IB Biology HLStage 3 — Conclude & EvaluateIA & Paper 2Practical skill~13 min read
Evaluating the Method
The evaluation is where you turn on your own work and take it apart. It sounds uncomfortable, but it is the section that separates a student who followed a method from one who understands why methods matter — and it is usually the easiest place left to pick up marks.
📚 What you need to know
An evaluation is a critical reflection on your methodology — the weaknesses and limitations of your own work.
Start by evaluating the hypothesis: even if the data supported it, judge how strong that support really was.
Identify specific sources of error, split into systematic and random.
Systematic errors are flaws in the method or apparatus that make results consistently wrong in the same direction.
Random errors are unpredictable variations, in biology usually caused by natural variability between samples. Repeats and means reduce them.
Also discuss weaknesses, limitations and assumptions — three different things.
Every weakness needs a specific, realistic improvement that a school lab could actually carry out.
Close the loop each time: weakness → impact → improvement.
Start with the hypothesis
The evaluation follows straight on from the conclusion, so begin with the thing you just decided. Even when the data supported your hypothesis, ask how strongly it did so once the uncertainties are taken into account.
What that sounds like. “The data supported the hypothesis, but the standard deviation at 50 °C was noticeably larger than elsewhere, suggesting the response was becoming unpredictable there. The exact temperature at which membrane damage begins is therefore less certain than the graph makes it look.”
Weakness, impact, improvement
This is the shape of every point in an evaluation. Miss out the middle box and it reads as a list of complaints; miss out the last one and you have not shown you know how to fix anything.
Notice how specific the improvement is. “Be more careful when cutting” would not be an improvement, because nobody could follow it.
The two kinds of error
You met these in Stage 1 when you were controlling variables. Now you have data, you can say what they actually did to it.
Systematic error
Random error
What it is
A flaw in the method or apparatus that makes every result wrong in the same direction — always too high, or always too low
Unpredictable variation that happens by chance, in different directions each time
Typical cause in biology
An uncalibrated instrument, or a step in the procedure that biases every reading the same way
Natural variability between samples, and judgements made by eye
Example
A pH meter reading 0.5 units high shifts the whole graph, so the optimum pH you report is wrong
Two beetroot discs from the same root have slightly different cell structures, so they leak differently
Effect on the data
Reduces accuracy. The trend may still look fine while every value is shifted
Reduces precision. Increases the standard deviation and makes nearby conditions hard to separate
How to reduce it
Calibrate the instrument, or redesign the biased step
Take more repeat trials and use the mean
🧠
Which one am I looking at?
Ask what more repeats would have done. If they would have helped, it was random. If every reading would still have been shifted the same way, it was systematic — and only a change to the method or the calibration fixes it.
Weaknesses, limitations and assumptions
These three get used interchangeably in conversation. In an evaluation they are three separate things, and naming them correctly shows you know the difference.
Limitations are worth writing even when nothing went wrong. They show you understand exactly how far your own result reaches.
Category
Beetroot example
What to write
Weakness
Discs were rinsed by hand for varying lengths of time before going into the water baths
Name the step, say it added error, and give a fixed protocol instead
Limitation
Only one variety of beetroot was used, at one exposure time of 20 minutes
State that the conclusion cannot be generalised beyond those conditions
Assumption
It was assumed the absorbance came only from pigment, and that all discs started with equal pigment content
Name the simplification and say what it could have hidden
Never write “human error”
It is the most common phrase in student evaluations and it earns nothing, because it names no step, explains no impact and points at no fix.
The same rule applies to “I could have been more accurate” and “there wasn’t enough time”. Neither points at anything a reader could act on.
Improvements that count
Relevant. It must fix the weakness you just described, not a different one.
Realistic. You should be able to do it in a normal school laboratory. Suggesting an electron microscope is not an improvement, it is a wish.
Specific. Give the actual protocol: the piece of apparatus, the fixed time, the number of replicates.
Prioritised. Deal properly with the one or two errors that had the biggest effect on your final result, rather than listing eight in one line each.
A test for whether an evaluation is really yours: could it be copied and pasted into someone else’s report on a completely different experiment? If yes, it is generic and will score poorly. Refer to your own graph, your own standard deviations, your own observations.
WE 1
A systematic error, closed out fully
In the beetroot investigation, discs were placed straight into the water baths after cutting, without rinsing. Evaluate this as a source of error. (4 marks)
Weakness
Cutting the discs damages cells at the cut surface, and that pigment was carried into the tube along with the disc.
Why it is systematic
It happened to every disc, in the same direction — each tube gained extra pigment that had nothing to do with the temperature treatment.
Impact
Every absorbance reading is inflated, so the whole curve is shifted upwards. The trend still appears, but the values are inaccurate, and the effect matters most at 20 to 30 °C where the real absorbance is smallest.
Improvement
Rinse each disc in distilled water for a fixed 30 seconds, blot it in a standard way, then transfer it to the water bath.
Systematic: every value too high, and no number of repeats would fix itsaying which part of the graph it affects most is what lifts this from a description to an evaluation
WE 2
A random error, closed out fully
The standard deviation at 60 °C was 0.03, larger than at any other temperature. Evaluate the likely cause and suggest an improvement. (3 marks)
Weakness
Beetroot tissue is naturally variable. Discs differ slightly in cell structure, pigment content and how much damage the borer caused.
Impact
This scatters the replicates around the mean, raising the standard deviation and lengthening the error bar at 60 °C, which makes it harder to say whether 60 and 70 °C really differ.
Improvement
Cut all discs from the same beetroot, from the same region of the root, and increase replicates from three to five at each temperature.
Random: repeats and a mean genuinely help herenatural variability is the classic random error in Biology — examiners expect to see it named
WE 3
A limitation and an assumption
State one limitation and one assumption of the beetroot investigation, and explain the effect of each on the conclusion. (4 marks)
Limitation
Only one variety of beetroot was tested, at a single exposure time of 20 minutes.
effect: the conclusion cannot be generalised to other plant species, or to longer exposures, where the membrane might behave differentlyHow to address it
Repeat the investigation with a second species, and at several exposure times, to see whether the same pattern holds.
Assumption
It was assumed that the absorbance came only from pigment leaking through the membrane.
effect: if other coloured substances entered the water, or discs started with different pigment contents, the absorbance overstates membrane damageA limitation narrows the claim; an assumption may quietly bias ita limitation is not a mistake — it is the honest boundary of what you tested
💡 Exam tips
Close the loop every time: weakness, impact, improvement. Three sentences minimum.
Say whether each error was systematic or random, and justify which.
Quote your own numbers — “the standard deviation at 60 °C was 0.03” beats “the results varied”.
Deal with one or two big errors properly rather than six in passing.
Discussing the natural variability of your biological samples is nearly always a scoring point.
Make improvements concrete enough that someone could follow them tomorrow.
⚠ Common mistakes
Blaming “human error”. It names nothing and fixes nothing.
Listing weaknesses with no impact, so the reader cannot tell whether they mattered.
Suggesting improvements no school could do, or vague ones like “be more careful”.
Confusing the two error types, and suggesting more repeats as the fix for a systematic error.
Writing a generic evaluation that would fit any experiment in the syllabus.
Ignoring limitations because nothing “went wrong”. They belong there regardless.
That completes The Scientific Inquiry Cycle. Across the three stages you asked a question worth asking, built a method that could be repeated, collected and processed the data honestly, drew a conclusion your evidence could carry, and then said out loud where it was weak. That last part is not an admission of failure — it is the whole reason anyone trusts scientific work at all.
Want this explained one-to-one?
Book a free session with an experienced IB Biology tutor and get your trickiest topics made simple.
