Senses, Instruments and Inferences

Every observation in science starts with noticing something. Most of the time we notice things with our five senses: sight, hearing, touch, smell and taste. Your senses are powerful. You can spot a friend in a crowd, hear a magpie warbling, or smell rain coming. But your senses also have real limits, and that matters a lot in science.

Here are some ways your senses can let you down:

• Touch cannot judge temperature well. If you hold one hand in cold water and the other in warm water, then plunge both into the same lukewarm bowl, one hand feels hot and the other feels cold. Same water, two answers.
• Your eyes can be fooled by optical illusions. Two lines can look different lengths even when a ruler shows they are exactly the same.
• You cannot see a fever. A person may feel hot to the touch, but you cannot tell whether their temperature is 37 degrees or 39 degrees without a thermometer.
• You cannot measure exact length by eye. You might guess a desk is "about a metre", but only a ruler or tape measure gives you the real number.

So our senses are a brilliant starting point, but on their own they can be unreliable and they cannot give us exact numbers. That is the problem instruments are made to solve.

A measuring instrument is a tool that extends and improves what our senses can do. It turns a rough feeling into an exact number that anyone can read and check. This is exactly why scientists compare observations made with the senses against those made with measuring equipment.

• A thermometer measures temperature in degrees Celsius, far more reliably than the back of your hand.
• A ruler or tape measure gives length in centimetres, instead of an eyeball guess.
• A set of scales (a balance) gives mass in grams, instead of "this feels heavier".
• A measuring cylinder gives the volume of a liquid in millilitres.
• A stopwatch measures time in seconds, far better than counting in your head.

Instruments give two big advantages. First, they are usually more accurate, meaning closer to the true value. Second, they give a number that other people can check and repeat. If you say "the water felt warm", nobody can check that. If you say "the water was 42 degrees Celsius", anyone with a thermometer can test it. Numbers that others can check are the backbone of science.

This is the most important idea in the lesson, so go slowly. There is a big difference between what you directly notice and the conclusion you work out from it.

• An observation is what you directly notice with your senses or measure with an instrument. It is just the raw fact. Example: the grass is wet.
• An inference is a conclusion you work out from your observations, using what you already know. It goes one step further than the raw fact. Example: it probably rained overnight.

Notice that the inference is not something you saw. You did not watch the rain fall. You used a known fact (rain makes grass wet) to explain the observation. An inference is a smart, sensible explanation, but it could still be wrong. The grass might be wet because a sprinkler was on, or because of morning dew. That is why scientists are always careful to keep their observations and their inferences clearly separate.

Here are more Australian examples to make the difference clear:

• Observation: there are footprints in the sand at Bondi Beach. Inference: someone walked here earlier.
• Observation: the kookaburra is sitting very still and puffed up. Inference: it might be cold or unwell.
• Observation: the creek is brown and flowing fast. Inference: it probably rained heavily upstream.

Once you have observations and a sensible inference, you can make a prediction. A prediction is a statement about what will happen next, before it actually happens. In science, the most useful predictions are testable, which means you could do an experiment or make a measurement to find out whether they come true.

Compare these:

• Testable prediction: "If I water this seedling every day for two weeks, it will grow taller than the one I do not water." You can measure both plants with a ruler and find out. This can be tested.
• Non-testable prediction: "Plants are happier when you water them." There is no way to measure plant happiness, so this cannot be tested.

A good testable prediction usually follows an "if… then…" shape and talks about something you can observe or measure: a length, a temperature, a time, a colour change. If you cannot think of any measurement that would check it, the prediction is probably not testable. Making testable predictions, then checking them with careful observations and instruments, is exactly how scientists turn an idea into real, trustworthy knowledge.

Imagine you are checking whether the water in a school fish tank is too warm for the fish. Watch how the three ideas from this lesson work together.

• Sense observation first: you put your hand near the tank and it feels warm. That is a useful start, but it is rough and could be fooled.
• Instrument observation next: you place a thermometer in the tank. It reads 29 degrees Celsius. Now you have an exact number that your teacher and classmates can all check.
• Inference: the recommended range for these fish is 22 to 26 degrees Celsius, so you infer that the tank is too warm for them.
• Testable prediction: "If we move the tank away from the sunny window, the temperature will drop below 26 degrees Celsius within a day." You can check this by measuring again tomorrow.

This is the everyday rhythm of science: observe with your senses, improve the observation with instruments, make an inference, then turn it into a testable prediction you can check. Australian scientists at the CSIRO and the Bureau of Meteorology do exactly this on a huge scale, swapping the back of a hand for high-precision instruments.