Science · Inquiry & data skills

Formulating Testable Questions

Good science starts with a good question. Here is how to turn a fuzzy "I wonder..." into a question an experiment can actually answer.

Here is something reassuring about the CAEC Science test: it is not a memory quiz. You are not asked to recite the parts of a cell or the periodic table. Instead, it checks whether you can think like an investigator, ask sharp questions, design fair tests, read data, and judge whether a conclusion holds up.

One of the most useful skills in that whole toolkit is writing a testable question. Master this and a lot of the other inquiry questions get easier, because almost every experiment begins here. Let's walk through it together.

What makes a question "testable"?

A testable question is one you could answer by collecting data, by measuring something and comparing the results. To get there, a strong question almost always has three ingredients:

One thing you change (the independent variable). This is the factor you deliberately vary, the amount of water, the temperature, the brand.
One thing you measure (the dependent variable). This is the outcome you record, and it has to be measurable, a height in centimetres, a time in seconds, a count.
A clear comparison. The question should point to two or more conditions you can line up against each other, so the data can actually settle it.

A quick test: ask yourself, "What would I measure, and what would I compare?" If you can answer both in one breath, your question is probably testable. If you cannot, it likely needs sharpening.

Why vague questions stall an experiment

Plenty of interesting-sounding questions cannot be tested as written. Watch for these traps:

Too broad. "What helps plants grow?" There is no single variable to change or measure, it could go a hundred directions at once.
No measurable outcome. "Is this plant food good?" "Good" is an opinion, not something you can read off a ruler or a stopwatch.
No comparison. "Does sunlight affect plants?" is closer, but it does not say what you would compare against, more sun versus less sun? Sun versus shade?
Not answerable by data. "Should everyone use fertilizer?" is a values question, not an experimental one.

Worked example: from a hunch to a testable question

The skill is the same no matter the science topic. Here the scenario happens to be about plants, but you would use the exact same steps for a chemistry, physics, or earth-science situation.

The situation: Priya notices that the basil plant on her sunny kitchen windowsill is much bushier than the identical one she keeps on a shadier shelf. She starts to wonder whether the light has something to do with it.

Priya's first instinct is to ask, "Does light make plants healthier?" It is a fine starting point, but it is too vague to test. Let's tighten it using our three ingredients.

Incorrect (too vague to test)

"Does light make plants healthier?"

"Healthier" is not measurable.
No clear comparison, how much light, versus what?
"Plants" is broad; which plant, under what conditions?

Correct (specific and testable)

"Do basil plants grown in 8 hours of light per day grow taller over 3 weeks than identical basil plants grown in 2 hours of light per day?"

Change: hours of light per day.
Measure: plant height, in centimetres.
Compare: 8 hours versus 2 hours.

Naming the parts

Once a question is sharp, the experiment almost designs itself. Here is how Priya's strong question breaks down:

Ingredient	In Priya's question
Independent variable (what you change)	Hours of light per day, 8 hours versus 2 hours.
Dependent variable (what you measure)	Plant height, measured in centimetres.
Comparison	Two groups of basil under different light, lined up against each other.
Kept the same (controlled)	Same plant type, same pot, same soil, same water, same 3 weeks.

Notice: a good testable question quietly tells you what to keep the same. Because only the light is meant to change, everything else, pot, soil, water, must stay constant. That is what makes the comparison fair.

The same skill, a different topic

To prove this is a transferable skill and not a plant trick, here is the same move applied to a physics situation. Notice the steps are identical, only the scenario changed.

The situation: Marcus thinks a tennis ball bounces higher on the gym's wooden floor than on the carpeted hallway.

Incorrect (vague)

"Does the floor matter for bouncing?"

Correct (testable)

"When dropped from 1 metre, does a tennis ball bounce higher (in centimetres) on a wooden floor than on a carpeted floor?"

Change: the surface (wood versus carpet). Measure: bounce height in centimetres. Compare: the two surfaces, dropping from the same 1-metre height every time. Same recipe, brand-new topic.

Tips for sharpening any question

Swap opinion words for numbers. Words like "better," "healthier," or "stronger" are signals to stop and ask, "measured how?" Replace them with something you can read off a ruler, scale, or clock.
Name what changes, just one thing. If you are changing two things at once, you will not know which one caused the result. Pin down a single variable.
Build the comparison into the question. Use "more versus less," "A versus B," or specific amounts so the data has something to settle.
Add the details that keep it fair. A phrase like "dropped from 1 metre" or "over 3 weeks" shows you have thought about keeping other factors constant.

Your turn: practice problems

For each situation, decide what is wrong with the weak question and rewrite it as a testable one. Try it before you peek.

A baker wonders if adding more yeast changes how bread rises. Weak question: "Is more yeast better for bread?"
A student notices ice seems to melt faster in salty water. Weak question: "Does salt do something to ice?"
Someone asks, "Should people drink coffee to focus?" Is this a testable question at all?

Tap to reveal the answers

1. The problem is "better", not measurable, and there is no comparison. Testable version: "Does dough made with 2 teaspoons of yeast rise higher (in centimetres) in one hour than identical dough made with 1 teaspoon of yeast?" Change: amount of yeast. Measure: rise height. Compare: 2 tsp versus 1 tsp.
2. "Does something" is vague and has no comparison or measurable outcome. Testable version: "Does an ice cube melt faster (in seconds) in salt water than in plain water at the same temperature?" Change: salt versus no salt. Measure: melting time. Compare: salt water versus plain water.
3. As written, it is not testable, "should people" is a values question, not something data can settle. But you could rewrite the idea behind it as a testable one: Testable version: "Do people who drink one cup of coffee complete a 20-question puzzle in less time than people who drink water instead?"

Why this matters for the CAEC

The CAEC Science test is 35 questions in 90 minutes, and a calculator is allowed. Crucially, it is a skills and inquiry test, not a fact-recall test, most of the marks come from things like formulating questions, designing fair experiments, and interpreting data. The biology, chemistry, or physics topic is just the setting; the real question is whether you can think like a scientist. Spotting and fixing a weak question is a skill that pays off across the whole test.

Ready for more inquiry practice? Browse our Science lessons, pick up the full CAEC Ready Workbook, or start with a free sample to test yourself.

Disclaimer

This article is a general study lesson. CAEC Ready is an independent study resource and is not affiliated with or endorsed by any government, ministry of education, or official CAEC testing provider.