IB Physics Internal Assessment: Mistakes That Cost Marks

The IB Physics internal assessment is one individual scientific investigation, written in at most 3000 words, marked out of 24 by your teacher and then moderated by the IB. It is worth 20% of the final Physics grade at both Higher Level and Standard Level, and around 10 hours of class time are set aside for it.

The four criteria carry equal weight.

Research design, Data analysis, Conclusion and Evaluation are worth 6 marks each. That symmetry is the single fact most students ignore. They spend three weeks in the laboratory and two evenings on the write up, then wonder why an elegant experiment scored 15 out of 24. Half the marks, the 12 in Conclusion and Evaluation, are awarded for what you write once the data already exists.

The IA is not the collaborative sciences project.

The collaborative sciences project is a separate activity of roughly 10 hours, shared with Biology and Chemistry students, and it contributes no marks to your Physics grade. Keep the two apart in your planning calendar.

Four legitimate kinds of investigation.

Hands on laboratory work, analysis of an existing database, a computer simulation, or a hybrid of these. A database investigation is not the soft option some students hope for: the same four criteria apply, so you still need a controlled comparison, a defensible uncertainty on values you did not measure yourself, and real processing rather than a downloaded graph pasted into the report.

The task is identical at HL and SL. Only the syllabus content you can draw on differs, which is one reason the difference between HL and SL Physics matters when you choose a topic. An SL student who builds an investigation on HL only theory has nothing solid to argue from in the Conclusion.

An IB Physics IA research question loses marks when it names no measurable independent variable, no numerical range and no controlled variables. Research design is worth 6 of the 24 marks, and a vague question then damages Data analysis and Conclusion too, because there is nothing precise left to analyse or to conclude about.

Mistake 1: the question has no numbers in it.

How does temperature affect resistance is a topic, not a research question. The version that scores is: how does the resistance of a 30.0 cm nichrome wire of diameter 0.27 mm vary with temperature between 20 °C and 80 °C, measured every 10 °C. A reader now knows the independent variable, the range, the interval and the fixed geometry.

Mistake 2: the controlled variables are listed but not controlled.

Writing room temperature was kept constant earns nothing. Saying how you kept it constant, and how you checked, earns the mark. If a variable cannot be controlled, monitor it and record it, then use that record in the Evaluation.

Mistake 3: choosing a relationship that is trivially linear.

If your prediction is a straight line through the origin and the data confirm it in one graph, there is little for you to do in Data analysis. Relationships with a power, a square root or an exponential decay give you a linearisation to perform, and linearisation is where analysis marks live. Good candidates include the pendulum, $T = 2\pi\sqrt{\frac{L}{g}}$, the discharge of a capacitor, $V = V_0 e^{-t/RC}$, and the inverse square fall off of intensity, $I \propto \frac{1}{d^2}$.

Mistake 4: no pilot run.

Twenty minutes of trial measurements tells you whether your range produces a visible change, whether your timer resolution is adequate, and whether five values will be enough. Students who skip the pilot discover on data collection day that their whole range fits inside their uncertainty.

Mistake 5: an unsafe or unfeasible topic.

The IA has to pass a safety, ethics and environment check. A question that needs equipment your school does not own is not ambitious, it is simply undoable, and the fallback experiment improvised in the last week is what ends up being submitted.

The most common IB Physics IA data analysis mistake is treating uncertainties as decoration: quoting the precision of an instrument, then never propagating it into the final result. Data analysis is worth 6 marks, and an uncertainty that never reaches the answer limits that criterion however neat the graph looks.

Justify the raw uncertainty.

Half the smallest division is a reasonable starting point for a ruler, but not for a stopwatch operated by a human hand. If your reaction time contributes about 0.2 s, say so and use it. An honest large uncertainty scores better than an unjustified small one.

Propagate with the standard rules.

Then quote the final uncertainty to one significant figure and match the number of decimal places in the value to it.

A fully worked example: measuring $g$ with a pendulum.

The period of a simple pendulum is $$T = 2\pi\sqrt{\frac{L}{g}}$$ Squaring gives $T^2 = \frac{4\pi^2}{g}L$, so a graph of $T^2$ against $L$ should be a straight line through the origin with gradient $m = \frac{4\pi^2}{g}$.

• Step 1. Time 20 oscillations rather than one. If the timing uncertainty is 0.2 s, then the uncertainty in one period is $\frac{0.2}{20} = 0.01$ s, a hundredfold improvement for no extra equipment.
• Step 2. Convert to the plotted variable. If $T = 1.42 \pm 0.01$ s, the percentage uncertainty is 0.7%, so $T^2$ carries $2 \times 0.7\% = 1.4\%$, that is $T^2 = 2.02 \pm 0.03$ s$^2$.
• Step 3. Draw error bars on every point and fit the steepest and the shallowest lines that still pass through all of them. Suppose the best fit gradient is $m = 4.05$ s$^2$ m$^{-1}$, the maximum gradient is 4.21 and the minimum is 3.89.
• Step 4. Uncertainty in the gradient: $\Delta m = \frac{4.21 - 3.89}{2} = 0.16$ s$^2$ m$^{-1}$, which is $\frac{0.16}{4.05} \times 100 = 4\%$.
• Step 5. Rearrange: $g = \frac{4\pi^2}{m} = \frac{39.5}{4.05} = 9.75$ m s$^{-2}$. Since $g$ depends on $m^{-1}$, the percentage uncertainty stays 4%, so $\Delta g = 0.04 \times 9.75 = 0.39$. Report $g = 9.8 \pm 0.4$ m s$^{-2}$.

Three smaller traps in the same criterion.

Putting units and uncertainty in every cell instead of once in the column header. Quoting a coefficient of determination such as $R^2 = 0.998$ as if it proved accuracy, when it only describes scatter about the fitted line. And reporting a mean to six decimal places because the spreadsheet did, when the raw data carry three significant figures.

IB Physics IA conclusions score half marks when they describe the graph instead of interpreting it, and when they never compare the measured value with a cited accepted value. The Conclusion criterion is worth 6 marks and asks for a result that is justified by your own data and then placed in a wider scientific context.

Mistake 1: describing rather than concluding.

As the length increased, the period squared increased is a caption, not a conclusion. The conclusion is the number your gradient produced, with its uncertainty, and the physical statement it supports: the data are consistent with $T^2$ being proportional to $L$, giving $g = 9.8 \pm 0.4$ m s$^{-2}$.

Mistake 2: no accepted value, or an uncited one.

Compare your result with a published value and cite where it came from. In the pendulum example, the accepted value $g = 9.81$ m s$^{-2}$ sits inside the range 9.4 to 10.2 that your uncertainty defines, so the correct sentence is that the result agrees with the accepted value within experimental uncertainty. The percentage difference, $\frac{|9.75 - 9.81|}{9.81} \times 100 = 0.6\%$, supports that statement but does not replace it.

Mistake 3: confusing agreement with accuracy.

If your uncertainty is 40%, almost anything agrees with anything. Say so. A candid line noting that the uncertainty range is too wide to distinguish competing models is worth more than a triumphant claim of confirmation.

Mistake 4: claiming proportionality when the intercept is not zero.

A relationship is proportional only if the line passes through the origin within uncertainty. If your intercept is $-0.05 \pm 0.01$ s$^2$, that is a systematic effect, and it belongs in the conclusion as a finding, not hidden.

Mistake 5: extrapolating beyond the range investigated.

You measured between 20 °C and 80 °C. You cannot conclude anything about 200 °C. Examiners notice the overreach immediately, and it undermines the parts of the conclusion that were sound.

The same discipline of answering the question you actually set is what makes the difference across the whole Diploma, from the IA to the Extended Essay.

You avoid the Evaluation trap by giving every weakness a direction of effect, an estimated size and a realistic fix. Evaluation is worth 6 of the 24 marks in the IB Physics IA. Phrases such as human error, air resistance and use more accurate equipment score nothing on their own because they carry no magnitude and no consequence.

Use a sentence template that forces the detail.

Because the pendulum bob was released by hand, the first oscillation was slightly larger than the rest, which makes the measured period systematically too long by roughly 0.5%, pushing my value of $g$ below the accepted value. Releasing the bob with an electromagnet would remove the extra initial displacement. That single sentence contains a cause, a direction, a size and a fix.

Separate random from systematic.

Random effects show up as scatter and as the spread between repeats, and more repeats reduce them. Systematic effects shift every point the same way and show up as a non zero intercept or a wrong gradient. Repeating the experiment more times does nothing to a systematic effect, which is why more repeats is such a weak improvement to propose.

Rank the weaknesses.

An examiner wants to see judgement, not a list of eight equally weighted apologies. If the timing uncertainty is 0.7% and the length measurement is 2%, the length is where your error budget actually lives, so discuss it first and give it the most space.

Propose improvements that a school can afford.

A light gate replacing a hand held stopwatch is realistic. A vacuum chamber is not. An improvement that is impossible to implement reads as a way of avoiding the problem rather than solving it.

Evaluate the method, not yourself.

I should have been more careful is not an evaluation point. The apparatus, the procedure, the range and the assumptions are what you are being asked to criticise. Building this habit early also pays off in written papers, where the same explain and evaluate command words appear, as covered in the IB exam preparation strategy.

Plan the IB Physics IA backwards from your school's final deadline, with the pilot run and the data collection finished well before it, because only one draft may be given written feedback by your teacher. The investigation is capped at 3000 words and around 10 hours of class time, and neither figure stretches under pressure.

The one draft rule is the tightest constraint.

Your teacher may comment on a single complete draft in writing, and may not correct or annotate the work itself. A draft submitted in fragments wastes that one chance. Submit something complete, even if imperfect, so the feedback covers all four criteria.

A realistic sequence.

Choose the topic and run the pilot in the second half of the first year, when the timetable is still light. Collect data early in the second year. Write the analysis within a week of collecting, while you still remember what the apparatus was doing. Leave the Conclusion and Evaluation for a period when you can think, not for the night before.

Manage the word count from the start.

At 3000 words maximum, the write up is short. Raw data tables, graphs, calculations and captions are where the evidence lives, and the prose should be doing argument rather than narration. Cut every sentence that recounts what you did minute by minute; the method needs to be reproducible, not autobiographical.

Keep a laboratory notebook with dates.

Records of your pilot, your discarded attempts and your monitored variables become Evaluation material later. Students who rely on memory end up inventing plausible weaknesses instead of reporting real ones.

Respect academic integrity.

Cite the source of any accepted value, any database, any simulation and any image that is not yours. The IA is submitted with a declaration of authenticity, and an uncited value in a Conclusion is exactly the kind of detail that turns a marking issue into a procedural one. If you want the wider picture of how internal assessment works across the Diploma subjects, the IB internal assessment overview sets out the common structure.