The Physics HL IA uses the same scientific investigation criteria as SL, but the analysis should normally show deeper theory, stronger modelling and more careful uncertainty treatment.
Focused question, variables, controls, reproducible method, safety/environmental notes and a setup that produces quantitative data.
Raw and processed data, uncertainty, sample calculations, model fitting, gradients and appropriate significant figures.
Direct answer to the question, comparison with theory and interpretation supported by the graph and uncertainty.
Specific weaknesses, their effect on results, realistic improvements, strengths and possible extensions.
A strong Physics HL IA should still be measurable, but the model can be richer. HL is a good place for exponential decay, log-linear analysis, electromagnetic induction, drag laws, proportionality testing and uncertainty in fitted constants.
The attached guide focuses on HL ideas with clean data and a real design endpoint. Capacitor discharge can become a timing circuit. Falling magnets can test Faraday's law. Coffee filters can reveal a drag exponent. Each topic is observable, but each also gives room for stronger analysis than a basic SL verification.
HL investigations become impressive when the graph transformation is meaningful. If a relationship is exponential, use a log-linear graph and interpret the gradient. If a power law is suspected, take logarithms and find the exponent. If induction is involved, discuss flux change rather than only voltage peaks. The conclusion should go beyond 'the equation fits' and calculate a designed resistor, proportionality exponent, threshold speed or model limitation.
This turns a standard exponential curve into a design problem: choose the resistor that makes a circuit wait exactly the time you want.
Research question: How does resistance affect the discharge time of a capacitor, and how can this be used to design a timing circuit?
Build a simple RC circuit with a capacitor, resistor, switch and voltmeter or data logger. Charge the capacitor to a known starting voltage, then record voltage over time for several resistor values. Fit the exponential discharge model and linearise the graph using natural logarithms. The IA becomes more than a verification when you calculate the resistor needed for the voltage to fall to a chosen threshold after a target time. This links to camera flashes, alarms, sensors, smoothing circuits and electronic delays.
IB topics used: electric circuits, capacitors, exponential decay, linearisation, uncertainty
Data to collect: Voltage-time data for several resistor values, compared with theoretical and experimental time constants.
Diagram to include: RC circuit diagram, voltage-time decay curve and log-linear graph.
The magnet falls for less than a second, but the voltage pulse contains a full story about speed, flux change and induction.
Research question: How does the speed of a magnet passing through a coil affect the peak induced voltage?
Connect a coil to a voltage sensor, oscilloscope or data logger, then drop the same magnet through the coil from different heights. Estimate entry speed from drop height or video analysis. Record the voltage-time pulse and extract the peak induced voltage for each drop. Test whether peak emf is proportional to speed, or fit a power model to estimate the exponent. A strong evaluation discusses alignment, magnet orientation, coil geometry, sensor response time and why pulse area can be a useful extension.
IB topics used: electromagnetic induction, Faraday's law, magnetic flux, power laws, uncertainty
Data to collect: Voltage-time pulses from repeated drops at different heights, with peak emf plotted against estimated speed.
Diagram to include: Magnet falling through coil, voltage pulse shape and peak voltage versus speed graph.
This is a deceptively simple parachute lab: stack filters, film the fall, then let the log graph reveal the drag exponent.
Research question: How does mass affect the terminal velocity of falling coffee filters, and what drag model best fits the data?
Stack different numbers of identical coffee filters, drop them from the same height in still air and use video analysis to find terminal velocity. At terminal velocity, weight equals drag, so the data can test a power-law drag model. Taking logarithms lets you find the exponent from the gradient of a log-log graph. The strongest version confirms terminal velocity rather than using average speed over the whole fall, and evaluates whether stacking filters really keeps area and shape constant.
IB topics used: forces, terminal velocity, drag, log-linear analysis, power models
Data to collect: Terminal velocity for different filter stacks, with repeated drops and a log-log model fit.
Diagram to include: Falling filter stack with forces, then log(mg) against log(v) to find the drag exponent.
Choose the investigation where you can collect the cleanest data, not the one with the most dramatic title. A high-scoring IA usually has one clear independent variable, one meaningful dependent variable, enough repeats, a sensible model and an evaluation that admits what the data cannot prove. Before starting, confirm your exact subject guide, safety rules and teacher expectations.
