WH-Questions Are a Hierarchy, Not a List
The WH-questions arrive in a fairly reliable order, and knowing it tells you where to start. "What" and "who" come first, because the answer is a thing or a person and it is usually right there. "Where" follows. "When" is harder because it needs a grasp of time. "Why" and "how" come last, because both require reasoning about something that was never stated.
Within any of those, questions still range from literal to inferential to predictive. "What did the boy carry?" is literal — the answer is in the text. "Why was he in a hurry?" is inferential. "What will happen when he opens the door?" is predictive. A child who is fine on the first and lost on the second has a very specific, workable target, and an assessment that reports only "comprehension difficulty" has not gone far enough.
That hierarchy is also how you grade an activity on the spot. If a child fails an inference, drop to the literal question underneath it, confirm they held the facts, then step back up. Most of the time the facts were there and the leap was not.
What Inferencing Actually Requires
An inference needs three things at once: the information from the text, relevant background knowledge, and enough working memory to hold both while combining them. When inferencing fails, it is worth working out which of the three is missing, because they need different responses.
Background knowledge is the one most often overlooked. A child cannot infer that a character is at a wedding from a mention of confetti if they have never been told what confetti is for. That is a knowledge gap wearing a comprehension costume, and it is why vocabulary and comprehension goals usually travel together.
The picture-based inferencing worksheet — look at the scene, say why the girl is sad — is the standard material and its limitation is that it removes the memory load entirely. Everything needed stays on the page. A spoken or narrative version is harder and closer to the real demand, which is a classroom where the information arrives once, out loud, and then is gone.
Comprehension, Reading, and the Third-Grade Cliff
Up to about third grade, school teaches reading. After it, reading is how everything else is taught, and text gets denser at exactly the point at which explanations stop. Children whose difficulty is language comprehension rather than decoding often look fine until here and then appear to "suddenly" struggle, which is why the drop is usually described as a cliff rather than a slope.
The practical test parents can run themselves takes thirty seconds: have the child read a paragraph aloud, then close the book and ask what happened. Accurate reading with an empty retell points at comprehension, not at reading. Inaccurate reading points somewhere else entirely.
This is also why listening comprehension is worth practising without text at all. Removing decoding from the task isolates the language, and for a child with both difficulties it lets you work on one of them at a time.
Running It in a Session or at Home
Books and films work as well as any material, provided you stop. Pause and ask what might happen next, why someone did that, and how they are feeling — three questions in a chapter beats twenty at the end, because the answer is still available and the child has not yet been asked to hold the whole plot.
Resist the urge to quiz. Comprehension practice that turns into an interrogation gets quiet children, and a question the adult already knows the answer to is a different task from a question that matters. Wondering aloud — "I think he's about to get in trouble" — invites a response and models the reasoning at the same time.
In a mixed group, comprehension is the goal that benefits most from other students being in the room. Ask the group to agree or disagree with an answer before it is confirmed, and to say why. That turns a one-player game into genuine practice for everyone waiting a turn.