Development of practical skills in physics

Q: What is the difference between accuracy and precision in physics measurements?

Accuracy is how close a measurement is to the true (or accepted) value, whereas precision is the closeness of agreement between repeated independent measurements of the same quantity.

Q: What distinguishes repeatability from reproducibility?

Repeatability is precision obtained by the same operator using the same equipment in the same lab over short time intervals, while reproducibility is precision obtained by different operators using different equipment or in different laboratories.

Q: How do random errors and systematic errors differ, and how can their effects be mitigated?

Random errors cause readings to scatter unpredictably above and below the true value and are reduced by taking repeated readings and calculating a mean. Systematic errors shift all readings consistently in one direction and must be resolved by recalibrating equipment or correcting zero errors.

Q: What is the difference in uncertainty between a single reading and a two-point measurement?

A single reading has an uncertainty of op pm 0.5 \times \text{smallest scale division}\) (or $\pm 1$ in the least significant digit for digital displays), whereas a two-point measurement has an uncertainty of $\pm 1 \times \text{smallest scale division}$.

Q: How must table column headings and raw data be formatted according to OCR conventions?

Column headers must use the solidus convention: quantity symbol / unit (e.g., $t\text{ / }\text{s}$). Raw readings in each column must be recorded consistently to the same decimal places matching instrument resolution.

Physics A - H156 · Cambridge OCR AS Level · 11 min read · Updated: Aug 27, 2026

Module 1: Development of Practical Skills in Physics

Welcome to the essential toolkit for every physicist! While physics is famous for its grand theories about the universe, every single law and formula you study was discovered through careful experiments. In your OCR AS Level Physics A (H156) exams, at least 15% of the total marks across Component 01 (Breadth in physics) and Component 02 (Depth in physics) test your practical skills. Mastering this chapter gives you an immediate boost across your entire written examination!

Don't worry if experimental design or calculating uncertainties feels a bit intimidating at first. We will break down every concept step-by-step with clear real-world examples, simple memory aids, and examiner-approved techniques.

1. Key Scientific Vocabulary (Precision, Accuracy, and Errors)

Examiners love testing scientific terminology. Using these terms accurately is one of the easiest ways to secure top marks.

Accuracy vs. Precision

• Accuracy: A measurement is described as accurate if it is close to the true (or accepted) value of the quantity being measured.
• Precision: The closeness of agreement between independent, repeated measurements of the same quantity under stipulated conditions. A small spread in repeated readings indicates high precision.

The Dartboard Analogy: Imagine throwing three darts at a bullseye. If all three hit right around the bullseye, your throws are both accurate and precise. If all three cluster tightly together way off in the top-left corner, your throws are precise (close to each other) but not accurate (far from the target).

Repeatability vs. Reproducibility

• Repeatability: Precision obtained when independent test results are gathered using the same method, on identical test items, in the same laboratory, by the same operator, using the same equipment, within short intervals of time.
• Reproducibility: Precision obtained when test results are gathered using the same method, on identical test items, in different laboratories, with different operators, using different equipment.

Memory Trick: Repeatable = Re-done by the same person. Reproducible = Produced by a partner or new person.

Resolution

• Resolution: The smallest change in the quantity being measured that causes a perceptible change in the instrument's reading (e.g., a standard metre ruler has a resolution of $1\text{ mm}$, whereas digital calipers often have a resolution of $0.01\text{ mm}$).

Random Error vs. Systematic Error

• Random Error: Unpredictable variations that cause individual readings to deviate randomly above and below the true value. These can be caused by fluctuating ambient temperature or human reaction time. How to fix: Take multiple repeat readings and calculate a mean to reduce the effect of random error.
• Systematic Error: Predictable, constant errors that cause measurements to differ consistently in one direction from the true value (always too high or always too low). Examples include a meter with a zero error (not reading zero when disconnected) or a ruler with a worn-off end. How to fix: Re-calibrate apparatus or subtract the zero error. Repeating readings does NOT reduce systematic error!

Common Examiner Trap: Never say "repeating readings makes the measurement more accurate". Repeating readings identifies anomalies and improves precision and the reliability of the mean, but it does not remove systematic errors.

Key Takeaway: Accuracy is closeness to the true value; precision is closeness of repeated readings. Random errors are reduced by averaging repeats; systematic errors shift all readings in one direction and must be corrected by recalibrating equipment.

2. Planning and Implementing Experiments

Identifying Variables

When designing an experiment to investigate a physical relationship, you must clearly identify:
• Independent Variable: The variable that you deliberately change (e.g., length of a wire $L$).
• Dependent Variable: The variable that you measure as a result of changing the independent variable (e.g., resistance $R$).
• Control Variables: All other variables that must be kept constant to ensure a fair test (e.g., temperature of the wire, cross-sectional area, material).

Safety and Risk Mitigation

In planning questions, you must specify the hazard, the associated risk, and a realistic precaution:
• Hot wires / heating elements: Risk of skin burns. Precaution: Switch off power between readings and allow apparatus to cool before touching.
• Falling masses: Risk of foot injury. Precaution: Place a sand tray or cushion underneath hanging masses and keep feet clear.
• Lasers / intense light: Risk of eye damage. Precaution: Do not look directly into the beam, wear appropriate safety goggles, and place warning signs.

Instrument Uncertainties: Readings vs. Measurements

• Reading: A single scale point is observed (e.g., digital balance, thermometer, digital voltmeter).
$$\text{Uncertainty} = \pm 0.5 \times \text{smallest scale division}$$ (For digital displays, it is $\pm 1$ in the least significant digit).
• Measurement: The difference between two separate scale readings (e.g., measuring length with a ruler requires aligning $0\text{ mm}$ at one end and reading the scale at the other end).
$$\text{Uncertainty} = \pm 1 \times \text{smallest scale division}$$ Example: A metre ruler marked in millimeters has an uncertainty of $\pm 1\text{ mm}$ (or $2 \times 0.5\text{ mm}$).
• Human Reaction Time: For manually operated stopwatches, the timing uncertainty is dominated by human reaction time, typically taken as $\pm 0.1\text{ s}$ to $\pm 0.2\text{ s}$.

Key Takeaway: Valid experiments keep control variables constant. A single reading has an uncertainty of $\pm 0.5 \times \text{division}$, while a two-point measurement (like a ruler) has an uncertainty of $\pm 1 \times \text{division}$.

3. Processing Data and Calculating Uncertainties

Calculating Uncertainties in Values

• Absolute Uncertainty: The interval around a measured value within which the true value is expected to lie (e.g., $25.0 \pm 0.5\text{ cm}$).
• Fractional Uncertainty:
$$\text{Fractional Uncertainty} = \frac{\text{Absolute Uncertainty}}{\text{Measured Value}}$$ • Percentage Uncertainty:
$$\text{Percentage Uncertainty} = \left(\frac{\text{Absolute Uncertainty}}{\text{Measured Value}}\right) \times 100\%$$

Uncertainty from Repeat Readings

When you have a set of repeated measurements for the same quantity, first discard any obvious anomalies, then use:
$$\text{Absolute Uncertainty} = \frac{\text{Range}}{2} = \frac{\text{Maximum Value} - \text{Minimum Value}}{2}$$

Worked Example: Three repeated timings for an oscillation are recorded: $3.42\text{ s}$, $3.48\text{ s}$, and $3.44\text{ s}$.
1. Calculate the mean: $\text{Mean} = \frac{3.42 + 3.48 + 3.44}{3} = 3.45\text{ s}$
2. Calculate the absolute uncertainty: $\text{Uncertainty} = \frac{3.48 - 3.42}{2} = \frac{0.06}{2} = \pm 0.03\text{ s}$
3. Calculate the percentage uncertainty: $\frac{0.03}{3.45} \times 100\% = 0.87\%$
4. Final recorded value: $(3.45 \pm 0.03)\text{ s}$

Rules for Combining Uncertainties

When you use measured values in formulas, their uncertainties combine (propagate). Follow these three golden rules:

Rule 1: Addition or Subtraction ($y = a + b$ or $y = a - b$)
Always ADD ABSOLUTE uncertainties:
$$\Delta y = \Delta a + \Delta b$$ Example: If initial temperature $\theta_1 = (20.0 \pm 0.5)\text{ }^{\circ}\text{C}$ and final temperature $\theta_2 = (45.0 \pm 0.5)\text{ }^{\circ}\text{C}$, the temperature change $\Delta \theta = 45.0 - 20.0 = 25.0\text{ }^{\circ}\text{C}$. The absolute uncertainty is $0.5 + 0.5 = \pm 1.0\text{ }^{\circ}\text{C}$. So, $\Delta \theta = (25.0 \pm 1.0)\text{ }^{\circ}\text{C}$.

Rule 2: Multiplication or Division ($y = a \times b$ or $y = \frac{a}{b}$)
Always ADD PERCENTAGE uncertainties:
$$\% \Delta y = \% \Delta a + \% \Delta b$$ Example: A potential difference $V = (12.0 \pm 0.6)\text{ V}$ gives a current $I = (2.0 \pm 0.2)\text{ A}$.
• $\% \Delta V = \frac{0.6}{12.0} \times 100\% = 5.0\%$
• $\% \Delta I = \frac{0.2}{2.0} \times 100\% = 10.0\%$
• Resistance $R = \frac{V}{I} = \frac{12.0}{2.0} = 6.0\,\Omega$
• Total percentage uncertainty in $R = 5.0\% + 10.0\% = 15.0\%$
• Absolute uncertainty in $R = 15.0\% \times 6.0\,\Omega = 0.9\,\Omega$. Final result: $R = (6.0 \pm 0.9)\,\Omega$.

Rule 3: Powers ($y = a^n$)
MULTIPLY the percentage uncertainty by the power index $n$:
$$\% \Delta y = n \times \% \Delta a$$ Example: The radius $r$ of a sphere has a percentage uncertainty of $2\%$. The volume is given by $V = \frac{4}{3}\pi r^3$. Since $r$ is raised to the power $3$, the percentage uncertainty in $V = 3 \times 2\% = 6\%$.

Key Takeaway: Add absolute uncertainties when adding/subtracting quantities. Add percentage uncertainties when multiplying/dividing quantities. Multiply percentage uncertainty by the exponent when raising a quantity to a power.

4. Tables, Graphs, and Error Analysis

OCR Table Presentation Conventions

• Column Headers: Must state the physical quantity symbol and the unit separated by a forward slash (solidus), e.g., $t\text{ / }\text{s}$, $V\text{ / }\text{V}$, or $v^2\text{ / }\text{m}^2\,\text{s}^{-2}$. Avoid writing units in brackets like $t\text{ (s)}$, as OCR specifically requires the solidus convention.
• Raw Data Consistency: All raw readings in a single column must be recorded to the same number of decimal places (matching the resolution of the instrument).
• Calculated Values: Should be quoted to the same number of significant figures as the least precise raw measurement used to calculate them (or at most one more).

Graph Plotting Guidelines

• Grid Size: Scales must be chosen so that the plotted data points occupy at least 50% of the available grid space along both the $x$-axis and $y$-axis.
• Plotting Points: Mark each point accurately using a sharp pencil with a small cross ($\times$) or a small encircled dot ($\odot$) to within half a small square.
• Error Bars: Draw vertical and/or horizontal bars through each data point representing $\pm \text{absolute uncertainty}$ in that variable.

Line of Best Fit and Worst Acceptable Line

When analyzing graphs with error bars:
1. Line of Best Fit (LOBF): Draw a straight line passing smoothly through the centre of the error bars with an even balance of points on either side.
2. Worst Acceptable Line (Worst Fit): Draw the steepest or shallowest possible straight line that still passes through all the error bars of valid data points.

To determine the uncertainty in the gradient and $y$-intercept:
$$\text{Uncertainty in Gradient} = |\text{Gradient of Best Fit} - \text{Gradient of Worst Fit}|$$ $$\text{Uncertainty in } y\text{-Intercept} = |y\text{-intercept of Best Fit} - y\text{-intercept of Worst Fit}|$$

Worked Example:
• Gradient of Best Fit = $4.25\text{ m s}^{-1}$
• Gradient of Worst Fit = $4.05\text{ m s}^{-1}$
• Absolute Uncertainty in Gradient = $|4.25 - 4.05| = 0.20\text{ m s}^{-1}$
• Final Gradient Quote: $(4.3 \pm 0.2)\text{ m s}^{-1}$ (quoted to matching decimal place!).

Key Takeaway: Tables must use the quantity / unit solidus format. Data points must cover over 50% of the graph grid. The uncertainty in a gradient equals the difference between the best-fit gradient and worst-fit gradient.

5. Evaluation and Drawing Valid Conclusions

Dealing with Anomalies

• An anomaly (outlier) is a reading that does not follow the general trend of the other data points.
• Handling Anomalies: Identify the anomaly, repeat the measurement if possible, and exclude it when calculating the mean and range.

Interpreting Gradients and Intercepts ($y = mx + c$)

In physics questions, you will often need to rearrange a theoretical formula into straight-line form $y = mx + c$:
• Example: For an object accelerating from rest, $s = \frac{1}{2}at^2$.
• Plot $s$ on the $y$-axis and $t^2$ on the $x$-axis.
• Comparing $y = mx + c$ with $s = \left(\frac{1}{2}a\right)t^2 + 0$:
- $y$-variable = $s$
- $x$-variable = $t^2$
- Gradient $m = \frac{1}{2}a \implies a = 2 \times \text{gradient}$
- Theoretical $y$-intercept = $0$.
• Evaluating Zero Errors: If theory predicts a line passing through the origin ($c = 0$), but your plotted graph has a non-zero intercept, this points to an uncorrected systematic error in your measurements!

Suggesting Realistic Improvements

When exam questions ask how to improve an experimental procedure, be specific:
• Instead of: "Use a computer" or "Be more careful."
• Write: "Use a light gate connected to a digital datalogger to record time intervals automatically, eliminating human reaction time error."
• Instead of: "Measure more values."
• Write: "Measure oscillations over $20$ complete cycles instead of $1$ cycle, then divide by $20$ to significantly reduce the percentage uncertainty in the period $T$."

Key Takeaway: Always relate your graph to $y = mx + c$. A non-zero intercept in a directly proportional relationship reveals systematic error. Propose concrete, apparatus-specific improvements rather than vague statements.

6. Summary: Quick Review & Top Exam Tips

• Vocabulary Check: Repeating readings improves precision, not accuracy. Repeatability is within the same lab by the same person; reproducibility is across different labs/people.
• Repeat Uncertainty: Always use $\frac{\text{Range}}{2} = \frac{\text{Max} - \text{Min}}{2}$.
• Uncertainty Operations: Add absolutes for sums/differences; add percentages for products/quotients; multiply percentage by the exponent for powers.
• Table Headers: Strictly use the solidus format: $Quantity\text{ / }Unit$ (e.g., $F\text{ / }\text{N}$, $d\text{ / }\text{mm}$).
• Graph Scales: Points must take up at least $50\%$ of both graph axes. Error in gradient is $|\text{Gradient}_{\text{best}} - \text{Gradient}_{\text{worst}}|$.
• Significant Figures: Express absolute uncertainty to $1\text{ s.f.}$ (e.g., $\pm 0.2$), and match the calculated value to the same decimal place (e.g., $4.3 \pm 0.2$).

Quick check

Can you answer these now?

Open each question to check the key ideas from this chapter.

What is the difference between accuracy and precision in physics measurements?

Accuracy is how close a measurement is to the true (or accepted) value, whereas precision is the closeness of agreement between repeated independent measurements of the same quantity.

What distinguishes repeatability from reproducibility?

Repeatability is precision obtained by the same operator using the same equipment in the same lab over short time intervals, while reproducibility is precision obtained by different operators using different equipment or in different laboratories.

How do random errors and systematic errors differ, and how can their effects be mitigated?

Random errors cause readings to scatter unpredictably above and below the true value and are reduced by taking repeated readings and calculating a mean. Systematic errors shift all readings consistently in one direction and must be resolved by recalibrating equipment or correcting zero errors.

What is the difference in uncertainty between a single reading and a two-point measurement?

A single reading has an uncertainty of op pm 0.5 \times \text{smallest scale division}\) (or $\pm 1$ in the least significant digit for digital displays), whereas a two-point measurement has an uncertainty of $\pm 1 \times \text{smallest scale division}$.

How is absolute uncertainty calculated from a set of repeat readings?

$$\text{Absolute Uncertainty} = \frac{\text{Range}}{2} = \frac{\text{Maximum Value} - \text{Minimum Value}}{2}$$ after discarding any anomalous readings.

What is the rule for combining uncertainties when adding or subtracting physical quantities?

When adding or subtracting quantities ($y = a + b$ or $y = a - b$), always add absolute uncertainties: $$\Delta y = \Delta a + \Delta b$$

What is the rule for combining uncertainties when multiplying, dividing, or raising quantities to a power?

For multiplication/division ($y = a \times b$ or $y = \frac{a}{b}$), add percentage uncertainties: $\%\Delta y = \%\Delta a + \%\Delta b$. For powers ($y = a^n$), multiply percentage uncertainty by the power index: $\%\Delta y = n \times \%\Delta a$.

How must table column headings and raw data be formatted according to OCR conventions?

Column headers must use the solidus convention: quantity symbol / unit (e.g., $t\text{ / }\text{s}$). Raw readings in each column must be recorded consistently to the same decimal places matching instrument resolution.

How is the absolute uncertainty in a gradient determined using a graph with error bars?

$$\text{Uncertainty in Gradient} = |\text{Gradient of Best Fit} - \text{Gradient of Worst Fit}|$$

What does a non-zero $y$-intercept indicate when theoretical equations predict direct proportionality ($y = mx$)?

A non-zero $y$-intercept in a directly proportional relationship indicates the presence of an uncorrected systematic error (such as a zero error in the measuring apparatus).

Ready to test yourself?

Turn these notes into exam-style practice. Get unlimited AI questions on this topic with instant marking and explanations.

Practice This Topic

* The content provided by thinka is generated by AI and may not always be accurate or up-to-date. Please use it as a supplementary resource and verify with official materials.

Put These Notes into Practice

Reading the notes is a great start. Now practice with unlimited AI-generated questions and get instant feedback. 100,000+ students are already improving their grades.

Start Practicing Now View Pricing

Done reading? Test yourself with AI practice questions

Practice This Topic Now