ACARA v9 CONTENT DESCRIPTION “pose questions to explore observed patterns and relationships and make predictions based on observations”
Builds on earlier looking and wondering about the world. Now an everyday observation grows into a question you can explore: you spot a pattern, ask about the relationship behind it, and make a prediction based on what you have already seen.
From an observation to a question you can explore
You roll a toy car across the smooth kitchen floor and it glides a long way. You roll the same car across the carpet and it stops much sooner. That is an observation, and the difference between the two is a pattern worth exploring. A good science question asks about the relationship behind the pattern: how does the surface the car rolls on relate to how far it travels? To explore that fairly, you change just one thing at a time, so you can tell what really caused the difference.
Frame the question: change only the surface
You saw the car roll further on a smooth floor than on carpet. To explore that pattern fairly, keep everything the same except the one thing you are testing.
You roll the same toy car on a smooth floor and on carpet to see which lets it travel further. To explore the pattern fairly, hold the other things steady so only the surface differs.
Variable being tested: The surface the car rolls on (this one we change)
You give the car the same push every time
You use the same toy car for every trial
The car starts from the same line each time
Not a fair test yet: more than one thing is changing, so you could not tell which change caused the result. Hold every other variable the same.
Decide what to measure
A good question also needs a clear thing to measure, chosen before you begin, so that every trial gives you a number to compare. For the rolling car, the natural measurement is how far it travels before it stops. Measure that distance the same way every time, from the same starting line, and each surface produces a value you can line up against the others.
Decide what to measure before you start
Lock in the one thing you will record, in the same units, before the first trial, so the surfaces can be compared fairly.
To answer whether a smoother surface lets the car roll further, you must record the same measurement on every surface, in the same way.
Variable being tested: The surface the car rolls on (this one we change)
The distance the car rolls before it stops, measured in centimetres each trial
Not a fair test yet: more than one thing is changing, so you could not tell which change caused the result. Hold every other variable the same.
Predict from what you have observed
A prediction is more than a wild guess. It says what you expect, and it gives a reason based on a pattern you have already observed. You have seen the car roll further on the smooth floor than on the rough carpet. So a reasoned prediction is: on an even smoother surface, such as a polished tile, the car will roll further still. The experiment then checks whether that relationship really holds when you measure it.
Look for the pattern in the readings
Once the question is clear and you know what to measure, you take readings and look for a pattern. Suppose you roll the car across surfaces that get smoother and smoother, and you measure how far it travels each time. A steady pattern would have the distance grow as the surface gets smoother. As you read the data, watch for any single reading that does not fit the rising pattern, because that is the one worth checking again before you trust it.
Spot the reading to check again
The same car was rolled across surfaces that got smoother and smoother. The distance it rolled was measured in centimetres. One reading does not fit the steady climb.
Click the point that does not fit the pattern of the others.
Which predictions are reasoned from the observation?
A prediction is only worth keeping if it follows from what you have actually seen. The observation was that the car rolls further as the surface gets smoother. Sort each statement by whether it is a prediction reasoned from that pattern, or just a guess or an opinion that does not follow from the evidence. Good questioning means telling a reasoned prediction apart from a statement that only sounds related.
Sort the reasoned predictions from the guesses
The observation: the car rolls further as the surface gets smoother. Decide which statements are predictions reasoned from that pattern.
Claim: The smoother the surface, the further the car will roll.
On a polished tile, smoother than any surface tried so far, the car should roll the furthest yet.
On rough gravel, rougher than the carpet, the car should stop sooner than it did on carpet.
A red car will surely roll further than a blue one.
Since each smoother surface gave a longer roll, a fresh smoother surface should give a longer roll again.
The car rolls best on Fridays.
Decide whether each statement is evidence for the claim, or not.
Why this matters
Every investigation starts with a question you can really explore. Learning to spot a pattern in what you observe, to ask about the relationship behind it, to decide what to measure, and to predict from what you have already seen turns everyday wondering into real science. It is the same first step an engineer testing tyres or a coach timing a race takes before any careful test.
Quick self-check
1. You notice a toy car rolls further on a smooth floor than on carpet. Which is a good science question to explore this?
2. To explore how the surface affects rolling distance fairly, what is the one thing you should change?
3. In this rolling test, what is the sensible thing to measure?
4. You have seen the car roll further on smoother surfaces. Which is a reasoned prediction for a brand new, very smooth tile?
5. Your rolling distances get longer as the surface gets smoother, but one reading is far too short. What should you do with that odd reading?