Inference Questions in English Reading Comprehension: How to Teach Children to Read Between the Lines

Of all the question types in English reading comprehension, inference questions are the ones that most reliably divide the class. Strong readers answer them with apparent ease. Others stare at the page, knowing the passage was understood but unable to bridge the gap between "what the text says" and "what the question wants."

Inference is one of the higher-order reading skills — it requires the reader to combine what is explicitly stated with background knowledge and contextual clues to reach a conclusion the text does not state directly. It is, in other words, thinking while reading, not just decoding.

Teaching inference is not about learning a trick. It is about developing a habit of active, questioning reading. But there are specific strategies that help, and I use them regularly in my classroom.

What Inference Actually Is

Let me start with a clear definition through example.

Passage: Marcus put his key in the lock three times before his trembling hands finally managed to turn it. He pushed the door open, hurried to the kitchen, and sat heavily at the table, pressing his cold palms flat against the surface.

Inference question: How is Marcus feeling? How can you tell?

The text never says "Marcus was frightened" or "Marcus was anxious." But a reader who understands how physical symptoms relate to emotions — trembling hands, rushing, needing to sit, cold hands — can infer that Marcus is significantly distressed, possibly frightened or panicked.

This is reading between the lines: drawing a conclusion from the evidence in the text rather than finding a directly stated answer.

Why HK Students Find This Difficult

Over-reliance on literal reading. From P1 to P3, most comprehension tasks reward finding stated information. Children develop the habit of scanning for keywords and copying. Inference tasks require the opposite — synthesis rather than location.

Cultural and experiential gaps. Inference requires background knowledge. A child who has never been genuinely frightened in a frightening setting may not have the emotional reference point to recognise the described symptoms as fear. Wide reading builds this experiential database.

Language processing demand. For Cantonese-first learners, the cognitive effort of reading English can crowd out the higher-order processing needed for inference. A child who is still working hard at the sentence level has less capacity for the meta-level thinking inference requires.

Strategy 1: The Clue + Inference Two-Step

I teach a specific sentence frame for inference questions:

"I know [inference] because the text says '[direct quote or evidence]'."

This forces the child to:

1. State their inference (what they concluded)
2. Anchor it in the text (what evidence supports it)

Both halves are necessary. An unsupported inference ("I know Marcus is scared") scores less than a supported inference ("I know Marcus is scared because the text says his hands were trembling and he sat heavily at the table, suggesting his body was reacting to stress").

The two-step frame is initially mechanical — but that is fine. Mechanical practice of the structure internalises the habit. Over time, the child produces the structure naturally.

Strategy 2: The Four Clue Types

I teach my students to look for four types of evidence that support inferences:

1. Physical descriptions of characters: Trembling, sweating, blushing, pale face, wide eyes → these suggest specific emotions with fairly predictable mappings.

2. Actions and behaviour: Running away, hiding, repeatedly checking something, avoiding eye contact → these suggest motivations and emotional states.

3. Dialogue and tone: What characters say, and how they say it (shouted, whispered, murmured, snapped) → reveals attitude and emotion.

4. What characters do NOT say or do: A character who stays silent when asked a question, who leaves a room abruptly, who changes the subject → inference from absence is sophisticated but important.

Teaching children to look for these four clue types gives them a systematic search strategy rather than relying on a vague "read between the lines" instinct.

Strategy 3: Emotion Mapping

One specific inference skill is recognising emotions from descriptions. I create an "emotion evidence" reference chart with my P4 students:

Physical description	Likely emotion
Trembling hands, racing heart	Fear, nervousness, excitement
Slumped shoulders, slow movements	Sadness, defeat, exhaustion
Wide eyes, quick breathing	Surprise, fear, excitement
Flushed face, raised voice	Anger, embarrassment
Avoiding eye contact	Guilt, shyness, nervousness
Bright eyes, quick steps	Happiness, excitement, eagerness

This is a scaffold, not a rule. Emotions are complex and context-dependent. But having explicit mappings helps children who are not making them intuitively.

Strategy 4: Practising with Non-Text Images

Inference is not exclusively a reading skill — it is a general thinking skill. Practising inference with images before applying it to text builds the habit in a lower-stakes context.

Show your child a photograph of a person or scene. Ask:

• "How does this person feel? What makes you think that?"
• "What do you think just happened? What's your evidence?"
• "What might happen next? Why do you think that?"

This is oral inference practice, and it directly transfers to reading comprehension. Children who practise inference through image interpretation often show immediate improvement in text inference tasks.

Helping at Home Without Over-Scaffolding

When your child does a reading comprehension at home and gets an inference question wrong, resist the urge to explain the correct answer immediately. Instead:

Ask: "What do you think the character was feeling there?" If they say "I don't know": "Well, what was happening to their body? What were they doing?" If they identify some evidence: "So what do you think that tells us?" If they reach the inference: "Now — can you say that as a full answer, with the evidence you just found?"

This Socratic approach is more demanding for you as a parent, but it develops the child's thinking process rather than just delivering the answer. The process is the learning.

A Realistic Progression

Inference is a skill that develops over years of reading and discussion. P3–P4 children are typically at the stage of identifying obvious emotional inferences with support. P5–P6 children should be able to make multi-layered inferences about motivation, theme, and implied meaning with relative independence.

The most powerful single thing you can do to develop inference is to read and discuss widely — books, stories, films, news events. "Why do you think she did that?" is an inference question. Ask it everywhere, not just in reading comprehension exercises. The thinking habit is what matters, and it is built in a thousand small moments.