1.7 Evaluating data

Summary: Top-level athletes like those in Figure 1 always look for ways to improve their performance. They identify areas where they can make changes to perform at a higher level in future events. Tags: science oxford-science student-book-7 Created: 2026-05-19 Last Updated: 2026-05-19

After this topic, you will be able to:

• describe the stages in evaluating data
• suggest ways to improve a practical investigation.

Think back

1. How do you show a pattern on a line graph?
2. What should a conclusion include?
3. What name is given to a measurement that is very different from other repeat results?

Key idea Scientists evaluate the data collected and the method used. They look at anomalies, the spread of data, and any errors. This helps them improve their investigation.

Key words evaluate, evaluation, confidence, spread, repeatable, random error, systematic error, range

Top-level athletes (like those in Figure 1**) always look for ways to improve their performance. They identify areas where they can make changes to perform at a higher level in future events.**

Figure 1Evaluating your performance after a race helps you identify areas to improve.

What is an evaluation?

There are three parts to a scientific investigation:

• Identify the strengths and weaknesses in your results.
• Decide how confident you are in your conclusion.
• Suggest and explain improvements to your method, so you can collect data of better quality if you did it again.

Katie and Rahim have collected data and analysed it. Now they need to evaluate their data and their method. This is called an evaluation.

Identifying strengths and weaknesses in data

Katie and Rahim had only one anomaly in their experiment – the third measurement at 20 °C (see page 11). If there were lots of anomalies, they would have less confidence in their conclusion.

The spread is the difference between the highest and the lowest readings in a set of repeat measurements. In Katie and Rahim’s results, the spread of the data for 0 °C is 10 cm (35 cm–25 cm). A small spread in the data will give you more confidence in your conclusion. Your method is repeatable if your repeat results are very similar.

There is an experimental error in any measurement you make. This is why there is often a small spread in experimental data.

However, there are two types of error that can cause much bigger changes in scientific measurements:

• random error – this is where a change you were not expecting affects your results. An example is the temperature of the room suddenly changing because someone opens a door.
• systematic error – this is where something affects all of your results all of the time. Systematic errors are often caused by faulty measuring equipment, such as in Figure 2.

Figure 2This newton meter shows a systematic error. It has a reading when there is no load. Full text alternative to the content

A A balance shows a reading of 0.2 g when it has no load. State whether this is an example of a random error or a systematic error.

You should think about errors, anomalies, and spread to help you decide how confident you are in your conclusion.

Improving an investigation

It is always possible to improve the quality of your measurements in an investigation. Improvements would make you more confident that your conclusion is correct.

Remember, the range is the smallest and largest values of the independent variable. The wider the range, the more certain you can be that your conclusion is always true. Rahim and Katie used a range of 0 °C to 40 °C. To be more certain of their conclusion, they could test hotter balls.

Remember – within the range, you should always investigate at least five values of the independent variable.

B Suggest another temperature of ball Katie and Rahim could test to improve their experiment.

If you have not taken repeat readings, you cannot be certain if any of the results were a coincidence. You should always take at least three repeat readings for each value of the independent variable.

If you had several anomalies, or a large spread of data, you might want to consider using different equipment to collect your measurements. For example, Katie and Rahim could have used a video camera to record the highest bounce point. It is difficult to take an accurate measurement of a moving object.

Table 1

1. Match the words to the definitions. range an error which consistently affects a set of results spread an error caused by an unexpected event random error the difference in readings within a set of repeat results systematic error the smallest and largest values of the independent variable
2. Draw a flowchart to show the stages you should follow when writing an evaluation.

Table 1

1. A group of students collect the data in Table 1 about how the mass of a paper aeroplane affects how far it can fly. Table 1Mass of aeroplane in g Distance travelled in m Measurement 1 Measurement 2 20 1.8 2.0 40 1.4 3.8 60 1.2 1.0 Identify the anomaly.The students conclude: The bigger the mass of a paper aeroplane, the less distance it can fly. Evaluate their results.