A2 planning, analysis and evaluation

Learning objectives

• Specify apparatus, controls, safe procedures and measurement ranges.
• Linearise relationships and estimate uncertainties.
• Evaluate limitations with specific improvements.

The big idea

A practical plan should let another investigator obtain useful data safely. Specify measurements, controlled conditions, useful ranges, repetitions and calibration checks, rather than only listing apparatus.

Connect plotted gradients and intercepts to physical quantities. Compare uncertainty with the size of an effect. Each suggested improvement should address a specific limitation; requesting better equipment alone does not explain how the result becomes more reliable.

1. A plan must be reproducible

Describe how another investigator would measure each quantity, vary the input and maintain controlled conditions. Choose apparatus range and resolution for the predicted effect, not simply because an instrument is available. Include a safe operating range and concrete precautions. For a spring investigation, specify the original-length reference, alignment, load increments, catch arrangement and how oscillations or settling affect the readings.

2. Linearisation links a graph to a model

Transform a proposed relationship so a graph can test it and yield a useful parameter. If T = 2π√(m/k), a plot of T² against m has gradient 4π²/k. This does not mean every observed line must pass through the origin: additional effective mass or a timing offset can affect the intercept. State the physical meaning of the plotted variables before interpreting gradient units.

3. Uncertainty needs the whole calculation

For small maximum-uncertainty estimates, fractional uncertainty in a power xⁿ is approximately |n| times that in x. Thus 2% timing uncertainty gives about 4% uncertainty in T². If estimating the uncertainty of a best-fit gradient, use the range of plausible lines supported by the plotted uncertainties under the method specified. Instrument resolution is only one contribution; calibration, alignment and environmental variation can also matter.

4. Evaluation should target the limiting effect

A useful improvement explains which uncertainty it reduces and why. Timing many cycles reduces the fractional influence of reaction time; fixing a pointer can reduce parallax. Repeating measurements can reveal scatter but does not automatically remove a systematic offset. Discuss whether the observed effect is large enough compared with uncertainty to support the model, and identify additional measurements that would distinguish competing explanations rather than merely producing more of the same data.

A method you can reuse

1. State the hypothesis and choose a useful linearised form.
2. Specify measurements, control conditions, range and safe procedure.
3. Record raw readings and uncertainty evidence before processing.
4. Interpret gradient and intercept with units and model assumptions.
5. Propose an improvement tied to the dominant uncertainty or unresolved mechanism.

Equations and relationships

Investigate and explain

Plan a mass–spring investigation and analyse teacher-provided data. Explain how one specific timing or alignment limitation could be reduced.

Use these tools to plan variables and analyse relationships. Simulations cannot replace real instrument handling, calibration, experimental uncertainty or assessed practical work, and their noise settings may omit systematic effects.

Open the interactive investigation

Check your understanding