Get access to over 1,300 articles and reader Q&As.
Much of what we get told about activities that impact cognitive development is not well supported by data. There is no evidence that you need to, for example, rotate your children’s toys. There is no compelling reason to think crawling impacts your child’s long-term ability to use a pencil. Even the oft-cited terrors of screen time are largely overrated.
When faced with this fact, parents often ask: What does matter? Is there anything that parents should definitely be doing? I’ve written more broadly on this, making the point that many of the most important things are the most basic: a stable place to live, enough to eat, and at least one supportive, loving adult. But among the individual behaviors, there is one that has some good data behind it: reading to kids.
How do we know that? Why are we more confident about the impact of reading than, say, the impact of screen time? To answer this, it’s useful to walk through the full set of evidence we have, to see how one builds a strong case for causality.
What does the data say about reading to kids?
The easiest way to study this question is to use data on parents reading to their children and connect it with measures of children’s cognitive development later on. There are several longitudinal (over time) and cross-sectional datasets that make this possible. The canonical summary of this literature is still a paper from 1995 that brings together many papers on the topic and shows a consistent relationship between reading aloud to children and later child outcomes. The effect size here is fairly large — the paper’s authors argue that 8% of the variation in outcomes can be explained by reading.
The problem with this evidence is similar to a lot of the issues I raise with evidence: it’s correlation, not causality. Parents who read more to their children differ in other ways (education, income) that might drive cognitive outcomes. It is hard, therefore, to draw any causal conclusions from this evidence alone. To establish causality, we need better data.
The impact of siblings on reading time
The first set of more plausibly causal evidence comes from data on siblings. In a 2018 paper, two researchers employ a clever design. They note that first children are read to more and that this difference is larger if births are spaced out more. Basically, kids get more time alone with their parents and, as a result, more reading.
The authors show that the gap in reading time and in child reading test scores between first- and second-born children is larger if the children are further apart in age. They use random variation in birth spacing — unfortunately, driven by some women having miscarriages before a second child — to reinforce their results. They also show that the effects are much smaller or not present for math scores, reinforcing the role of reading in language in particular.
This evidence is better than the correlations, but it’s not perfect. More one-on-one time with a parent means more reading, but it may mean more of other good things too. Those could drive the cognitive effects we see. This type of sibling study is one step toward causality, but not the gold standard.
How parent coaching makes a difference
The gold standard for evaluating causality in any treatment is a randomized controlled trial. In this type of trial, individuals are enrolled and some are randomly selected to receive a treatment. Because the selection is random, we can be confident that the groups are similar in other ways and that any differences we see after the intervention are due to the effects of the experimental treatment.
In studies of reading to kids, the treatment that parents are given is generally either support to read to kids, information about the value of reading, or access to books. Experimenters cannot force parents to read to their children, so they intervene to try to encourage this behavior. They then observe whether parents read more and whether there are positive outcomes for kids.
One large-scale example of this was run in Brazil and published in 2018. Researchers took 586 families in 12 child care centers. In half of the centers, they coached parents on reading to their children and provided access to books. Nine months after the intervention, they found that the children of the parents in the intervention group scored better on receptive vocabulary, working memory, and IQ.
This is one of many studies. A shorter but more intensive study in Hong Kong showed that explicit training in paired reading over a period of seven weeks improved child reading fluency. A very light-touch intervention in Paris — just providing parents information about the value of reading — showed impacts on vocabulary that persisted for at least six months.
Putting the literature together, a meta-analysis of 30 randomized controlled trials showed an average positive impact on children’s vocabulary. As that study notes, the size of this outcome is fairly small, although there are larger impacts for studies that focus on interventions where parents are encouraged to talk about books with their kids.
It is important to note that some of these studies are as much about providing access to books as they are about telling parents to read. A paper from February 2025 showed almost a 0.3 standard deviation increase in literacy skills (a sizable effect) when parents were given access to a digital library of books. This is a good reminder that, for many parents, literal access to books may be a barrier to reading.
How big are these effects? In the randomized trials, it is common to compare the effects with other interventions. These reading effects are similar in size to what we see in programs like Head Start, and they are slightly larger than some of the estimated impacts of teacher quality. But they are not as big as the impact of extensive one-on-one tutoring. Put more simply: the effects are meaningful but moderate.
Closing thoughts
Two key takeaways here.
First: This is an answer to the question parents ask me of what behaviors seem to matter for kids. Reading is great. I will caution — don’t take it obsessively. You do not need to chase your child around with books and hold them down to force-read to them. Try having books around, offering to read, and incorporating it into the bedtime routine.
The second implication is a policy lesson. For many families, the barrier to reading is book access. Digital libraries, or real libraries, are one solution. There are also programs, most notably Dolly Parton’s Imagination Library, that deliver free books to kids. Programs like this deserve support.
The bottom line
- In a paper that studies siblings, researchers find that first children are read to more and that this difference is larger if births are spaced out more. Basically, kids get more time alone with their parents and, as a result, more reading.
- In other studies of reading to kids, parents are either given support to read to kids, information about the value of reading, or access to books. Outcomes of these studies showed an average positive impact on children’s vocabulary.
- It is important to note that some of these studies are as much about providing access to books as they are about telling parents to read. For many parents, literal access to books may be a barrier to reading.
- As a general rule, having books around is a good thing. Offer to read to your child, and consider incorporating it into the bedtime routine.
Log in
Many thanks for this! One thing I was curious about: is there any evidence that reading in the very first months (say before 6 month old) has any benefit? “Read to your child from birth” (usually with the addendum “or even before”) is a recurring recommendation we got. While I don’t object–there’s little cost–I was struggling to imagine that someone made an RCT specifically designed to identify the benefits of reading in the first 6 months as opposed to from 6 months onward.
I’m in total support of reading to kids. I just don’t want that info or coaching presented to me hours after childbirth. I got an absolutely ridiculously huge pile of papers at discharge. All I could really process at that point was the basics of keeping a newborn alive. I was annoyed to spend my precious mental bandwidth on papers coaching me to read to my child. I will! Just maybe when I have enough strength to sit up! Maybe save the coaching for an early doctor visit.
Does it matter how much you read to your child? Our older child (18 months apart) had undiagnosed sleep apnea, so all he wanted to do was read instead of play. He started recognizing letters by two, and he started reading at 3. Our daughter, on the other hand, doesn’t know any letters even though she’s two and a half, but we read to both of them every night. We feel guilty that our daughter is behind in her cognitive development (although I still believe she’s in the normal range) – should we be reading to her more?!
I had the same question. When my child was an infant I read maybe 5-6 books to her throughout the day. Now she’s a very energetic toddler, and the daytime book reading is more haphazard. So only bedtime reading is consistent now. But after reading this article, maybe “only” at bedtime is okay after all?
I’m curious how important the one-on-one piece is. All of my memories of being read to as a child include my older brother (2-year age gap). Maybe the books were a little above my level or a little beneath his, but we both ended up highly verbal adults (probably driven by genetics too). We just had our second child, a 3-year age gap, and the best way for her to get reading time is alongside her older brother. I wonder if that’s just as good as his 3-year head start.
I was just wondering the same, because my first two are 18 months apart, and we typically read to them together at bedtime.