Practical Investigation: Waves or Motion Data
In 2010, CSIRO researchers at the WA Wave Energy Project measured wave converter output 48 times per second — valid data demands a rigorous fair-test design first.
Printable Worksheets
Print or save as PDF — or build a custom worksheet from any module's questions.
You want to find out whether a heavier ball bounces higher than a lighter ball. List three things you would keep the same and one thing you would change in your experiment.
Why do scientists repeat their measurements multiple times instead of just doing an experiment once?
● Know
- A fair test changes only one variable at a time while keeping all other conditions constant.
- Reliable results require repeated measurements and careful recording.
● Understand
- Scientific investigations follow a method: aim, hypothesis, method, results, conclusion.
● Can do
- Plan and conduct a simple practical investigation, record data in tables, and identify sources of error.
Two students both test whether ramp angle affects trolley speed: one changes the angle but also changes the mass of the trolley between trials; the other changes only the angle and keeps everything else identical. Only the second student's results can answer the question, because only one variable changed. This is the fair-test principle that sits at the heart of every scientific investigation: to find out how one factor affects an outcome, you must change only that factor while keeping everything else constant.
Independent variable: The factor you deliberately change. It goes on the x-axis of your graph.
Dependent variable: The factor you measure as a result. It goes on the y-axis.
Controlled variables: All other factors that must be kept constant to ensure a fair test. If you do not control these, you cannot be sure what caused any changes you observe.
Reliability: Repeating measurements multiple times reduces the impact of random errors. Calculate the mean and range to summarise your data.
Validity: Your method must actually test what you claim to test. A method with poor validity gives meaningless results regardless of how carefully you follow it.
Investigation: How does the angle of launch affect the distance travelled by a paper airplane? Independent variable: launch angle (0, 15, 30, 45, 60 degrees). Dependent variable: horizontal distance travelled (metres). Controlled variables: same paper airplane design, same paper type, same launch force, same launch height, same indoor environment (no wind), same measurer. Method: Launch the plane 5 times at each angle, measure distance each time, calculate mean distance for each angle. Plot angle (x) versus mean distance (y). Expect a peak at some intermediate angle, likely around 30-45 degrees.
Australian science education: The Australian Curriculum Science Inquiry Skills strand requires students to plan fair tests, identify variables, and evaluate measurement reliability. The Science Teachers Association of Victoria and other state bodies provide investigation planning templates that scaffold students through hypothesis, method, risk assessment, and evaluation. These frameworks ensure that practical work develops scientific reasoning, not just procedural following.
If an experiment gives unexpected results, the method must be wrong. This is false. Unexpected results are often the most valuable in science. They may reveal flaws in the hypothesis, point to uncontrolled variables, or even indicate a genuinely new phenomenon. The response to unexpected results should be to repeat the experiment, check for errors, and if results persist, revise the hypothesis. Dismissing unexpected data is a form of confirmation bias that hinders scientific progress.
Put these steps of a fair-test investigation in order.
- Design a method that changes only the independent variable.
- Analyse results using graphs and calculations.
- Draw conclusions and evaluate the method reliability.
- Conduct the experiment and record data systematically.
- Identify the research question and hypothesis.
- List the independent, dependent, and controlled variables.
No measurement is perfectly accurate. Understanding the types and sources of error helps you assess the reliability of your data and improve your methods.
Random errors are unpredictable variations that affect measurements in both directions. They arise from limitations in measurement precision, environmental fluctuations, and human judgement. Random errors can be reduced by taking multiple measurements and calculating the mean. The spread of measurements (range or standard deviation) indicates the magnitude of random error.
Systematic errors are consistent biases that shift all measurements in the same direction by the same amount. They arise from faulty equipment, incorrect calibration, or flawed experimental design. Systematic errors cannot be reduced by repeating measurements. They require identifying and correcting the source of bias.
Parallax error occurs when you read a scale from an angle rather than straight on. The apparent position of the needle or liquid surface shifts, causing incorrect readings. Always view measuring instruments perpendicularly.
Zero error occurs when an instrument does not read zero when it should. A ruler with a worn end, or a scale that does not tare properly, introduces zero error. Check and adjust instruments before use.
Measuring the period of a pendulum with a stopwatch introduces human reaction time error (typically 0.2-0.3 seconds). For a pendulum with period 1.0 seconds, this is a 20-30% error - enormous. Using a photogate timer reduces this to milliseconds. Alternatively, timing 10 complete swings and dividing by 10 spreads the reaction time error across 10 periods, reducing its impact to 2-3%. These are standard techniques taught in Australian physics classrooms.
Australian metrology: The National Measurement Institute (NMI) maintains Australia primary measurement standards and calibrates scientific instruments. Their laboratories achieve uncertainties of parts per billion for time, length, and mass standards. These calibrations trace back through an unbroken chain to international standards, ensuring that measurements made in Australian schools, hospitals, and industries are consistent and comparable worldwide.
Error means mistake. In science, error does not mean a mistake or blunder. It means the inevitable uncertainty in any measurement. Even expert scientists using the best equipment cannot eliminate error completely. The goal is not to achieve perfect accuracy (which is impossible) but to quantify the uncertainty and ensure it is small enough for the purpose at hand.
- Random error
- Systematic error
- Parallax error
- Human reaction time
- Delay in starting/stopping a timer; reduced by electronic timing
- Unpredictable variation; reduced by repeating measurements
- Consistent bias in one direction; reduced by calibration
- Reading a scale from the wrong angle; reduced by viewing straight on
Data analysis transforms raw measurements into scientific understanding. The process involves organising data, visualising patterns, performing calculations, and drawing evidence-based conclusions.
Data tables should have clear column headings that include units. Raw data should never be altered - if you make calculations, put them in separate columns. Include all repeated trials, not just averages.
Graphs are the most powerful tool for revealing patterns. Choose appropriate scales so the data fills most of the graph area. Plot the independent variable on the x-axis and dependent variable on the y-axis. Draw a line of best fit (for linear relationships) or a smooth curve (for non-linear relationships) that shows the overall trend without connecting every point.
Conclusions should state what the data shows, explain the pattern using scientific concepts, and acknowledge limitations. A good conclusion does not overclaim - it matches the confidence level of the evidence.
A student investigates how string length affects pendulum period. They collect data for lengths from 10 cm to 100 cm and plot period versus length. The graph shows a curve, not a straight line. Squaring the period and plotting T2 versus length produces a straight line through the origin. The student concludes that T2 is proportional to length, which matches the theoretical formula T = 2 * pi * sqrt(L/g). They acknowledge that their timer measurements had about 5% uncertainty, and that air resistance was ignored in the theoretical model.
Australian data science: The Australian Bureau of Statistics runs CensusAtSchool, a program that teaches students data analysis using real census data. Students collect their own data, enter it into a national database, and analyse patterns across Australia. This program develops statistical literacy and connects classroom investigations to authentic large-scale data analysis - skills increasingly important in science, business, and government.
If my results match the theory, my experiment was good; if they do not, my experiment was bad. This is false. Results may differ from theory for many legitimate reasons: approximations in the theory, uncontrolled real-world factors, or measurement limitations. The goal of practical work is not to confirm textbook answers but to learn how real measurements relate to idealised models. Discrepancies between theory and experiment are often more educationally valuable than perfect agreement.
1. In an experiment to test how the length of a pendulum affects its period, identify the independent variable, dependent variable, and two variables that should be controlled.
2. Why is it important to repeat measurements in a scientific investigation?
- Changing more than one variable at a time. — If you change multiple variables, you cannot tell which one caused the change in results. Only change the independent variable.
- Recording results without units. — Numbers without units are meaningless. Always include the correct unit for every measurement.
📓 Copy Into Your Books
▼Independent Variable
The variable you deliberately change in an experiment.
Dependent Variable
The variable you measure or observe to see the effect of your changes.
Controlled Variables
Variables that are kept constant to ensure a fair test.
Reliability
Repeat measurements and calculate an average to improve reliability and identify anomalies.
You learned how to plan and conduct a fair scientific investigation, identifying variables and controlling conditions.
A student measures the bounce height of a ball dropped from different heights but does not control the surface it bounces on. Why might their results be unreliable?
The hook described CSIRO researchers collecting hundreds of data points per second on wave-energy converters — but pointed out that all that data is meaningless without proper variable control.
Now that you've designed your own investigation and worked through the data skills, what would you say is the most important step in turning raw data into reliable science? Did the CSIRO example change how you think about the difference between data collection and scientific evidence?
1. In a fair test, how many variables should be changed at a time?
2. The variable that is deliberately changed in an experiment is called the:
3. Why should experiments be repeated?
4. A hypothesis is:
5. Which of these is a source of error in a timing experiment?
Design an experiment to investigate how the tension in a string affects the speed of a wave on the string. Identify the independent variable, dependent variable, and at least two controlled variables. (3 marks)
Hint: Consider what you would change, measure, and keep the same.
Explain why it is important to identify and control variables in a scientific investigation. (3 marks)
Hint: Think about what would happen if multiple things changed at once.
Describe two sources of error that could affect the accuracy of a measurement, and suggest how each could be minimised. (3 marks)
Hint: Consider human error, equipment limitations, and environmental factors.