Experimental Skills » Planning Experiments and Variables

From idea to hypothesis

Every experiment starts with a question. Scientists use theories and explanations they already know to suggest an answer. That answer, which can be tested, is the hypothesis. You met this word in the antibiotics practical, so here is how to build one.

Start with an observation, then use your biology to explain it. For example, you notice that bubbles of oxygen come off a piece of potato dropped into hydrogen peroxide. A drop of washing-up liquid traps the bubbles as a foam you can measure. You know that cells make an enzyme called catalase, and that enzymes work faster when they are warmer, up to a point. So you suggest: "The higher the temperature, the faster catalase breaks down hydrogen peroxide, up to the optimum temperature."

A good hypothesis names what you will change, names what you will measure, and gives a reason from science. You can then test it with an experiment.

Key terms:

• Valid: an investigation is valid if it measures what it claims to measure and only the factor you chose changes the result.
• Precise: measurements are precise if repeats give values that are very close together.

Planning a fair test

To test a hypothesis you plan an experiment or devise a procedure. The plan needs to say what you will do, step by step, so that someone else could follow it.

The key is the variables. Remember the three types from the antibiotics practical, and see how they work for the catalase hypothesis.

Why control variables? If you changed the temperature and the amount of potato at the same time, you could not tell which one caused the change. You need to manipulate only the independent variable and control everything else, so the result is valid.

An experiment is well designed for its purpose when it changes only one variable, measures the right thing, uses a sensible range of values and includes repeats.

Choosing apparatus and techniques

You need to select the technique, instrument, apparatus or material that suits the job, and explain why. Ask yourself two questions: what am I measuring, and how accurately do I need to measure it?

• A measuring cylinder gives volumes more accurately than a beaker.
• A water bath keeps a temperature steadier than a beaker heated over a flame, and a thermometer measures it.
• A top-pan balance measures mass, and a ruler measures length in mm.
• A stopwatch times reactions. A pH meter or indicator measures acidity.

A measurement is accurate if it is close to the true value, so choose apparatus with a fine enough scale. Measuring a 3 mm change with a ruler marked only in centimetres would give poor results.

Working safely and carefully

You must carry out experiments appropriately: handle apparatus correctly, measure accurately and think about health and safety.

First, identify the main hazards. Then suggest ways to reduce the risk of harm.

Take measurements carefully. Read a scale with your eye level with it, and record each reading to the same number of decimal places.

Sampling

Sometimes you cannot measure everything, so you take a sample. The sample must be representative, which means it should match the whole population.

Suppose you want the mean length of holly leaves on a large hedge. If you only pick the leaves you can reach at the front, they may be longer or shorter than the rest. A better technique is to pick leaves at random from all sides and heights. Use enough leaves, such as 30, because a larger sample is more likely to be representative. Choosing at random avoids bias.

For plants and animals in a habitat, quadrats and transects are sampling techniques you have already met. The main point here is to be able to suggest and describe one that suits the situation.

Recording and checking measurements

Record your results in a table as you go, with headings and units. Repeat each measurement so you can spot anomalous results and calculate a mean.

Then ask: have I taken sufficient measurements, meaning enough repeats and enough values of the independent variable to draw a conclusion? Are they precise? For the catalase test, one reading at each of five temperatures would not be enough. Three readings at each temperature would be much better.

Evaluating the method

At the end, evaluate the method to decide whether the results are valid. Think about what went wrong and what you would change.

• Was it a fair test? Did any control variable change? For example, the water bath cooled during the test.
• Were there enough readings? Too few repeats, or too few temperatures, gives weak results.
• Was the measurement good enough? Foam height is hard to read, so measuring the volume of oxygen collected would be better.

Suggest improvements, such as using a water bath with a thermostat to keep the temperature steady, adding more temperatures between 30 and 50 °C, or repeating more times. You can also suggest further investigations, such as testing a different enzyme or a different pH.

Exam-style question

A student investigates whether the mass of yeast affects how many bubbles of gas are made in a sugar solution in 5 minutes. She uses 1 g, 2 g, 3 g and 4 g of yeast.

(a) Name the independent variable and the dependent variable. [2 marks]

(b) Give two control variables. [2 marks]

(c) Explain why she should repeat each test. [2 marks]