A second-grade argument about a lima bean
In one second-grade class filmed for a study of science talk, a group cut open lima beans and announced there was a seed inside the seed. A classmate named JoAnn was not convinced1:
I disagree that you could see a seed in a seed.1
What happened next is the useful part. The group described what they saw, a classmate pointed out the stem they had found, and a girl named Erin offered a new idea: maybe the "seed" was a hard coat protecting what was inside. The talk ran more than seventeen turns, then about thirty more, and ended without an answer. The class planned to check nonfiction books next time1.
The talk was left unsettled, and the researchers noted it still seemed satisfying to the class1.
Claim, evidence, reasoning
Science educators Douglas Clark and Pratim Sengupta describe scientific argument as examining how evidence connects to the ideas used to explain it1. For a child, we break that into three parts with plain names. The breakdown and the examples below are ours.
| Part | Kid words | At dinner | On a walk |
|---|---|---|---|
| Claim | What I think | "Ice melts faster in my cocoa." | "It rained last night." |
| Evidence | What I saw, heard or measured | "My cube was gone first. Yours is still floating." | "The sidewalk is wet, but the sky is clear." |
| Reasoning | Why that shows it | "My cocoa is warm and your milk is cold." | "The water had to come from somewhere, and nobody watered the grass." |
Most five-year-olds give a claim and stop. That is fine. Just ask for the next part, kindly, every time.
What you saw comes before what it means
Darci Harland's research handbook, written for high school students, warns that a note like "the solution looks normal" is an inference, a conclusion, and is useless without the observations behind it. She tells students to record what they actually see and save inferences for after the data2. Our adaptation for younger kids: ask for the "what" before the "why".
- "Is that right?"
- "No, that's wrong."
- "What's this called?"
- "Was it good?"
- "How do you know?"
- "Interesting. What did you see that made you think that?"
- "What do you notice about it?"
- "What did you see happen?"
In those low-level classrooms, questions asked for a fact and stopped, with no follow-up, and ideas were not grounded in how claim and evidence connect1. Quiz questions at home do the same: one answer, then silence.
Questions for dinner, walks and play
In the classrooms where children carried the talk themselves, the researchers found the first question mattered less than what came after it. Teachers and students kept asking follow-up questions, whatever the first question was1. So keep a few follow-ups ready.
- At dinner. "What do you think will happen to the ice in your drink? How will we know?"
- On a walk. "Which way is the wind blowing? What are you seeing that tells you?"
- During play. "Why do you think the tower fell? What would change your mind?"
- Any time. "Did anyone see something different?" and "How could we test it?"
Predict, observe, explain: a 3-minute routine
Teacher-education books describe inquiry as a cycle: ask a question, investigate, gather data, use evidence to interpret it, then communicate. One table even lists "argue an interpretation" as a student skill3. Here is our pocket-size home version.
-
01Predict.
"What do you think will happen?" Ask your child to say it out loud or point before anything moves.
-
02Observe.
Do it. Then ask "What did you see?" Accept only what they saw for now, not why2.
-
03Explain.
"Why do you think it happened?" Follow with "How do you know?"
-
04Revise.
If the prediction missed, ask "What would you predict now?" A changed mind is the goal, not a loss.
When your child's idea is wrong
Clark and Sengupta note that students often rely on their personal views rather than the data in front of them1. Our view: telling a child the right answer fixes one idea. Sending them back to the evidence builds the habit of checking.
In the lima bean class, the teacher did not name a winner. The teacher reminded children to "listen first", invited other voices in, and sent the open question to books for next time1. In the lowest-level classrooms, by contrast, a discussion seemed "won" or "settled" once it ended1.
- "That's an interesting idea. What makes you think so?"
- "What would we see if you're right? What would we see if you're not?"
- "Let's test it."
- "It's okay that we don't know yet. Where could we look?"
- "I used to think that too. Here's what changed my mind."
Science means getting the right answer.
Clark and Sengupta describe scientists constructing, revising and defending models, then testing and refining them as a community1. Revising is the job.
An unfinished argument means the conversation went badly.
The second graders' seed discussion stayed unresolved, and the class planned its next step: checking books1.
Real research can be messy.2
Draw it, then fix the drawing
Clark and Sengupta define models as physical, computational or mental stand-ins for something real, tools that let you describe, predict and explain1. In their computer-modeling design for students, learners fix a model by comparing what they meant to happen with what actually happened, instead of by random trial and error1.
A drawing is the simplest model a child can make. In an Australian pilot study of one Year 4 class (average age 9 years 9 months), four children corrected their own mix-up of protons and electrons once they started using a hands-on atom model1. The authors also advise explaining how a model does and does not resemble the real thing1.
Draw what you think is happening
- Paper and crayons or pencils
- A clear glass of water
- Two or three ice cubes
- A clock or timer
- Before adding the ice, ask your child to draw what they think will happen to an ice cube in the water. Label it Drawing 1.
- Ask them to explain it: "What are these lines? What happens next?"
- Drop in the ice and watch for 10 minutes. Ask only "What do you see?" and write down their words.
- Put Drawing 1 next to a fresh sheet. Ask: "Is there anything you want to change? What did you see that tells you?"
- Make Drawing 2 and keep both. Ask: "How is your drawing like the real ice, and how is it different?"
What you think is true.
Say it: “What's your claim?”Something you saw, heard or measured that backs up a claim.
Say it: “How do you know?”Why your evidence shows your claim is true.
Say it: “Why does that prove it?”A drawing, build or idea that shows how you think something works1.
Say it: “Can you draw what you think is happening?”What to expect by age
Can: Tell you what they think and show you one thing they saw. A drawing counts as evidence.
Try: Use just two questions: "What do you think?" and "How do you know?"
Can: Work through results that disagree. In one grade 5 class, two groups that heard different pitches from jars talked it through until the water levels in the jars helped explain both results1.
Try: Add "What would change your mind?" and keep a two-column page: what I saw, and what I think it means2.
If you want something new to talk about
Everything above works with a glass of ice and a walk. If you would like a new building project arriving each month to make claims about, there is our Club.
ClubA sealed box for every monthMonthly at $39, or $29 a box on the 12-month plan. Siblings pay less.from $29GoP Toys sells the Club mentioned above. None of the sources cited tested our products.