QCAA Physics IAs Explained: IA1, IA2 and IA3 (2025 Syllabus)

The data test, the student experiment and the research investigation, worked through for Physics: what each is worth, what the top band needs, and how uncertainty, gradients and claims decide your marks.

ISMGenius35 min read
A flat geometric illustration of a dotted projectile arc landing as a coral ball, a straight wire ringed by concentric field circles, and a small graph with a coral line of best fit through error bars between two grey trend lines
On this page

Every QCAA Physics student sits the same three kinds of internal assessment: a 10-mark data test, a 20-mark student experiment and a 20-mark research investigation. They use the same criteria as the other 2025 sciences, but Physics changes what top-band work looks like. Most of your relationships have an equation behind them, so the strongest analysis turns data into a straight line and asks whether its gradient matches the model. Every measurement carries an uncertainty, and the marks follow it through each calculation. And in the research investigation, the claim is about relativity, quantum theory or particles, so the evidence is often a number you can check yourself.

This guide is for Year 12s starting Units 3 and 4 in Term 1 2027. It works through each IA under the current Physics 2025 syllabus (v1.3, January 2026): the conditions, the marks and criteria, what the top band looks like in Physics, and what QCAA's latest reports say students and schools get wrong. For the skills every science shares, such as the structure of a justified modification or the return to the claim, it links to our student experiment guide and research investigation guide rather than repeating them.

The three Physics IAs at a glance

Internal assessment is 50 of your 100 marks. The external exam is the other 50.

TaskMarks and conditions
IA1 Data test, Unit 310 marks, one criterion. 60 minutes plus 5 minutes perusal, supervised, unseen data, formula and data book allowed
IA2 Student experiment, Unit 320 marks, four criteria of 5. About 10 hours of class time, up to 2000 words or 11 minutes
IA3 Research investigation, Unit 420 marks, four criteria of 5. About 10 hours of class time, up to 2000 words or 11 minutes
External exam, Units 3 and 450 marks. Two papers, each 90 minutes plus 5 minutes perusal

Unit 3, Gravity and electromagnetism, has two topics. Gravity and motion covers projectiles, inclined planes, circular motion and orbits. Electromagnetism covers electrostatics, magnetic fields and electromagnetic induction. Unit 4, Revolutions in modern physics, has three: Special relativity, Quantum theory and The Standard Model. IA1 and IA2 draw on Unit 3. IA3 draws on Unit 4. To see how the four results combine, use the subject result calculator.

The syllabus also labels every block of Unit 3 and 4 subject matter with the assessment it feeds. Science understanding "can be assessed in the external assessment". Science as a human endeavour and science inquiry "may be assessed in the internal assessments". Our guide to reading a QCAA syllabus shows where those labels sit.

When each IA falls in 2027

Your school sets the exact dates, and your task sheet is the authority. QCAA's Senior Education Profile calendar sets the deadlines schools work back from.

  1. 27 January

    Term 1 starts and Unit 3 begins. The practicals you run in these weeks are the source of the IA1 data and the base for IA2.

  2. Term 1

    The data test. Schools must enter provisional marks for the first internal assessment by Tuesday 13 April 2027, the second day of Term 2, so expect IA1 in Term 1.

  3. Terms 1 and 2

    The student experiment, usually with a draft checkpoint. Schools that opt in to QCAA's second confirmation event must enter IA2 provisional marks by Tuesday 8 June 2027. Other schools submit them with the remaining internal assessments on Tuesday 17 August 2027.

  4. Terms 2 and 3

    The research investigation on Unit 4. Provisional marks for the remaining internal assessments, including IA3, are due on Tuesday 17 August 2027.

  5. 25 Oct to 16 Nov

    The external assessment block. QCAA sets the day of the Physics papers.

For how the whole cycle runs from task sheet to submission, see the internal assessment timeline.

IA1: the data test

The data test asks you to work with tables, graphs and trend lines from Unit 3 experiments, presented in a form you have not seen before. It is the only internal assessment sat under exam conditions.

About 30% of the marks go to working out unknown scientific quantities or features of datasets, about 30% to identifying trends, patterns, relationships, limitations or uncertainty, and about 40% to drawing conclusions based on analysis. In Physics, that means reading values off a trend line and substituting them into a formula, finding a gradient and its uncertainty, naming a relationship such as inverse square from the numbers, and saying what a gradient tells you about a physical quantity.

QCAA's own sample data test for this syllabus shows the range. Its three datasets are the force between two bar magnets at separations up to 0.050 m, the range of a golf ball launched at angles from 10° to 80°, and the force on a current-carrying wire in a magnetic field, given as a graph with its trend line and its maximum and minimum trend lines. The last question asks for the magnetic field strength with its uncertainty.

How the data test is marked

There are no criteria to balance. Your raw score on the paper becomes a percentage, and the percentage becomes a mark out of 10 through fixed cut-offs. The top mark needs "consistent demonstration, across a range of scenarios", and more than 90% of the paper.

Your raw scoreIA1 mark
More than 90%10
More than 80%9
More than 70%8
More than 60%7
More than 50%6
More than 40%5
More than 30%4
More than 20%3
More than 10%2
More than 1%1
1% or less0

Read "more than" literally. QCAA's 2025 Physics report uses the example itself: 16 out of 20 is 80%, which is not more than 80%, so it earns 8. QCAA's sample paper is out of 20, so each raw mark there is worth 5%, half an IA mark.

What full-marks answers do

  • Quote the data that shows the relationship. QCAA's 2025 report shows a two-mark "identify the relationship" question marked as one mark for the relationship and one for using appropriate data. The excerpt it chose: as the separation doubles from 2 m to 4 m, the force approximately quarters, so the relationship is inverse square.
  • Read from the trend line, not a data point. In the circular motion excerpt QCAA's report chose, the marking scheme expected values identified on the trend line, then substituted into the formula.
  • Uncertainty of the mean is half the range. For three trials of 1.52 s, 1.47 s and 1.55 s, the mean is 1.51 s and the absolute uncertainty is ±(1.55 − 1.47) ÷ 2 = ±0.04 s, a percentage uncertainty of about 2.6%. Both formulas are in the formula and data book.
  • Get gradient uncertainty from the maximum and minimum trend lines. It is half the difference between their gradients. The intercept's uncertainty is half the difference between their intercepts.
  • Powers multiply the percentage uncertainty. The syllabus rule is to add absolute uncertainties when you add or subtract, and add percentage uncertainties when you multiply, divide or raise to a power. If r is uncertain by 2.5%, r3 is uncertain by 7.5%.
  • Supply the unit and the precision yourself. QCAA's June 2026 endorsement report asked schools to drop response boxes, embedded units, decimal-place cues, formulas already in the formula book, and cues such as "refer to Graph 1". Expect to find the evidence, choose the formula and write the unit on your own.

Here is a four-mark item of the kind you will meet, with an original dataset. A simulation tracks five moons orbiting an invented planet. A graph of the square of each moon's orbital period against the cube of its orbital radius is a straight line through the origin. The trend line's gradient is 2.50 × 10−14 s2 m−3, and the maximum and minimum trend lines have gradients of 2.60 × 10−14 and 2.40 × 10−14 s2 m−3. The question: Draw a conclusion about the mass of the planet, including its uncertainty. Show your reasoning.

2 of 4 marks

"M = 4π2 ÷ (G × gradient) = 4π2 ÷ (6.67 × 10−11 × 2.50 × 10−14) = 2.37 × 1025 kg."

4 of 4 marks

"Kepler's third law gives T2/r3 = 4π2/GM, so the gradient of T2 against r3 is 4π2/GM. The gradient's uncertainty is ±(2.60 − 2.40) × 10−14 ÷ 2 = ±0.10 × 10−14 s2 m−3, which is 4.0%. M = 4π2 ÷ (6.67 × 10−11 × 2.50 × 10−14) = 2.37 × 1025 kg, and dividing by the gradient carries its 4.0% uncertainty to M. The planet's mass is (2.37 ± 0.09) × 1025 kg."

The first answer finds the right number but never says why the gradient gives the mass, and ignores the uncertainty the question asked for. QCAA's sample marking scheme for its four-mark magnetic field question is built the same way: a mark each for the gradient's percentage uncertainty, the physical meaning of the gradient, the calculated value, and the conclusion with its uncertainty. Our guide to what QCAA's cognitive verbs ask for covers draw a conclusion and the other verbs on the paper.

How IA1 prepares you for the external exam

The data test is assessed on three of the external exam's four objectives: applying understanding, analysing data and interpreting evidence. The exam adds describing ideas and findings, covers Units 3 and 4, and runs as two 90-minute papers. QCAA provides the formula and data book in both papers, so the book you practise with for IA1 is the one you will use in November. Our guide to QCAA formula sheets lists what the Physics book has and what it leaves out, such as right-hand rules, vector components and a table of particles.

The syllabus also names practicals the exam can come back to. Among the Unit 3 subject matter that can be assessed externally are three dot points that begin "Interpret data relating to": the horizontal distance travelled by an object projected at various angles, the force acting on a conductor in a magnetic field, and the strength of a magnet at various distances. Unit 4 adds data related to the photoelectric effect. Treat those practicals as exam preparation, not just class activities.

QCAA's 2025 subject report shows the overlap. Students responded well to analysis questions that required them to use data to identify trends and determine unknown values, and to calculation questions when they communicated their mathematical reasoning. One 2025 question asked for Saturn's orbital period from Earth's orbital data. Effective responses recognised Kepler's third law, converted Earth's orbital period into seconds and set out their reasoning. The report asked for two things next time: describe and explain answers that go beyond basic descriptions, and an understanding of the concepts and contexts of formulas, not just their symbols.

IA2: the student experiment

You modify, which means refine, extend or redirect, an experiment relevant to Unit 3 to address your own related hypothesis or question. The base is a practical or simulation performed in class.

  • Forming5 marks
  • Finding5 marks
  • Analysing5 marks
  • Interpreting and Evaluating5 marks

Choose a practical with a model to test

The Unit 3 investigations the syllabus lists are: the horizontal distance of an object projected at various angles; the parallel component of an object's weight down an inclined plane at various angles; the net forces on an object in horizontal circular motion on a string; orbital radius and mass for orbiting objects, using a simulation; the force on a conductor in a magnetic field; the strength of a magnet at various distances; electrostatic charge on materials such as trickling water; and induction, with a magnet and coil or an AC generator. QCAA's sample task gives students three of them as the starting point: the projectile, the conductor and the magnet.

Pick one whose data can be tested against an equation in the syllabus. QCAA's June 2026 endorsement report praised tasks with a context that let students collect sufficient primary data to analyse trends, justify conclusions and evaluate uncertainty, such as how the force between two magnets varies with separation. A relationship you can linearise, and a gradient you can compare with a prediction, give Analysing and Interpreting and Evaluating something to work on.

Projectiles: launch from a bench

Launched from floor level, a ball's range peaks at 45° and the graph is symmetrical, so the class data mostly confirm what everyone expected. Launch it from a 0.90 m bench at 3.5 m s−1 instead. The projectile equations predict the peak moves to about 33°, and the symmetry breaks: a 20° launch should land about 1.87 m away and a 50° launch about 1.76 m, where on level ground the 50° launch would go further.

What it tests: a prediction the class version could not make. Measure the launch speed at each angle with a light gate, because a spring launcher can lose speed when it fires steeply upwards.

Circular motion: measure the string, not the radius

In the class practical, a stopper whirls on a string through a tube, with a hanging mass providing the tension. If you measure the radius of the circle, the string's dip below horizontal makes the calculated centripetal force come out below the hanging weight, even with perfect timing. Measure the length L of string from the tube to the stopper instead, and time ten revolutions at six lengths. The dip angle cancels, leaving T2 = (4π2m/Mg)L, where T is the period, m the stopper's mass and M the hanging mass.

What it tests: whether the gradient of T2 against L matches 4π2m/Mg, without measuring the angle at all.

Magnetic fields: from a bar magnet to a wire

A bar magnet's field has no formula in the syllabus, so the class data can show only that the field weakens with distance. A long straight wire's field does: B = μ0I/2πr. Measuring it at six distances turns a description into a test.

What it tests: whether B against 1/r is a straight line with gradient μ0I/2π. The rest of this section follows this example.

Forming: build the modification around the model

The top band of Forming asks for "a considered rationale for the experiment", "justified modifications to the methodology", "a specific and relevant research question", "a methodology that enables the collection of sufficient and relevant data", and "appropriate use of genre and referencing conventions". QCAA's 2025 report spells out what the 2025 wording means. A specific research question "explicitly states the relationship between an independent and dependent variable and is directly related to the data that will be collected". A considered rationale shows how the question was developed, "giving consideration to how the dependent, independent and control variables relate to the topic of investigation".

Relevant, not specific

"How does distance affect the magnetic field around a wire?"

Specific and relevant

"What is the relationship between the perpendicular distance from a long straight wire carrying 4.00 A, from 0.010 m to 0.050 m, and the magnetic field strength measured by a smartphone magnetometer?"

The second names the independent variable, the dependent variable, the range and the controlled current. It also points straight at a prediction: B should be proportional to 1/r, with a gradient of μ0I/2π = 2 × 10−7 × 4.00 = 8.0 × 10−7 T m.

The modification is where most of Forming is decided. Our student experiment guide explains the general rule: a feasible change caps you at 2 to 3, and a justified one names the limitation, the change, and what the change makes possible. In Physics, the strongest justification is usually that the change lets you test a model.

Feasible

"We repeated the class magnet experiment with a wire instead of a bar magnet, at six distances."

Justified

"The class experiment used a bar magnet, whose field has no simple model, so its data could only show that the field weakens with distance. A long straight wire has one, B = μ0I/2πr. This experiment measures the field at six distances from 0.010 m to 0.050 m from a 1.5 m wire carrying 4.00 A, with three trials at each. Each reading is the current-on value minus the current-off value, which removes Earth's field, and the return lead runs at least 1.0 m away, so its own field adds less than 1 μT. Plotting B against 1/r should give a straight line whose gradient can be compared with the predicted 8.0 × 10−7 T m."

QCAA's 2025 report chose an excerpt that sets out its modifications in a two-column table, each change beside its justification, and its praise was that the response "explicitly addresses how the modifications address the reliability and validity of the experiment". That is the same three-part move, made visible.

Finding: risk management with a reason

Finding is worth a quarter of the task, and in Physics its first line is mostly about risk. Conducting a risk assessment and accounting for risks in the methodology is one of the task's specifications in the syllabus.

  1. 4–5

    considered

    considered management of risks/ethical issues/environmental issues; collection of sufficient and relevant raw data; fluent and concise use of scientific language and representations

  2. 2–3

    managed

    management of risks/ethical issues/environmental issues; collection of relevant raw data; competent use of scientific language and representations

  3. 1

    inadequate

    inadequate management of risks/ethical issues/environmental issues; collection of insufficient and irrelevant raw data; simplistic use of language and representations.

What separates "considered" management from plain management is the reason each control is there. Ethical and environmental issues rarely arise in a Unit 3 experiment, so say so briefly and spend the words on the risks.

The wire carried 4.00 A, so the power supply's current limit was set to 5 A, so that a short circuit could not drive the current higher. Current flowed only for the ten seconds each reading took, so the wire could not overheat and soften its insulation. The phone sat in a clamp on a ruler rather than in someone's hand, so no one touched the wire while current flowed, and the supply was switched off before each change of distance, so the circuit was never adjusted while live.

Why this scores: every control comes with its reason. The same list without the reasons describes management, which is the 2–3 band.

The other two lines of Finding are lost by omission. Record every trial, not only the means, with the instrument's uncertainty in each column heading, such as "Distance (m) ±0.0005 m". QCAA's 2025 report describes the language and representations it rewards as correct symbols, units and prefixes, indicators of measurement uncertainty such as significant figures, and graphs with lines of best fit. That sat in the old Communication criterion. Under the 2025 syllabus, it is Finding's third line. Spreadsheet output such as 7.6E-07 is not the convention either: QCAA's own advice to schools prefers "7.0 × 10−15 kg" to "7.0e−15 kg", and the same goes for your report.

Analysing: linearise, then let the gradient work

The top band of Analysing asks for "correct and relevant processing of data", "thorough identification of relevant trends/patterns/relationships", and "thorough and appropriate identification of the uncertainty and limitations of evidence". QCAA's 2025 report says what "thorough" means under the 2025 syllabus. Trends are thoroughly identified when they "accurately describe the processed data and lead to a valid conclusion to the research question". Uncertainties and limitations are thoroughly identified when they are "correctly calculated and appropriately represented", with "calculation of percentage uncertainty in processed data, inclusion of appropriate error bars and/or maximum and minimum trendlines" as its examples.

The sample QCAA chose shows a car released down a ramp at six angles. Acceleration is plotted against the sine of the angle, with maximum and minimum trend lines, the gradient's uncertainty is half the difference between them, and the result is written as an equation with an uncertainty on both the gradient and the intercept. Most Unit 3 models can be made straight the same way, if you choose the axes.

ModelPlotThe gradient is
a = g sin θ (no friction)a against sin θg
F = BIL sin θF against IBL sin θ
B = μ0I/2πrB against 1/rμ0I/2π
F = kq1q2/r2F against 1/r2kq1q2
T2 = (4π2/GM)r3T2 against r34π2/GM
T2 = (4π2m/Mg)LT2 against L4π2m/Mg

Here is the wire experiment's analysis, with original data. Mean field strengths ran from 76.3 μT at 0.010 m to 15.9 μT at 0.050 m, each with an uncertainty of half the range of its three trials.

Figure 3 plots the mean field strength against 1/r. The points lie on a straight line (R2 = 0.9999), so between 0.010 m and 0.050 m the field is inversely proportional to the distance from the wire. The gradient is (7.55 ± 0.23) × 10−7 T m, taking its uncertainty as half the difference between the maximum and minimum trend lines, a 3.0% uncertainty. The predicted gradient, μ0I/2π, is (8.0 ± 0.1) × 10−7 T m once the ±0.05 A uncertainty in the current is allowed for, so the two ranges do not overlap. The offset found below is about the size of the ruler's ±0.0005 m reading uncertainty, but it pushes every reading the same way, which is what makes it systematic rather than random. Every reading falls below its prediction, by 4.6% at 0.010 m but only 0.6% at 0.050 m. A wrong current would lower every reading by the same fraction, so the cause matters more close to the wire. The intercept, (1.0 ± 1.1) μT, is consistent with zero, so subtracting the current-off reading removed Earth's field, and the return lead's field, under 1 μT, is too small to detect here. The ±0.0005 m uncertainty in distance is 5% of the closest distance and 1% of the furthest, which makes the closest readings the least certain.

Why this scores: the relationship is quantified, the gradient carries an uncertainty from the trend lines, the comparison with the model uses ranges rather than single numbers, and every limitation comes from this data rather than from a generic paragraph about human error.

Its 2019 predecessor, Analysis of evidence, was marked down in 3.86% of the IA2 samples QCAA checked in 2025, second only to Research and planning. More on that below.

Interpreting and Evaluating: let the model judge your data

The top band needs "justified conclusion/s linked to the research question", a "justified discussion of the reliability and validity of the experimental process", and "suggested improvements and extensions to the experiment that are logically derived from the analysis of evidence". The middle band accepts reliability and/or validity, and improvements and/or extensions. The top band needs both halves of each.

QCAA's 2025 report describes the Physics responses that earned those judgments. Justified conclusions "referred to the trends, patterns or relationships identified in the analysis of evidence to indicate how the evidence matched with the theoretical concepts identified in the rationale". Validity was discussed "with reference to the accepted and constant values derived from the relevant equation in the rationale". Improvements were linked to reliability and "considered the calculated uncertainties in the raw data".

In the wire experiment, the accepted constant is μ0. The gradient gives μ0 = 2π × gradient ÷ I = 1.19 × 10−6 T m A−1, a percentage error of 5.6% against the accepted 4π × 10−7 ≈ 1.26 × 10−6 T m A−1. A 5.6% error beside a 3.0% uncertainty is the start of the validity discussion: the method is precise, but something is shifting it. The syllabus asks you to keep four pairs of ideas apart.

Reliability

Would repeating it give the same result? The spread of your trials, the uncertainty of each mean, and how closely the points follow the line. Here, half-ranges of 1.6% to 3.8%.

Validity

Did the method measure what the question asked, and does the result match the model? Compare the derived constant with the accepted value, and check what your distance really measured.

Random error

Scatter in both directions. It shows as error bars, and repeated trials reduce it.

Systematic error

A shift in one direction that repeats cannot remove. It shows as a gradient or intercept that misses the model, like this one.

The syllabus also asks you to tell absolute uncertainty from percentage error, and precision from accuracy. The 3.0% is an uncertainty, a statement about precision. The 5.6% is an error against an accepted value, a statement about accuracy.

Now the improvement can come from the analysis. A shortfall that grows close to the wire is the signature of a distance offset: if the true distance from the centre of the wire to the sensor is a little more than the measured one, the closest readings suffer most. Replotting 1/B against r puts any offset into the intercept, and here it comes out at about 0.5 mm, with the corrected gradient matching the prediction. So the improvement is specific: measure from the centre of the wire, and find the sensor's position by sliding the phone across the wire until the reading peaks. "Take more trials" would not help, because repeated trials reduce random error, not systematic error. The extension is the next question the data raises, such as whether B stays proportional to I at a fixed distance.

IA3: the research investigation

You gather evidence related to a research question to evaluate a claim relevant to Unit 4. Your teacher provides a list of claims. QCAA's sample instrument offers six, including "neutrinos are massless particles", "all experimental evidence indicates that light is a particle" and "a machine on the ISS would have a longer lifespan than a machine on Earth".

  • Forming and Finding5 marks
  • Analysing5 marks
  • Interpreting5 marks
  • Evaluating5 marks

There is no Communication criterion any more. Genre and referencing conventions are marked in Forming and Finding, and use of scientific language/representations in Interpreting. For referencing, see our guide to referencing and using sources.

IA3 is also where Unit 4's science as a human endeavour subject matter is assessed. Its dot points make natural sources of claims: how technologies such as GPS assist with accurate navigation, how theories are contested, refined or replaced when new evidence challenges them, and the evidence relating to the Standard Model that supports the Big Bang theory.

From claim to a research question with numbers in it

QCAA's 2025 report asks for a rationale that "clearly breaks the claim into its constituent parts and outlines the logical development from the claim to the research question", and research questions that "include measurable independent and dependent variables". QCAA's sample instrument gives three tests for a draft question: does it consider only one independent variable, does it include an element that can be measured using data, and is its scope suitable for a detailed 1500 to 2000-word answer?

Take an invented claim: "The effects of special relativity only matter for objects travelling close to the speed of light."

As broad as the claim

"Does special relativity matter at everyday speeds?"

Specific and relevant

"How does the time dilation of a moving clock, in microseconds per day, change with speed from 250 m s−1 (an airliner) to 7.66 km s−1 (the International Space Station)?"

"Matter" became a time dilation you can measure in microseconds per day. "Everyday speeds" became a range with a real example at each end. The question can be answered with the syllabus's time dilation equation and published clock measurements inside 2000 words, and your rationale should show that path, step by step.

Analysing: check what each number includes

The top band of Analysing asks for "the identification of sufficient and relevant evidence", "thorough identification of relevant trends/patterns/relationships in evidence", and "thorough and appropriate identification of limitations of evidence". In Physics, the trap is a number that measures more than your question does.

Clock experiments are the classic case. A clock in orbit runs slow because of its speed, which is special relativity, and fast because it sits higher in Earth's gravitational field, which is general relativity. For a GPS satellite, the speed effect is about 7 μs per day slow and the gravity effect about 45 μs per day fast, a net gain of about 38 μs per day. A source that quotes only the net figure is not evidence about special relativity until you separate the two.

Before you claim a trend across sources, make a short table for yourself: what each source measured, at what speed, over what time, with what uncertainty, and whether its number is a measurement or a calculation. A pattern that holds across measurements at different speeds is strong evidence. A trend built from one measurement and three calculations is a limitation you should name, and naming it is how you reach "thorough and appropriate".

Interpreting: reproduce the number before you use it

Interpreting needs "justified scientific argument/s", a "justified conclusion linked to the research question" and "fluent and concise use of scientific language/representations". In Physics, a justified argument often starts with a calculation that reproduces a source's number from the syllabus equation.

For the ISS at 7.66 km s−1, v/c = 2.55 × 10−5. Your calculator will probably show γ = 1 exactly at that speed, because γ differs from 1 by about 3 × 10−10. Use the approximation γ − 1 ≈ v2/2c2, which holds when v is much less than c, and say that you have. Over one day, 86 400 s × (2.55 × 10−5)2 ÷ 2 ≈ 2.8 × 10−5 s, about 28 μs. At 250 m s−1, the same calculation gives about 30 ns per day.

Then compare like with like. Your calculation is the special relativity part only, so set it beside a source's special relativity figure, not its net figure. If they agree, you have an argument. If they don't, the difference is something to explain. Write the numbers in scientific notation with units every time, and keep the syllabus symbols, such as t0 for proper time.

Evaluating: judge the evidence, then return to the claim

  1. 4–5

    justified and extrapolated

    justified discussion of the quality of evidence; extrapolation of credible findings of the research to the claim; suggested improvements and extensions to the investigation that are considered and relevant to the claim

The 2–3 band asks only for a "reasonable description of the quality of evidence", "application of relevant findings of the research to the claim", and improvements and/or extensions. QCAA's 2025 report says a justified discussion of the quality of evidence "includes verifying the credibility, bias, relevance, accuracy and recency of multiple sources", and that findings should address "not only the research question but also the claim". For the quality of evidence, Physics gives you four questions that generic advice misses:

  • Measured or calculated? A web page that works out the dilation from the equation shows what the theory predicts. Only a measurement is evidence that it happens.
  • What does the number include? Special relativity alone, or combined with gravity? A count of muons reaching sea level, or a lifetime in the muon's own frame?
  • Is the effect bigger than its uncertainty? An effect is demonstrated only if it is larger than the uncertainty of the measurement. Say whether it is.
  • How recent is the technology? Atomic clocks have improved by orders of magnitude since the first flown-clock experiments in 1971. A newer measurement at a lower speed can reach further than an older, famous one.

Then go back to the claim. Your findings might show that at 7.66 km s−1 the effect is about 28 μs per day: far too small for a person to notice, yet large enough that satellite navigation must correct for effects of this size. So the claim holds if "matter" means something a person can perceive, and fails if it means something instruments can measure. Saying which reading your evidence supports, and why, is extrapolation. An extension that tests the other end of the claim, such as relativistic momentum in particle accelerators, does more for the evaluation than more sources on clocks. Our research investigation guide explains why the conclusion to the research question and the return to the claim are separate jobs.

What QCAA's 2025 and 2026 reports flag

QCAA's Physics subject report: 2025 cohort, published in January 2026, reviewed the last cohort assessed in Units 3 and 4 under the 2019 syllabus. Its criterion names are the old ones, but it also lists the key differences for the 2025 syllabus, which is the one you will be marked against. When QCAA checked schools' marks through confirmation, it agreed with this share of the samples:

100%IA1 data test
88.65%IA2 student experiment
92.75%IA3 research investigation

Where QCAA changed a mark in IA2 or IA3, it was usually downwards, and some criteria were lowered more often than others.

Criterion (2019 name)Marked downClosest 2025 home
IA2 Research and planning4.59%Forming, and parts of Finding
IA2 Analysis of evidence3.86%Analysing
IA2 Interpretation and evaluation2.17%Interpreting and Evaluating
IA3 Research and planning3.14%Forming and Finding
IA3 Conclusion and evaluation2.42%Evaluating, with the conclusion in Interpreting
IA3 Analysis and interpretation1.45%Analysing and Interpreting

In Physics, the planning criterion was the one most often lowered in both IA2 and IA3. For both tasks, the report notes that the criteria have been reorganised but the way teachers judge each characteristic has not changed. Its advice for the 2025 criteria is built into the sections above: a research question that states the relationship between measurable variables, a rationale that shows how the question grew from the topic or the claim, uncertainties that are calculated and shown, and evaluations that check sources and return to the claim.

QCAA's June 2026 endorsement report, on the instruments schools wrote for the 2026 cohort, changes what the tasks themselves look like. Only 1% of Physics data tests were endorsed at their first application, most often over item construction, so expect your 2027 paper to follow its advice: datasets placed before the questions that use them, no "refer to Graph 1" cues, no question that needs the answer to the one before, and no units or formulas printed for you. For the student experiment, it asks for checkpoints linked to draft data, analysis and evaluation. For the research investigation, it asks for claims explicitly linked to Unit 4.

QCAA's report on the 2026 cohort, the first assessed in Units 3 and 4 under the 2025 syllabus, should follow the same January pattern. Read it in Term 1. Our guide to QCAA subject reports shows where to find it and how to use it.

Are there cut-offs for Physics assessments?

Only in IA1. The data test turns a percentage into a mark through the cut-offs above. IA2 and IA3 have no percentage cut-offs: your teacher matches your response to the descriptors for each criterion, and our guide on how to read an ISMG explains how a mark is chosen within a band. For why an IA mark out of 20 has no letter grade, see what your ISMG mark means.

The A to E cut-offs apply only to your final result out of 100. For the 2025 Physics cohort they were A for 87 to 100, B for 72 to 86, C for 50 to 71, D for 20 to 49 and E for 0 to 19. Of 8,302 students who completed Units 3 and 4, 49.12% received an A, 37.50% a B and 13.01% a C. The boundaries are set each year and can move, so treat them as a guide. For how the internal and external marks combine, see how your QCAA subject result is built.

A Term 1 plan for 2027

  • Before Term 1: revise the uncertainty skills from Units 1 and 2: the uncertainty of the mean, percentage uncertainty and percentage error, maximum and minimum trend lines, and linearising a dataset. Check your calculator is on QCAA's approved list, and print a clean copy of the formula and data book.
  • After each Unit 3 practical: add a line to your gradient log: the model, the straight-line plot, and what its gradient and intercept mean.
  • Before the data test: practise four-mark "draw a conclusion" items until meaning, uncertainty, value and conclusion come out every time, with a unit on every number.
  • When IA2 starts: choose a practical with a model to test, write the risk assessment in the first week, and record every trial with the instrument's uncertainty.

Key takeaways

  • IA1 rewards the whole chain. Say what the gradient means, carry its uncertainty, give the value with a unit, and finish with a conclusion.
  • IA2 starts with a model. Modify a Unit 3 practical so its data can be tested against a syllabus equation, and justify each change by what it lets you test.
  • Linearise, then compare ranges. A straight-line plot, maximum and minimum trend lines, and a predicted value with its own uncertainty tell you whether an error is random or systematic.
  • IA3 turns on the numbers in the claim. Check what each source's figure includes, reproduce it from the equation, and take your findings back to the claim.

Questions about this topic? Ask students taking the same subject in the QCE study group.

Frequently asked questions

What are the three internal assessments in QCAA Physics?

Under the Physics 2025 syllabus, IA1 is a 10-mark data test on Unit 3, IA2 is a 20-mark student experiment that modifies a Unit 3 practical, and IA3 is a 20-mark research investigation that evaluates a claim about Unit 4. Together they are 50 of your 100 marks. The other 50 come from the external exam, two papers of 90 minutes each.

How is the QCAA Physics data test marked?

The data test has 5 minutes perusal and 60 minutes working time, uses two to four unseen datasets from Unit 3, and allows a calculator and the QCAA Physics formula and data book. Your raw score becomes a percentage, and fixed cut-offs turn it into a mark out of 10: more than 90% earns 10, more than 80% earns 9, and so on. Exactly 80% earns 8.

Which experiment do you modify for the Physics student experiment?

You modify a practical or simulation performed in class that is relevant to Unit 3, gravity and motion or electromagnetism. The syllabus lists investigations such as projectile range at various angles, forces on an inclined plane, horizontal circular motion on a string, the force on a conductor in a magnetic field, the strength of a magnet at various distances and electromagnetic induction.

How do you calculate uncertainty in a Physics IA2?

The syllabus uses half the range of your trials for the absolute uncertainty of the mean, and maximum and minimum trend lines for the uncertainty of a gradient or intercept. QCAA's sample marking scheme takes half the difference between their gradients. When you add or subtract, add absolute uncertainties. When you multiply, divide or raise to a power, add percentage uncertainties.

What claims can you use for the Physics research investigation?

Your teacher provides a list of claims relevant to Unit 4: special relativity, quantum theory or the Standard Model. QCAA's sample instrument includes claims such as "neutrinos are massless particles" and "all experimental evidence indicates that light is a particle". You then pose your own research question about one aspect of the claim.

What were the grade cut-offs for QCAA Physics in 2025?

For the 2025 Physics cohort, the final result out of 100 needed 87 for an A, 72 for a B, 50 for a C and 20 for a D. Of 8,302 students who completed Units 3 and 4, 49.12% received an A. These boundaries apply only to the final result and can move each year.

Back to all articles

Read next

A square quadrat frame lying on pale ground among small plants, one of them coral, beside three cards: a column graph with error bars and a coral bar, a Punnett square grid with one coral cell, and a double helix with one coral strand

QCAA Biology IAs Explained: IA1, IA2 and IA3 (2025 Syllabus)

The data test, the student experiment and the research investigation, worked through for Biology: what each is worth, how field sampling and standard error earn marks, and what QCAA's latest reports say costs them.

Jackson Wright6 November 2026
An ancient coin, a fragment of a carved inscription and an open scroll on a warm off-white surface, with a coral thread pinned from the inscription through the coin to an open essay page with one coral underline

QCAA Ancient History IAs Explained (2025 Syllabus)

A source essay exam, an independent source investigation and a research essay: 75 of your 100 marks. What each Ancient History ISMG rewards, why ancient is not the same as primary, and where QCAA marked schools down.

Jackson Wright6 November 2026