IB Physics SL Internal Assessment Example

The research question is too broad for the actual procedure: it mentions “various angles,” but the study uses only 10° and 30° while changing length at the same time. The student should rewrite the question so the independent variables and expected comparison are stated precisely.

The student situates the investigation in a clear physics context and links the question to SHM and pendulum motion. This works well because it immediately tells the reader what relationship is being tested and why the topic matters scientifically.

The hypothesis would be stronger if it predicted the expected pattern in scientific terms, not just “greater slope.” The student should state what should happen to $T$ as $L$ increases and how the 10° and 30° conditions are expected to differ according to the pendulum model.

The control-variable table is a strong feature because it shows the student has thought about how to reduce confounding effects. That supports a more valid design by making the major variables and their control methods visible to the reader.

This release method introduces a reproducibility problem because “simultaneously” starting the software and releasing the bob can vary from trial to trial. The student should describe a release procedure that is time-consistent and explain how timing begins at a clearly defined physical event.

Five trials is a sensible minimum, but the design would be stronger if the student justified why five is enough to capture variation and whether additional repeats were needed for noisy conditions. Explaining that choice would show deeper methodological awareness.

What effect might changing both the clamp height and the amount of string wound around the rod have on the true pivot-to-bob length? Thinking about that could help the student check whether the length variable is being controlled as intended.

The student opens the analysis with a clear statement that uncertainty matters. This is a strong analytical habit because it shows the numbers are not being treated as exact and prepares the reader for later error discussion.

The outlier removal is not justified with a criterion, so it risks appearing arbitrary. The student should state a rule for identifying outliers and explain why this point was excluded, especially because removing one value can shift the mean and uncertainty.

The student is transparent about excluding an outlier, which is better than silently removing it. This helps the reader see that the data set was not selectively hidden and makes the processing easier to follow.

The stated horizontal uncertainty is inconsistent with the earlier measuring-stick uncertainty of $\pm 0.00005\text{m}$. The student should reconcile the units and explain how this uncertainty was calculated, otherwise the processing loses precision and credibility.

Several processed entries show the same length uncertainty value without a clear calculation, even though the raw uncertainty statement differs elsewhere. The student should show the propagation or explain why the same uncertainty is reused for every row so the processing can be checked.

The student could make the processing route more convincing by explicitly linking each graph to a purpose: one for raw trend, one for uncertainty bounds, and one for model comparison. That would help the reader understand why each transformation is needed.

The graph is being treated as linear in $L$, but the pendulum model is not linear in $L$; it is proportional to $\sqrt{L}$. The student should not use a straight-line fit on $T$ vs $L$ to derive $g$ from the textbook equation.

If the model is $T=2\pi \sqrt{\frac{L}{g}}$, what variable transformation would make the comparison with the data mathematically appropriate? Reflecting on that would help the student choose a processing method that matches the theory.

This interpretation depends on the invalid linearisation used earlier, so the derived $g$ values cannot be trusted as evidence of “slight error.” The student should reprocess the data with a model that matches the pendulum equation before drawing conclusions from $g$.

The student makes a clear connection between the observed result and the relevant physics idea. Linking the result to the small-angle approximation and restoring-force model is exactly the kind of scientific reasoning expected in a conclusion.

The conclusion overreaches here: only $10^{\circ }$ and $30^{\circ }$ were tested, so the investigation does not establish what happens for angles smaller than $10^{\circ }$. The student should limit the claim to the angles actually measured.

This claim is not fully supported because the $g$ range comes from the flawed processing method used earlier. The student should only state that the result supports the model after the data have been processed using a valid comparison to the pendulum equation.

The conclusion stays focused on the investigation’s main trend and refers back to the graph, which is good practice. This keeps the summary anchored in the student’s own data rather than drifting into unsupported general statements.

The percentage deviation is not convincingly justified because it depends on the earlier invalid $g$ calculation. The student should compare the 30° data directly with the accepted pendulum context and use the actual experimental trend, not a flawed derived value, to support the claim.

The conclusion would be stronger if it briefly stated whether the observed increase in period with angle was expected from the finite-amplitude pendulum model, not just from the small-angle approximation. That would give the comparison with accepted scientific context more depth.

The student correctly connects a methodological issue to a specific data-processing consequence. That is valuable because it shows the weakness affected the final averages, not just the procedure in theory.

The student identifies a concrete weakness rather than a vague “error,” which strengthens the evaluation. Naming the software delay helps explain why the timing data may be systematically shifted and shows awareness of how the method affected the results.

This limitation is described, but its effect on the data is not fully explained. The student should state whether protractor misalignment would mainly shift the measured angle, increase scatter, or bias one condition more than the other.

This is a useful evaluation because it links the control-variable strategy to result quality. The student shows understanding that uncontrolled angle or length changes would affect the dependent variable, which is important for judging validity.

The improvement would be stronger if the student named what kind of timing method would replace the current one, such as measuring between equivalent photogate events rather than starting a timer manually. Specificity would make the recommendation more actionable.

The weakness is identified well, but the evaluation does not quantify how large the delay was or how much it could change the measured period. The student should explain the likely impact on the results so the limitation is evaluated, not just stated.

This limitation is important, but the student does not distinguish clearly between a systematic offset and random outliers. The evaluation should separate these effects and explain why each one matters differently for reliability and validity.

If the aim is to locate the failure point of the small-angle model, how many angles would be needed to make that threshold convincing? Considering sample density could help the student turn a general extension into a more rigorous one.

The student correctly identifies angle variation, but the impact is only described qualitatively. The evaluation should explain whether this mainly increases random scatter or creates a systematic bias, and why that distinction matters.

This is a sensible improvement because it directly addresses reading uncertainty in length. Repeating measurements is a realistic way to improve precision without changing the whole experiment.