading, possibly due to being bilingual" sits in a drawer, and the child loses a year, sometimes two, because her bilingualism became the explanation that stopped everyone from looking further.
It seems prudent to "wait and see" with bilingual children, because a referral might label a child who simply needs more time in English. The opposite is true. Waiting costs more. The phonological deficit does not resolve with time, and every year of inadequate reading practice widens the gap in vocabulary, knowledge and self-concept. The cautious option is not waiting; it is measuring, in both languages, now.
Back to Camila
Camila's teacher timed her reading in Spanish: 31 correct words per minute, with only two errors, where her grade's norm is above 60. She timed the pseudowords: painful. She checked listening comprehension: at grade level. She asked the family: an older cousin "never learned to read well". She taught her, in a group of three, twenty minutes a day of explicit syllable-level work for ten weeks, and charted it. Camila gained seven words per minute. Her group-mates gained twenty-five.
None of that is a diagnosis. All of it is a referral that a school psychologist can act on in one meeting, because the differential between "not yet taught" and "not working" has already been done, in both languages, with numbers and dates.
That is what the book calls the teacher's intuition made into a certainty that guides help. It is also, in the end, what the right to learn means for a child like Camila: being taught the way she learns, and not in spite of how she learns.
What this post does NOT say
- It does not say that a teacher can diagnose dyslexia. Diagnosis requires a specialized evaluation. The teacher observes, documents and refers.
- It does not say that bilingual children are more likely to have dyslexia. There is no evidence for that.
- It does not say that slow reading in Spanish is always dyslexia. Slowness can also come from limited practice, attention difficulties, hearing problems or a language disorder. That is exactly why the record matters.
- It does not say that English screeners are useless with bilingual children. It says they are incomplete, and that they systematically miss the Spanish profile when they count only errors.
- The effect sizes quoted come from meta-analyses of heterogeneous studies; they describe averages across groups, not what any individual child will do.
Go deeper
Everything in this post is developed, with the primary studies and the classroom protocols, in Chapter 12 of Mente bilingüe, "Patologías de la lectura y la escritura": how to separate pedagogical deprivation from disorder (§12.1), what the phonological core looks like in the brain (§12.2), the clinical profiles and how they differ between Spanish and English (§12.3), the teacher's referral protocol, and the section on assessment in two languages that closes the book (§12.7). Reading fluency and why speed is the measure that matters in Spanish are in Chapter 9; the auditory processing behind the "she hears it wrong" complaint is in Chapter 5.
The Bilingual Mind: Neuroscience of Literacy — Andrés Marín-Palomar. Paperback and Kindle.
Outside the US? Search The Bilingual Mind or Mente bilingüe on your local Amazon.
If the book helps you with a student, a short review on Amazon helps the next teacher find it.
References
Hall, C., Dahl-Leonard, K., Cho, E., Solari, E. J., Capin, P., Conner, C. L., Henry, A. R., Cook, L., Hayes, L., Vargas, I., Richmond, C. L., & Kehoe, K. F. (2023). Forty years of reading intervention research for elementary students with or at risk for dyslexia: A systematic review and meta-analysis. Reading Research Quarterly, 58(2), 285–312. https://doi.org/10.1002/rrq.477
Melby-Lervåg, M., & Lervåg, A. (2014). Reading comprehension and its underlying components in second-language learners: A meta-analysis of studies comparing first- and second-language learners. Psychological Bulletin, 140(2), 409–433. https://doi.org/10.1037/a0033890
Paulesu, E., Démonet, J.-F., Fazio, F., McCrory, E., Chanoine, V., Brunswick, N., Cappa, S. F., Cossu, G., Habib, M., Frith, C. D., & Frith, U. (2001). Dyslexia: Cultural diversity and biological unity. Science, 291(5511), 2165–2167. https://doi.org/10.1126/science.1057179
Serrano, F., & Defior, S. (2008). Dyslexia speed problems in a transparent orthography. Annals of Dyslexia, 58(1), 81–95. https://doi.org/10.1007/s11881-008-0013-6
Snowling, M. J., Hulme, C., & Nation, K. (2020). Defining and understanding dyslexia: Past, present and future. Oxford Review of Education, 46(4), 501–513. https://doi.org/10.1080/03054985.2020.1765756
Ziegler, J. C., Bertrand, D., Tóth, D., Csépe, V., Reis, A., Faísca, L., Saine, N., Lyytinen, H., Vaessen, A., & Blomert, L. (2010). Orthographic depth and its impact on universal predictors of reading: A cross-language investigation. Psychological Science, 21(4), 551–559. https://doi.org/10.1177/0956797610363406
