Welcome to Working Scientifically
Science is not just a list of facts to memorise—it is a way of asking questions, testing ideas, and finding out how the universe works. In your GCSE Combined Science: Trilogy exams, at least 15% of the total marks across all six papers (Biology, Chemistry, and Physics) test your understanding of practical skills and working scientifically.
Whether you are investigating the rate of a chemical reaction, measuring the resistance of a wire, or looking at plant cells under a microscope, the rules of scientific enquiry remain the same. This guide breaks down every key skill, term, and calculation step-by-step so you can tackle exam questions with total confidence!
1. Development of Scientific Thinking
Scientific Models and Theories
A hypothesis is a scientific idea proposed to explain certain facts or observations. When scientists test hypotheses through experiments and gather evidence, they develop theories and models.
Scientific theories and models change over time. When new experimental evidence is discovered that does not fit the current model, scientists must adapt or replace the model.
Example: Think of how our understanding of the atom changed over time from solid spheres to the plum pudding model, then to the nuclear model, and finally to the modern electron shell model as new experimental results emerged.
Scientists use different types of models:
• Representational models: Physical or visual diagrams that simplify complex systems (e.g. diagrams of cell structures or electric circuits).
• Spatial models: Help us visualise 3D structures (e.g. models of molecules or the solar system).
• Descriptive models: Word-based descriptions of how processes work.
• Mathematical / Computational models: Equations and computer simulations used to predict real-world outcomes (e.g. climate change models).
Science, Society, and Ethics
Science can explain how things work, predict potential consequences, and develop new technologies. However, science alone cannot make ethical, moral, or social value judgements.
Example: Science can explain how to generate electricity from nuclear power and calculate the risks of nuclear waste, but society, politicians, and communities must decide whether building a nuclear power station is ethically and economically acceptable.
Hazards, Risks, and Precautions
Every practical investigation requires a risk assessment. Students often mix up hazards and risks, but they have very precise meanings:
• Hazard: Anything that has the potential to cause harm (e.g. a hot Bunsen burner flame, concentrated acid, broken glassware).
• Risk: The chance or probability that harm will occur, along with how severe that harm could be.
• Precaution: An action taken to minimise or control the risk.
Everyday Analogy: A puddle of water on a floor is a hazard. The risk is that you might slip, fall, and hurt your wrist. The precaution is putting up a warning sign or mopping the floor dry immediately.
Common lab precautions include wearing safety goggles (for chemical splashes), using heatproof gloves or tongs (for hot apparatus), and working in a fume cupboard (for toxic gases).
Peer Review
Before a scientist's findings are accepted by the wider scientific community, they must go through peer review. This means other independent, expert scientists read the report, check the experimental design, look for bias, and verify whether the conclusions match the data. Other scientists will also try to reproduce the experiment to see if they get the same results.
Section 1 Key Takeaway: Scientific models evolve as new evidence emerges. Science explains how things happen, but ethical decisions involve human values. Always link a hazard to a realistic risk and a practical precaution.
2. Experimental Skills and Variables
The Three Key Variables
To design a valid, fair experiment, you must understand variables. A fair test is one where only the independent variable is allowed to affect the dependent variable.
• Independent Variable (IV): The variable that you (the investigator) choose to change or select.
• Dependent Variable (DV): The variable that is measured or observed for each change in the independent variable.
• Control Variables (CV): All other factors that must be kept constant so they do not influence the outcome.
Memory Trick:
• I change the Independent variable.
• Dependent is the Data you measure.
• Control variables stay Constant.
Examiner Warning: Never write vague answers like "keep the amount of liquid the same" or "keep the environment constant". Always state the exact physical quantity, such as "keep the volume of hydrochloric acid the same" or "keep the temperature of the water bath constant".
Categoric vs Continuous Variables
• Categoric variables: Values that are words, types, or distinct categories (e.g. type of metal, colour of light, blood group).
• Continuous variables: Values that can take any numerical value along a continuous scale (e.g. temperature in \(^\circ\text{C}\), time in \(\text{s}\), mass in \(\text{g}\), length in \(\text{cm}\)).
Selecting Equipment
Choosing the right piece of equipment is crucial for good data:
• Range: The maximum and minimum values an instrument can measure (or the spread of values tested in an experiment, e.g. from \(10\text{ cm}^3\) to \(50\text{ cm}^3\)).
• Resolution: The smallest change in the quantity being measured that produces a noticeable change in the reading on the instrument.
Example: A standard school ruler has a resolution of \(1\text{ mm}\) (\(0.1\text{ cm}\)), whereas a digital calliper might have a resolution of \(0.01\text{ mm}\). A standard stopwatch measures to \(0.01\text{ s}\), but a thermometer might measure to \(1^\circ\text{C}\) or \(0.1^\circ\text{C}\).
Section 2 Key Takeaway: Change only one independent variable, measure the dependent variable, and keep all control variables strictly constant using apparatus with appropriate range and resolution.
3. The Language of Measurement
AQA exams use strict definitions from the Association for Science Education (ASE). Mixing up these terms is one of the most common ways students lose marks!
True Value, Accuracy, and Precision
• True value: The value that would be obtained in an ideal measurement with no errors at all.
• Accuracy: A measurement result is accurate if it is judged to be close to the true value.
• Precision: Measurements are precise if repeat values cluster closely together (they have very little spread), regardless of whether they are close to the true value.
The Dartboard Analogy:
• Accurate and Precise: All darts land tightly clustered in the bullseye (close to true value and close to each other).
• Precise but NOT Accurate: All darts land tightly clustered together in the top-left corner far from the bullseye (close to each other, but far from the true value).
• Accurate but NOT Precise: Darts are spread out loosely around the bullseye (the average is near the centre, but the spread is wide).
• Neither Accurate nor Precise: Darts are scattered randomly all over the board.
Repeatability vs Reproducibility
Both describe precision, but under different conditions:
• Repeatable: The precision obtained when the same investigator using the same equipment and method in the same laboratory repeats the experiment and obtains similar results.
• Reproducible: The precision obtained when different investigators using different equipment or methods repeat the investigation and obtain similar results.
Memory Trick: Repeatable = Redo it yourself. Reproducible = Peer / Person else does it.
Interval
• Interval: The gap or quantity between successive readings (e.g. taking a temperature reading every \(10\text{ s}\), or testing acid concentrations at intervals of \(0.5\text{ mol/dm}^3\)).
Section 3 Key Takeaway: Accuracy means close to the true value; precision means repeats are close to each other. Repeatable means you can do it again; reproducible means someone else can confirm it.
4. Errors, Uncertainties, and Calculations
Types of Measurement Error
A measurement error is the difference between a measured value and the true value. Errors are not "blunders" or "mistakes"; they are features of measuring in the real world.
1. Random Error:
• Causes readings to be spread unpredictably above and below the true value.
• Caused by human reaction time variations, minor fluctuations in room temperature, draughts, or reading a scale from slightly different angles.
• How to reduce random errors: Take repeat readings, identify and discard anomalies, and calculate a mean.
2. Systematic Error:
• Causes readings to differ from the true value by a consistent amount or proportion each time.
• Caused by flawed equipment, incorrect calibration, or a fundamental problem with the experimental setup.
• Zero Error: A specific type of systematic error where an instrument gives a reading when the true value is zero (e.g. a top-pan balance reading \(0.05\text{ g}\) before anything is placed on it).
• How to deal with systematic errors: Taking repeat readings and calculating a mean will NOT fix a systematic error! You must recalibrate the instrument, subtract the zero error from all readings, or redesign the method.
Anomalies (Outliers)
An anomaly is a measurement that does not fit the pattern of the rest of the data and is judged not to be part of the natural random variation.
• When calculating means, you must exclude (ignore) anomalies from your calculation!
Calculating the Mean
To calculate the mean of repeat trials:
\(\text{Mean} = \frac{\text{Sum of concordant (non-anomalous) repeats}}{\text{Number of valid trials}}\)
Calculating Uncertainty
Every measurement has an uncertainty. For a set of repeated measurements in AQA GCSE Science, use the formula:
\(\text{Uncertainty} = \pm \frac{\text{Range}}{2} = \pm \frac{\text{Maximum value} - \text{Minimum value}}{2}\)
Note: Always remove any anomalies before finding the maximum and minimum values!
For a single reading from an instrument, the uncertainty is typically taken as \(\pm \frac{1}{2} \times \text{resolution}\) (or \(\pm \text{resolution}\)).
Step-by-Step Worked Example
A student measures the time taken for a cross to disappear in a reaction. They record the following times: \(32\text{ s}\), \(34\text{ s}\), \(48\text{ s}\), and \(33\text{ s}\).
Step 1: Identify and discard any anomalies.
Looking at the data, \(48\text{ s}\) is much higher than the other values (\(32, 34, 33\)). It is an anomaly. We exclude \(48\text{ s}\).
Step 2: Calculate the mean of valid trials.
\(\text{Mean} = \frac{32 + 34 + 33}{3} = \frac{99}{3} = 33\text{ s}\)
Step 3: Calculate the uncertainty.
\(\text{Range of valid trials} = 34 - 32 = 2\text{ s}\)
\(\text{Uncertainty} = \pm \frac{2}{2} = \pm 1\text{ s}\)
Final Result: The recorded time is \(33 \pm 1\text{ s}\).
Section 4 Key Takeaway: Repeats and means reduce random errors, but cannot fix systematic errors. Always remove anomalies before calculating the mean and uncertainty (\(\pm \frac{\text{Range}}{2}\)).
5. Presenting Data: Tables and Graphs
Designing Data Tables
When presenting experimental data in a table, follow these official conventions:
1. The independent variable belongs in the first (left-hand) column.
2. The dependent variable goes in the next columns, with sub-columns for repeat trials (e.g. Trial 1, Trial 2, Trial 3) and a final column for the Mean.
3. Every column header must state the quantity name and the unit, separated by a forward slash or placed in brackets (e.g. \(\text{Time } / \text{ s}\) or \(\text{Length } (\text{cm})\)).
4. Never write units inside the data cells—write only numbers in the table body.
Choosing the Correct Graph
• If the independent variable is categoric \(\rightarrow\) Draw a Bar Chart.
• If the independent variable is continuous \(\rightarrow\) Draw a Scatter Graph / Line Graph with a line of best fit.
Graph Plotting Rules
• Axes: Independent variable on the horizontal (\(x\)-axis), dependent variable on the vertical (\(y\)-axis).
• Labels: Clearly label both axes with the quantity and unit (e.g. \(\text{Current } / \text{ A}\)).
• Scale: Choose sensible, linear scales (e.g. going up in \(1\text{s}\), \(2\text{s}\), \(5\text{s}\), or \(10\text{s}\)). The plotted points must occupy more than 50% of the grid area.
• Plotting: Plot points accurately using small neat crosses (\(\times\)).
• Line of Best Fit: Draw a single smooth straight line (using a ruler) or a smooth curve (freehand). Balance the points so there is an even spread of points above and below the line. Do not join points dot-to-dot!
Describing Relationships on Graphs
Exam questions frequently ask you to describe the relationship shown on a graph:
• Linear: The line of best fit is a straight line (\(y = mx + c\)).
• Directly Proportional (\(y \propto x\)): The line of best fit is a straight line that passes through the origin \((0,0)\). This means that if the independent variable doubles, the dependent variable also doubles.
• Non-linear: The graph is a curve (e.g. the rate of reaction decreases as reactants are used up).
• Inversely Proportional: As \(x\) increases, \(y\) decreases such that \(x \times y = \text{constant}\) (produces a downward curve).
Examiner Warning: Never say a relationship is "directly proportional" just because the line goes upwards! It is only directly proportional if it is a straight line AND goes directly through \((0,0)\).
Section 5 Key Takeaway: Independent on the \(x\)-axis, dependent on the \(y\)-axis. Use a bar chart for categories, and a scatter graph with a smooth line of best fit covering \(>50\%\) of the grid for continuous data.
6. Top Exam Pitfalls & Misconceptions
Avoid these common traps that catch students out every year:
• Trap 1: Confusing Accuracy and Precision
Incorrect: "My results are accurate because all three repeats gave 12.1."
Correct: "My results are precise because all three repeats are close together."
• Trap 2: Confusing Repeatable and Reproducible
Incorrect: "The experiment is reproducible because I repeated it three times."
Correct: "The experiment is repeatable because I got similar results when repeating it, and reproducible because another student got similar results using their own equipment."
• Trap 3: Trying to fix Systematic Errors with Repeats
Incorrect: "To fix the zero error on the balance, repeat the experiment and calculate a mean."
Correct: "To fix a zero error, tare (reset) the balance or subtract the zero reading from all measurements. Repeating trials only reduces random errors."
• Trap 4: Including Anomalies in Calculations
Incorrect: Adding all values including the outlier when calculating the mean.
Correct: Identify the anomaly, cross it out, and divide only by the number of valid trials.
• Trap 5: Drawing Jagged Lines on Graphs
Incorrect: Playing connect-the-dots between data points.
Correct: Draw a single continuous line of best fit (straight with a ruler or smooth curve freehand) ignoring anomalies.