IB Biology HL Internal Assessment Example

The research question is strongly framed in a specific biological context, naming the yeast species, sweeteners, concentration, temperature, duration, and CO₂ measurement. This makes the investigation focused and clearly tied to a measurable outcome.

The raw data table is well structured, with initial and final CO₂ readings recorded for each of the five trials per sweetener. This makes the data trail transparent and lets the reader see how the processed results were derived.

The crossed-out values are identified as outliers, but no rule for removing them is given. The student should state the criterion used for exclusion so the processing remains transparent and not selective.

What objective criterion justified removing each crossed-out value, and would the conclusion change if those points were retained? This would help the student evaluate the robustness of the dataset.

The standard deviation example is not fully transparent because the individual deviations are shown only as summed values, with no clear link back to the raw trial numbers. The student should show the substituted values step by step so the calculation can be checked.

The worked rate calculation produces $374.7{\text{ppm s}}^{-1}$ from a 10-minute interval, which is dimensionally inconsistent. The student should either report the result as ppm per minute or convert 10 minutes into seconds before calculating a per-second rate.

The rate formula is not communicated precisely because the unit is written as ${\text{ppm s}}^{-1}$ even though the calculation divides by 10 minutes. The student should align the unit with the time base used, or convert the denominator to seconds consistently.

The processing steps are communicated clearly through the worked calculations for change in CO₂ and mean value. Showing the formula and one full example helps the reader follow the logic of the analysis.

The student includes a standard deviation formula and uses it to describe spread in the repeated trials. This is a valuable step because it goes beyond averages and begins to communicate variability in the dataset.

If the CO₂ sensor contributes a ±10% uncertainty, how does that affect the confidence in the ranking between sweeteners? Thinking about this would strengthen the interpretation of the processed data.

The uncertainty example shows a large mismatch between the stated combined uncertainty and the final reported value, which suggests an arithmetic or transcription problem in the propagation. The student should recalculate the uncertainty chain and ensure the same method is used in the example and the summary table.

The report mentions a one-tailed t-test but does not justify why a one-tailed test is appropriate here. The student should explain the predicted direction of each comparison and the assumptions behind the test to make the statistical approach defensible.

The summary table still contains inconsistent units and notation, which weakens precision. In particular, the rate column should match the definition used in the calculations, and the mean CO₂ values should be labelled consistently across the report.

The student correctly attempts to quantify measurement uncertainty using the apparatus uncertainties rather than ignoring them. This is a meaningful improvement because it shows awareness that the processed rates are approximate, not exact.

The uncertainty table appears incomplete, with several blanks and unclear entries. Because uncertainty propagation is central to the analysis, every row should show the measured uncertainty, percentage contribution, and final combined uncertainty clearly and consistently.

The t-test results are presented in a way that allows pairwise comparisons across sweeteners, which is useful for testing whether observed differences are statistically significant. This moves the analysis beyond descriptive trends alone.

The “R values” table is not clearly defined, so the reader cannot tell what each number represents or how it was derived. This ambiguity reduces the clarity of the analysis and should be replaced with a labelled table that uses standard statistical terminology.