What the Research Says: How We Read the Evidence
This section will report what the studies we hold found. Our rules for each page: say who was studied and what was measured. Say what the study does not show. Name the study in the same sentence as the claim. Where we hold no evidence, say so. This page also shows you how to check a program's claim.
TL;DR
- This section will report studies in plain words: who was studied, what was measured, and what was found. Its pages are not written yet.
- Every research page here must say what its studies do not show. Read that part first.
- Our rule is to name the study in the same sentence as the claim. A claim of research with no name attached is not a citation.
- Where we hold no evidence on a question, our rule is to say so.
- Our rule is that a research page carries no sales pitch. Each must say whether a Learning Success program builds on its finding. If a program does, the page must say how.
Why this section exists
A parent looking into a learning difficulty meets two kinds of writing. One is the study itself. It is hard to read without training. The other is someone's summary. A summary can claim more than the study found.
This section is meant to be a third kind. It will report what was found, as strongly as it was found, in words you can use.
What every research page must have
Every research page here must follow the same shape. The shape is what lets you check it.
What the researchers looked at. Who was studied, how many, and what was measured. Here is an example. "Broc and colleagues (2025) studied 305 French children in grades 4 to 7. They compared handwriting with typing. None had a learning disorder. They measured spelling." A parent can follow that. It gives the group and the measure.
What they found. As strongly as the evidence allows, and no more. A solid finding is stated plainly, with no hedge. A finding with a limit carries the limit in the same sentence.
What this does not show. The limits. Read this part first. "This was measured in adults" is that kind of sentence. So is "nothing here is evidence about a child's brain." Sentences like these keep a real finding from being made to say more.
Why it matters for your family. The part you can act on.
Does Learning Success build on this? One of three answers applies. Yes: here is how the program uses it. Partly: here is the part we use. Or no: we do not build on it, and here is why. In no case do we claim a study's result for our program.
The studies. The full citation for every study the page names.
What is in this section
There are ten areas. Nine will hold pages on studies. The tenth will explain how our programs are designed. None of those pages is written yet.
- Reading. What the 2025 dyslexia definition says, and what it leaves out on purpose. Practicing a reading skill directly: what was tested, and in whom.
- Writing. What dysgraphia is, and why the definitions disagree. Why a child can say more than they can write. How strong the handwriting evidence is.
- Math. What dyscalculia is. Teaching math: what was tested, and in whom. What the math-anxiety brain scans measured, and what they did not.
- Movement and coordination. What coordination training improves. And whether movement improves reading or math.
- Attention and motivation. What the brain chemical dopamine does in learning. Why rewards can backfire. When a training gain may be expectation, not the training.
- Identity and mindset. How children see themselves, and how they do at school. Labels and praise. What happened when famous findings were tested again.
- Screens and technology. What has been measured about screens and school results. What school phone bans do. What learning technology teaches when it works.
- Programs and methods. How to read a program's claim about evidence. Then one page on each program our review covers. Each will say what was studied, if anything. Where there is a study, it will say who paid.
- Mental health. Learning difficulties and mental health: what the link is, and what it is not. How to read that kind of study.
- How Learning Success is built. The design choices behind our programs. These are facts about the design, not research findings. And where the evidence for them stops.
How we read evidence
Our rule is to ask four questions before we write a claim.
Was it measured? If a study measured it, our rule is to name the study in the same sentence. We give no effect sizes, the numbers for how big a change was. We do not turn one into a ranking either. "The largest measured impact" is the same claim without its limits.
Is it true because of how the thing is built? Some things need no study. Suppose a program shows each word letter by letter. The picture comes only once the word is complete. So guessing from the image is not available. That is a fact about the design. Hedging it would make it less accurate.
Do we see it repeatedly in the families we work with? In our view, a lot of useful knowledge is of this kind. It comes from years of practice. It has not been tested in a study with a comparison group. Our rule is to write it as ours. Our rule is never to present it as a research finding. Two mistakes are possible here. One is calling an observation a finding. The other is deleting an observation because nobody has studied it yet.
Or is it not established? Then our rule is to say so.
What we will not do here
No appeal to evidence without a name. A claim of outside support with no source is not a citation. It only looks like one. Our rule is to name the study in the same sentence as the claim. If we cannot, we write the claim as ours.
No claim of wider gains without a measure. Practice reaches the thing it is practice at. Some programs promise better reading or math from an activity that is not reading or math. Simons and colleagues (2016) reviewed brain-training programs sold on the promise of wider gains. They found good evidence for the trained task, less for closely related tasks, and little for distant ones. They also found little evidence that these programs make daily thinking better. Researchers still argue about whether some limited carry-over happens. We may describe a program like that. We do not call it evidence-based until a study measures the gain. Reading improves with reading instruction.
No claim about brain structure. Practice builds the pathways for reading. Only one study we hold measured the brain's structure. The change it found tracked only the narrow skill that was taught. Our rule is that no page says teaching changes the brain's structure.
No timeline of change. Our rule is that no page says how your child will change by a given week or month.
No selling. Our rule is that a research page carries no membership offer and no sales pitch. In our view, a research page that sells is harder to trust.
Subjects with no area of their own
Some subjects have no area of their own in this section. Sleep, memory, daily routines and executive function are among them. So are focus and attention span. The attention area covers motivation, not focus. Our guides on sleep and on executive function cover those two subjects.
How to read a program's evidence claim
In our experience, parents often meet evidence first on a program's own website. Four questions help.
Who paid for the study, and who ran it? A program tested by the people who sell it is not worthless. It is also not independent. Keep both facts in mind.
What was measured, and is it what you want? A program can improve the exact skill it trains and not change reading at all. That is not fraud. It is a real result. The marketing can make it sound bigger. Ask what changed, in what, and how it was measured.
Who was studied? Their age, how many there were, and whether they were like your child. A finding from a small group of adults is a finding about a small group of adults.
What does "no qualifying studies" mean? You may see it on What Works Clearinghouse, a US Department of Education review site. There, it means no study passed the review's screen for design. It does not mean the program was tested and failed. No evidence is not the same as evidence that it does not work. It is also not a recommendation.
Watch for a real but narrow finding used to back a broad claim. Something was measured. The claim on the homepage is about something else.
Key Takeaways
Ask four things of any claim. Was it measured? Is it true because of how the thing is built? Is it what people who do the work see again and again? Or is it not established?
A named source is stronger than an unnamed one, because you can check it.
Read what a study does not show before you read what it found.
Suppose a program promises better reading or math from an activity that is not reading or math. That is the claim to check.
Take the last claim you read about a program you are thinking about. Look for two things: who was studied, and what was measured. If the page says neither, you have not read a finding. You have read a sales line that looks like one.
”What to do tonightCommon questions from parents
Why would you report studies that do not help you sell?
Because they exist. A page that reports only helpful findings hides the rest. If we would not print an unhelpful result, our helpful ones would be worth little. It also means you can ask these questions of any program, ours included.
Is "evidence-based" a meaningful phrase?
Only with the evidence attached. Ask what was measured, and in whom. A program can be built on a real finding about one skill. It can then be sold on a result nobody has tested. The phrase covers both cases the same way.
There is no page here on the thing I searched for. Why?
The research pages are not written yet. Some questions may never get one, because we hold no study on them. The ten areas on this page say what this section will cover. For a practical question, the subject guides may help.
Do you write these pages from the studies or from summaries?
From our own records of each study. Each research page will carry the full citation, so you can find the study yourself. Some studies we know only through a review. Our rule is to say so when that is the case. Our rule is to quote a definition from the document that defines it.