How to plan an experiment
Here is an example of a process for planning an experiment. Click on each stage to find out more information.
If you can't access the interaction below, you can use this alternative instead.
A detailed step-by-step process for designing a fair test
The following is a more detailed step-by-step process that briefly outlines what needs to be planned for and included in designing an experiment.
Step 1: Design the test
A. Write down your preliminary ideas, for example:
- What do you want to do? (make an educated guess or hypothesis)
- What do you already know?
- What do you think the answer is to your hypothesis?
- What do you think will happen, and why?
- What are the things that are going to change (variables)?
- What are the things that can be measured and recorded (data)?
- What material you will need?
B. Sort out your ideas, for example, write:
- a hypothesis
- the one thing that you will change is……. (known as an independent variable)
- the one thing that you will measure is……. (known as a dependent variable)
- all the things you will keep the same are…. (known as constants).
C. Create the procedure:
- Set the parameters of your trial and write the steps you need to do to carry out it out. Include:
- Variables (independent and dependent)
- Constants
- Sampling (duration, interval, e.g. daily, sample size)
- Assumptions
- Controls
- What are your predictions?
- Create data sheets for recording the information
Step 2: Carry out the test
- Use the procedures you designed to carry out the test.
- Use the datasheet to record the information.
Step 3: Collate and analyse the data
- What type of data are you collecting - quantitative or qualitative?
- Change the raw data into meaningful statistics - percentages, averages, graphs, comparisons.
Step 4: Interpret the statistics
- Does the data support or refute your hypothesis?
- Does the data reveal a flaw in your design?
- Was the hypothesis incorrect?
- Was there an alternative hypothesis?
Step 5: Write a report
- Make arguments based on your data and separate the observations from interpretation.
- Provide sufficient detail to enable others to repeat the same experiment.
- Summarise your findings and observations.
- Do not present raw data.
- Describe the outcome of the trial.
- Do not use the phase 'made a mistake' or say that the experiment failed, instead, explain the outcome in terms of observable facts.
- Do not make excuses.
- Explain alternative hypotheses (if applicable).
- Explain issues that you did not consider in your initial design or hypothesis.
- Evaluate whether the test was valid, i.e. if the test was repeated would the same conclusions be reached?
Step 6: Review the experiment
- Look at how the test can be improved, i.e. what would you do differently next time?
|
