Physics unit guide
Practical physics and data analysis
Uncertainty, graph analysis, linearization and evaluation - the skills behind Paper 3 and the IA.
This unit is worth disproportionately more than the time most students give it. Uncertainty and graph skills appear in Paper 3, in the internal assessment, and scattered through Paper 2, which means marks here are spread across the whole qualification rather than concentrated in one place. It is also the most improvable part of the course, because the techniques are finite and mechanical in a way the conceptual units are not.
Topics in this unit
What gets examined
- Propagating absolute and percentage uncertainties through sums, products and powers.
- Deciding how many significant figures an answer can justify from the precision of the data.
- Linearizing a non-linear relationship so that a straight-line graph tests it.
- Extracting a physical quantity from a gradient or intercept, with its unit.
- Drawing maximum and minimum gradient lines through error bars to find the uncertainty in a gradient.
- Distinguishing random from systematic error, and identifying which a described flaw would produce.
How to revise it
Learn the three propagation rules cold
Add absolute uncertainties when adding or subtracting; add percentage uncertainties when multiplying or dividing; multiply the percentage by the power. Nearly every uncertainty mark in the course comes from these three.
Ask what to plot to get a straight line
If the relationship is T ∝ √L, plot T against √L. Questions are usually phrased as "suggest what should be plotted", and the answer is whatever makes the relationship linear.
Always give the gradient a unit
A gradient is a physical quantity with dimensions. Omitting the unit is a routine and entirely avoidable mark loss.
Evaluate specifically, never generically
"Human error" and "use better equipment" score nothing. Name the measurement, say why it was imprecise, and give an improvement that would actually address it.
Where marks get lost
- Quoting an answer to more significant figures than the least precise measurement allows.
- Adding percentage uncertainties when quantities are added rather than multiplied.
- Drawing a line of best fit that ignores the error bars.
- Calling a systematic error random, or vice versa, in an evaluation.
- Reading a gradient off two data points instead of two points on the drawn line.
Practise this unit
Reading about a unit only goes so far. Work through real IB-style questions on it, with mark schemes.