What is reading comprehension?

Decoding and comprehension are two different jobs

Decoding is the work of turning printed symbols into spoken words. A child who has learned that the letters s, h, i, p make the word ship has decoded it. Decoding can be taught directly, it improves with practice, and it becomes automatic for most readers somewhere between grade 2 and grade 4.

Comprehension is the work of turning those words into meaning. It asks what the sentence is saying, how it connects to the sentence before it, what the writer has left unstated, and what the whole passage adds up to. Comprehension does not become automatic. A capable adult reader still fails to comprehend a dense legal contract or an unfamiliar medical paper, despite decoding every word on the page without effort.

The practical consequence for a homeschooling parent is that fluent oral reading is not evidence of understanding. A child can perform a page beautifully and retain almost nothing from it. Listening to a child read aloud tells you about decoding. Asking the child to tell you what happened and why tells you about comprehension.

The two halves of skilled reading

The most widely used model in reading research is the Simple View of Reading, proposed by Philip Gough and William Tunmer in 1986. It states that reading comprehension is the product of decoding and language comprehension rather than their sum. Because the relationship is multiplicative, a near zero in either capacity produces a near zero result. A child with excellent decoding and weak language comprehension will not understand what they read, and neither will a child with strong language comprehension who cannot get the words off the page.

Hollis Scarborough's Reading Rope, published in 2001, expands the same idea into the individual strands that have to be woven together. Word recognition is made of phonological awareness, decoding, and sight recognition of familiar words. Language comprehension is made of background knowledge, vocabulary, language structures, verbal reasoning, and literacy knowledge. The word recognition strands become increasingly automatic. The language comprehension strands become increasingly strategic, and they keep developing well past the point where decoding is finished.

The Rope explains why comprehension problems rarely have a single cause. A child struggling to understand a chapter book might be missing vocabulary, might lack the background knowledge the author assumed, or might not yet follow the sentence structures literary prose uses. These are different problems with different remedies, and they are not fixed by more decoding practice.

Word recognition

Phonological awareness, decoding, and sight recognition. Becomes automatic with practice, usually well established by around grade 3.

Language comprehension

Background knowledge, vocabulary, language structures, verbal reasoning, and literacy knowledge. Keeps growing for as long as a reader keeps reading.

Why background knowledge decides so much

One of the most quoted studies in reading research was run by Donna Recht and Lauren Leslie in 1988. They gave middle school students a passage about a baseball game and sorted the students two ways: by measured reading ability and by how much they knew about baseball. Knowledge of baseball predicted comprehension of the passage better than measured reading ability did. Weak readers who knew baseball outperformed strong readers who did not.

The finding generalises. Comprehension is not a free floating ability that transfers evenly across every text. It is domain sensitive. A reader understands a passage about the Roman Republic, a passage about tide pools, and a passage about a family argument to the degree that the reader already knows something about republics, tide pools, and families.

This has a direct implication for how a reading curriculum should be built. A sequence of disconnected passages chosen only for reading level builds very little knowledge, because nothing accumulates from one week to the next. A sequence of whole books read over time builds a great deal, because each book leaves the reader with people, places, vocabulary, and situations that make the next book easier to understand.

Vocabulary and comprehension grow together

Vocabulary knowledge is among the strongest predictors of reading comprehension, and the relationship runs in both directions. A reader who knows more words understands more of what they read. A reader who reads more encounters more words in context and learns them.

The important detail is where the words come from. Most vocabulary is not learned from definitions or word lists. It is learned incidentally, from meeting a word repeatedly in contexts that make its meaning gradually clear. A child who reads widely meets tens of thousands of words this way. A child who reads little does not, and the gap between the two widens each year.

This is one reason reading level matters less than it appears to. A book slightly above a child's independent level, read with support or read aloud, exposes the child to language they would not otherwise meet. A book comfortably at their level builds fluency but stretches vocabulary less.

The three levels of understanding

Comprehension questions are usually described as working at three levels, and a child can be strong at one and weak at another. Knowing which level a child is struggling at is more useful than knowing that comprehension in general is weak.

Literal

What the text says directly. Who did what, where, and when. A child answers by locating information that is present on the page.

Inferential

What the text implies but does not state. Why a character acted, what a character feels, what is likely to happen next. A child answers by combining what is on the page with what they already know.

Evaluative

What the reader makes of the text. Whether an argument holds, whether a character was right, how this book compares to another. A child answers by judging the text against something outside it.

Why comprehension is not a skill you can drill

A great deal of instructional time is spent practising named comprehension strategies: finding the main idea, making predictions, summarising. The research on these strategies is not that they are useless. It is that they are quickly learned and reach their ceiling fast. Teaching a child what an inference is takes a short time. After that, more practice at inferring does not produce much more comprehension, because the limiting factor was never strategy awareness.

The limiting factor is usually knowledge and vocabulary. A child who cannot infer why a character in a historical novel is frightened may understand inference perfectly and simply not know what the character is frightened of. More inference worksheets will not help. Reading more, and reading things that build knowledge, will.

The practical takeaway is to spend most reading time reading, and to use questions to check and deepen understanding of a real book rather than to rehearse strategies against short disconnected passages.

How comprehension gets measured

Two measures come up constantly and they answer different questions. A Lexile measure describes text difficulty, and it can also describe a reader. It is calculated from sentence length and word frequency, which makes it useful for matching a reader to a book and unhelpful for judging whether a book is worth reading. A short sentence book can carry difficult ideas and a long sentence book can carry simple ones.

Standards describe what a reader should be able to do. The Common Core reading standards, used by most US states and by many homeschooling families for record keeping, set out expectations at each grade for things like citing textual evidence, determining a theme, and analysing how characters develop. They are a description of capability rather than a measure of difficulty.

For a homeschooling family, the useful combination is a book at a sensible difficulty and a record of which standards the child's reading actually covered. Difficulty on its own tells you nothing about what was understood. Reading Rockets publishes accessible summaries of the underlying research, and the Nation's Report Card publishes the national reading assessment data if you want to see how these capabilities are assessed at scale.

Is reading comprehension the same as reading fluency?

No. Fluency is the ability to read text accurately, at a reasonable pace, and with appropriate expression. Fluency supports comprehension, because a reader who is not struggling to identify words has attention left over for meaning. But a child can be perfectly fluent and comprehend very little, which is why listening to oral reading is not enough to assess understanding.

Can a child decode well and still comprehend poorly?

Yes, and it is common enough to have a name. Readers in this pattern are sometimes described as poor comprehenders. They read aloud accurately and fluently, which often means the difficulty goes unnoticed for years, and they struggle with vocabulary, background knowledge, or following extended text. The remedy is language and knowledge work rather than more phonics.

At what age should comprehension instruction start?

Long before a child reads independently. Listening comprehension develops first, and reading comprehension draws on the same language capacities. Reading aloud to a child, talking about the story, and asking why questions builds comprehension years before the child decodes their first sentence.

Do comprehension worksheets help?

A little, and then they stop. Learning what an inference is or what a main idea is takes a short time. Once a child knows, more worksheet practice adds little, because the real constraint is usually vocabulary and background knowledge rather than strategy awareness. Time spent reading whole books generally does more.

Is Chalky secular?

Yes. Chalky is a secular reading curriculum. Books are chosen for literary quality and for what they build in a reader, not for religious content or doctrinal alignment.