QCAA Physics common mistakes
QCAA Physics · Understand the traps and how to avoid them.
Writing a final numerical answer with no unit, or with the wrong unit (for example giving an acceleration in joules, or dropping the vector direction on a velocity).
Under time pressure candidates treat the number as the answer and copy it straight from the calculator, so unit tracking is skipped; they also confuse the unit of the quantity they computed with the unit of the quantity asked for.
What to do instead
Carry units through every line of working and cancel them as if they were algebraic symbols; before writing the final line, re-read the question stem and match the unit to the quantity requested, including direction or sign where the quantity is a vector. Train by marking your own practice answers against the unit alone, ignoring the number.
Substituting numbers into a formula before rearranging it, then making an algebra slip that cannot be traced in the working.
Candidates learn formulas in their substituted form and rush to get a number on the page; the rearrangement step feels like wasted time, and once numbers are in, errors are hard to spot.
What to do instead
Rearrange symbolically first, write the rearranged equation as its own line, then substitute. This gives an unbroken line of reasoning that earns method marks even when the arithmetic fails, and makes dimensional checking possible at the symbolic stage.
Reading a value off a graph or table without accounting for the scale, the axis origin, or the uncertainty in the plotted points, and quoting far too many significant figures.
Candidates treat graph reading as trivial and default to copying every digit shown by a calculator, forgetting that the precision of the answer is limited by the precision of the data and the resolution of the grid.
What to do instead
State the resolution of each axis or instrument, take the reading to that resolution, and round the final answer to the least number of significant figures in the given data. Practise by reading the same point twice from a printed grid and checking your two readings agree within one grid division.
Designing an experiment that changes two variables at once, or that has no identified controlled variables, so the data cannot support a causal claim.
Candidates focus on getting a measurable result and treat controls as paperwork; they also copy a familiar practical procedure without adapting it to the specific research question being tested.
What to do instead
Before writing the method, list the independent variable, the dependent variable, and at least three variables to hold constant, and state how each will be held constant. Justify each control by explaining what confounding effect it removes; if a control cannot be justified, it probably does not belong in the design.
Presenting a research investigation that summarises sources one after another without evaluating their reliability, validity, or relevance to the stated research question.
Candidates equate research with collecting information, so the report becomes a literature summary; they also avoid criticising sources because criticism feels negative rather than analytical.
What to do instead
For each source, record the author's expertise, the publication venue, the date, whether the claims are peer reviewed, and any conflict of interest, then state explicitly how that source answers part of the research question. Build a short evidence table so the evaluation is visible and each claim in the discussion can be traced to a judged source.
Treating a vector as a scalar by adding magnitudes directly, for example adding perpendicular forces or velocities as plain numbers.
Everyday language uses 'speed' and 'force' as single numbers, and candidates default to arithmetic addition because it is faster than drawing a diagram.
What to do instead
Draw a free-body or vector diagram before any calculation involving more than one direction, resolve into components, and add components axis by axis. Check the reasonableness of the resultant by comparing it with the largest single magnitude, which it can never exceed when vectors are added.
Describing a trend in words without quantifying it, such as saying the value 'increases' when the relationship is linear and could be stated with a gradient and intercept.
Candidates are trained to spot patterns rather than extract parameters, and they fear committing to a numerical relationship that might be slightly off.
What to do instead
For every linearised graph, state the gradient with its unit, the intercept with its unit, and the physical meaning of each. Use the phrase pattern 'as X increases by one unit, Y changes by [gradient] [unit]' so the description carries quantitative content.
Leaving uncertainty analysis as a single percentage error with no indication of which measurement dominates, or omitting uncertainty entirely from a conclusion.
Candidates see uncertainty as an add-on to satisfy the marking guide rather than as the tool that decides whether a result supports a hypothesis.
What to do instead
Propagate uncertainty through the calculation, then compare the contribution of each input and name the dominant source. Finish by comparing the experimental value with the accepted or theoretical value in terms of the stated uncertainty, and say plainly whether they agree within that uncertainty.
Quoting a formula or physical constant that is not on the supplied data sheet, or misremembering a constant's value or unit.
Candidates memorise a handful of constants from practice and assume they will be provided, and they rarely check the data sheet during the paper because they are afraid of losing time.
What to do instead
During the reading time, scan the data sheet and note which constants and formulas are supplied and which must be recalled. For recalled constants, learn the value together with its unit and typical magnitude, and sanity-check any computed result against a known physical scale.
Stating a conclusion that goes beyond the evidence, for example claiming a relationship is proven when the data cover only a narrow range or when systematic error was identified but not corrected.
Candidates want a decisive answer and believe a confident conclusion scores better; they also fail to carry identified limitations through to the final paragraph.
What to do instead
Write the conclusion as a claim plus its scope plus its caveat: state the relationship observed, the range of the independent variable over which it holds, and the main limitation that constrains generalisation. If a systematic error was found, say how it would shift the result and whether the conclusion survives that shift.