How Do I Know If a Dyslexia Program’s Results Are Real?

In January 2016, the Federal Trade Commission settled with the maker of Lumosity, a brain-game app that promised gains at school and at work. “Lumosity simply did not have the science to back up its ads,” the agency said.

Reading program sales pages make promises too, with rising charts, shiny badges and tearful videos. If one has you doubting before you pay, that doubt is good judgment. You don’t need a credential to test a program’s claims.

Why every program’s numbers go up

Here is the trap. A program tests a child in the fall, teaches for a few months, tests again, and shows you the jump. The jump is real. What it does not tell you is how much of it the program caused.

Three things push scores up with no program at all. The first is ordinary growth. Children’s reading improves across every school year, and fastest in the early grades. Hill, Bloom, Black and Lipsey (2008) mapped those normal yearly gains. So a big before-and-after jump in first grade says little on its own.

The second is a quirk of retesting called regression to the mean. A score on any one day is partly luck. A child picked because of a low score tends to score higher next time by chance alone. Dorothy Bishop makes this point in her guide Evaluating What Works. The worse a group starts, the bigger the gain you should expect on a retest.

This matters most for children showing signs of dyslexia. They are often the children a program picks for their low scores.

The third is practice. Children get better at a test they have already seen. Bishop’s fix for all three is the same: a comparison group of similar children who did not get the program. That group grows, drifts and practices too. Whatever gap opens between the two groups is the part the program earned.

“Evidence-based” has no referee

Schools and clinics are told to choose evidence-based programs. So who decides what counts? The answer is thinner than it sounds. There is no official body that certifies a program under ESSA, the federal education law. Its evidence tiers are standards, not awards. Some publishers hire a commercial research firm that both runs the study and hands out the tier badge. So an “ESSA Tier 1” seal on a sales page tells you less than it seems to.

A few other lines deserve the same second look.

• “Hundreds of studies.” A pile of studies is not the same as independent evidence. Many are before-and-after reports from school sites, with no comparison group. “Independent” sometimes means only that the company did not run the study. It does not mean the study was rigorous. That is a point about the kind of evidence, not about anyone’s honesty.
• “Built on the science of reading.” That describes how a program was designed. It is not a result.
• “Statistically significant.” That means a difference is unlikely to be down to chance. A significant difference is sometimes tiny. Ask how big it was, in plain words.
• “Brain scans show it works.” A change on a scan is not the same as better reading. Perdue and colleagues (2022) reviewed the brain imaging studies of reading programs. The studies did not agree, and pooled together they found no significant overall effect.
• “Matched to your child’s learning style.” Pashler and colleagues (2008) found the evidence did not justify matching teaching to learning styles. A 2024 review by Clinton-Lisell and Litzinger judged any benefit too small and too rare to be worth adopting. A program sold on this promise is selling the part two reviews did not back.

Some programs do clear the bar, and it is worth saying so. Lexia Core5, for one, has an independent evaluation by RAND behind it. And thin evidence is not the same as failed evidence. A low-cost program you run at home with thin evidence is a different bet from an expensive clinic intensive with the same thin evidence.

If you want to go deeper, here is a plain guide to judging a program’s evidence.

What to do this week: five questions to ask before you pay

Send these to the program or clinic by email, and ask for written answers. A company that has the evidence usually has the answers ready.

1. Who were the children compared with? Ask whether a similar group who did not get the program was tested at the same times. With no comparison group, you are looking at growth, chance and practice mixed in with whatever the program added.
2. Who ran the study, and who paid for it? Ask whether it was peer-reviewed. Ask whether anyone outside the company checked it.
3. What was measured? Ask whether the test was the program’s own or an outside reading test. A rising score on the program’s own drills is not the same as a child reading a new page better.
4. Which scores are you showing me? A percentile rank is a child’s place among children the same age. It is not the percent of answers right. Percentile ranks are not equal steps, either. They bunch up in the middle and spread out at the ends. So the same jump on a chart means a smaller score change near the middle than near the bottom. A big grade-level gain is the kind of score to ask an evaluator to explain. Ask for standard scores, before and after, with their ranges. Every score carries some error, and a small gain inside that range might be noise.
5. What do the independent registries say? Look the program up yourself on the What Works Clearinghouse and on Evidence for ESSA from Johns Hopkins. Listings change, so check the live page. “No qualifying studies” means no study passed their screen. It is not a finding that the program fails.

When a page leans on a video of one family, ask what happens for the typical child. US advertising rules expect an ad with a standout result to make clear what results people should generally expect. One family’s best story is real. It is not a study.

Ask these questions of us too. Learning Success sells a program, and no controlled study of it exists. When we share a parent’s words, read them for what they are: one family’s story, in their own voice.

You want your child reading with ease, and you want their months and your money to go somewhere real. The villain is not the teacher or the tutor. It is a system that tells everyone to pick evidence-based programs and then lets anyone print the badge. You are the one who gets to ask the hard question before you sign. Nobody will ever advocate for your child as hard as you will.

When reading is hard, we look at more than reading. The sounds, the page, holding words in mind. Our assessment asks about all of them. Start with All Access here.