IB Physics IA: ideas, criteria and how to score well
The IB Physics internal assessment is your own scientific investigation, worth 20% of your grade. A good idea is one you can measure well, analyse properly and evaluate honestly. Here is how it is marked, plus ideas that work.
On this page
How the IA works
From the 2025 syllabus, the internal assessment is one task: the scientific investigation. It is worth 20% of your final grade at both SL and HL, and the IB plans about 10 hours of class time for it. Your teacher marks it and the IB moderates a sample from your school. It is marked out of 24, with four criteria worth 6 marks each.
Check your school's deadlines and the current IB guidance on length and format with your teacher, because the IB can update its instructions.
The four criteria
- Research design (6 marks): a focused research question in context, and a method that can actually answer it, with controlled variables and enough data.
- Data analysis (6 marks): clear, correct processing of your raw data, with uncertainties handled properly and appropriate graphs.
- Conclusion (6 marks): a conclusion that answers your research question, is justified by your analysis, and is compared with accepted physics.
- Evaluation (6 marks): the strengths and weaknesses of your method, how they affected your results, and realistic improvements.
Notice what is missing: there is no separate mark for “personal engagement” or for presentation. Those still matter, but they earn marks through clear design, analysis and evaluation.
What makes a good IA idea
- One independent variable you can change over a wide range, with at least five values and repeats.
- A dependent variable you can measure precisely with school equipment, a phone sensor or video analysis.
- A link to theory, so that your graph can be compared with a prediction, ideally by linearising it.
- Enough uncertainty to discuss, but not so much that the trend disappears.
- Something you are curious about. Investigations built on a real question are easier to write well.
IA ideas by theme
Use these as starting points and make them your own by changing the context or the variable.
- Mechanics (Theme A): how the drop height affects the bounce height of a ball; how the angle of a ramp affects the acceleration of a trolley; how the mass on a spring affects its period; how the length of a pendulum affects its period at large angles.
- Thermal physics (Theme B): how the surface area of water affects its rate of cooling; how insulation thickness affects heat loss from a beaker; how the temperature of a thermistor affects its resistance.
- Circuits (Theme B): how the length of a wire affects its resistance, to find resistivity; how the current drawn affects the terminal voltage of a cell, to find internal resistance.
- Waves (Theme C): how the tension in a string affects the frequency of its fundamental; how the length of an air column affects its resonant frequency; how the concentration of a sugar solution affects its refractive index.
- Fields (Theme D): how the distance from a magnet affects the magnetic field strength measured with a phone magnetometer; how the number of turns on a coil affects the induced e.m.f. (HL).
- Nuclear and light (Theme E): how the distance from a light source affects the measured intensity, testing the inverse-square law; how absorber thickness affects the count rate from a safe source, if your school allows it.
Turning an idea into a research question
Idea: pendulums. Weak question: “How does a pendulum work?”
- Pick one independent variable: the length of the string, from 0.20 m to 1.20 m in six steps.
- Pick one dependent variable: the period, timed over 20 swings and repeated three times.
- Link to theory: T = 2π√(L/g), so a graph of T² against L should be a straight line through the origin with gradient 4π²/g.
- Strong question: “How does the length of a simple pendulum affect its period, and what value of g does this give?”
Handling uncertainty in the analysis
You time 20 swings as 28.4 s with a reaction-time uncertainty of ±0.2 s. What is the period and its uncertainty?
- Period = 28.4 / 20 = 1.42 s.
- Absolute uncertainty = 0.2 / 20 = 0.01 s, so T = 1.42 ± 0.01 s.
- Timing many swings divides the uncertainty: a point worth making in your evaluation.
Writing it up
- Introduction and context: why the question matters to you and the physics behind it.
- Method: variables, apparatus, procedure, safety and how you reduced uncertainty.
- Raw data: a clear table with units and uncertainties.
- Processing and graphs: one worked calculation, a linearised graph with error bars, and maximum and minimum gradient lines.
- Conclusion: the answer to your question, your value compared with the accepted one, and whether the difference is within your uncertainty.
- Evaluation: specific weaknesses, their effect on your result, and realistic improvements.
Common mistakes
- Choosing a question with too narrow a range of the independent variable to show a trend.
- Collecting too few data points, or no repeats.
- Listing generic errors such as “human error” instead of specific ones.
- Copying a textbook experiment without any personal angle or extension.
- Getting a teacher or tutor to write sections: the IA must be your own work.
Practice questions
Try each one before opening the answer.
Q1. Your graph of T² against L has gradient 4.05 s² m−1. What value of g does it give?
Show answer
Q2. Name the four IA criteria.
Show answer
Q3. Why plot T² against L instead of T against L?
Show answer
Frequently asked questions
How much is the IB Physics IA worth?
20% of your final grade at both SL and HL, marked out of 24.
Can I use a simulation or database instead of an experiment?
The IB allows investigations that use secondary data or simulations in some cases, but check with your teacher first, since the research design must still be strong.
Can a tutor help with my IA?
A tutor can explain the physics, the criteria and how to handle uncertainties, and give general feedback, but the investigation and the writing must be your own, as the IB requires.
Key takeaways
- One scientific investigation, 20% of the grade.
- Marked out of 24: four criteria of 6 marks.
- Choose one variable you can change widely and measure precisely.
- Linearise to compare with theory.
- Evaluate specific weaknesses, not generic errors.