Methods · Visual, spatial & data representation

Plot my own data

Turn numbers you generated yourself into a chart, and find out that reading a graph and making one are different skills.

When to use this: While you practise
Learning impact: Reliable gain
Learning evidence: Credible with caveats
Where you are with the topic: Building understanding
How often to use it: Worth repeating often

When to use it

You have numbers from a lab, survey, or homework and only a vague sense of the pattern.
You can read other people's charts but freeze when you must choose axes yourself.
You are about to paste data into a plotter and want the thinking to stay yours.

What to ask AI

Type something like: “Here's my data [paste] and what I think it shows: [prediction]. Before plotting: ask me what shape I expect and why. Then help me choose axes. After the plot, ask what surprised me, do not interpret first.”

Ask it to interrogate your axis choices and expectations, not to decorate a chart for you.

Keep the hard part on your side: you predict the shape before you see the plot.

How it helps

Reading a finished graph and making one are different skills.

When you choose the axes, scale, and transforms, you confront decisions that stay invisible in a textbook figure.

Your own data also carries a prediction you can falsify, that mismatch is the learning, not the pretty output.

How to run it

Paste or list your data and write one sentence on what you think it shows.
Before any plot: state the expected shape and why.
Choose variables for axes; defend the scale (linear vs log, truncated or not).
Plot (or describe the plot carefully), then name what matched and what surprised you.
Ask what a misleading version of the same data would look like, so you notice design choices.
Write a one-sentence claim the chart actually supports, not the claim you hoped for.

Example exchange

You

I've got reaction times at three caffeine levels. I think higher caffeine means faster times, but I haven't plotted yet.

Before you plot: do you expect a steady linear drop, a step drop, or a U-shape, and why?

You

A steady linear drop. I'll plot means with error bars and keep outliers marked.

Plot it, then tell me one thing that contradicts the linear story.

Copyable prompt

Here's my data [paste] and what I think it shows: [prediction]. Before plotting:
ask me what shape I expect and why. Then help me choose axes and scale. After the
plot, ask what surprised me, do not interpret the chart before I do.

The Tell

Here is how you know this method has flipped: you have a nice chart and cannot say what you would have expected before you made it.

If the AI chose the story and the axes, you practised software, not graph thinking.

Predict first, then plot, and own the claim the picture actually supports.

Principle evidence

Strength of the underlying learning idea, not a claim about AI products.

The underlying learning idea is rated moderate. Graph construction, not only graph reading, is a distinct literacy skill: axis choice, scale, transforms, and outliers are decision points that finished figures hide.

AI delivery evidence

Whether an AI tutor delivers this method well is a separate question.

AI can suggest plot types fluently and still push a misleading scale. Treat plotting advice as provisional; check against your data and a real plotting tool when stakes are high.