IB Chemistry SLTopic 8 — Concluding and EvaluatingInternal assessmentPractical skill~12 min read
Drawing Conclusions
A conclusion is short. That is the part students find hardest to believe, so they pad it out with everything they wish had gone better. All it has to do is answer the question you asked at the start, using the numbers you have in front of you.
📚 What you need to know
Open with a direct answer to your research question. One sentence.
Back it with your processed data — the final value, the gradient, the enthalpy change.
Always quote the result with its uncertainty.
Say plainly whether the results support or refute your hypothesis.
Compare with a literature value, cite where you got it, and work out the percentage error.
Then compare that percentage error with your percentage uncertainty. This is the sentence that scores.
No new ideas here. Improvements and excuses belong in the evaluation.
Answer the question you actually asked
Go back and read your research question, word for word. Your first sentence should read like a reply to it. If your question named a range, a chemical and a measurement, your conclusion names the same things.
Weak conclusions drift into general chemistry: “this shows that concentration affects the rate of reaction.” True, and known since before you were born. A conclusion is about your experiment and your numbers.
Layer four is the one most students leave out, and it is the one that lets you say anything at all about accuracy.
Say what happened to your hypothesis
You made a prediction at the start with a reason attached. Now say straight out whether the data backed it up. Do not leave the reader to work it out from the graph.
And if the data went against your prediction, write that down plainly. A hypothesis that turns out to be wrong is a result, not a failure, and pretending otherwise is far more damaging than being wrong was. What you must not do is quietly change the hypothesis afterwards so it matches what happened.
There is a middle case worth knowing. Sometimes the data neither supports nor refutes the prediction, because the scatter is too big to tell. Saying so — “the trend is in the predicted direction but the spread of results is too large to claim it with confidence” — is a much stronger answer than forcing a conclusion the data cannot carry.
Compare with the accepted value
You can only comment on accuracy if there is something to be accurate against. So find a literature value, say where you found it, and turn “quite close” into a number.
Cite the source in the same sentence: the IB data booklet, a named textbook, a chemical database. “I looked it up” is not a citation, and an assessor cannot check it.
The comparison that separates a good IA from an average one
Here is the move most students never make. You have two percentages by this point: the percentage error, which says how far you are from the accepted value, and the percentage uncertainty, which says how far off you could be just because of your equipment. Put them next to each other.
Both students had the same 3.2% uncertainty. What differs is the gap: 0.5% for A, 8.7% for B. That gap is the whole story.
What you find
What it means
How to say it
Percentage error is smaller than your percentage uncertainty
The literature value sits inside your range, so the gap could be explained by your equipment alone
“the result agrees with the accepted value within experimental uncertainty”
Percentage error is much larger than your percentage uncertainty
Your equipment cannot account for the gap, so something in the method is pushing every result the same way
“the difference is too large to be random, which points to a systematic error”
They are roughly the same size
Borderline. Your result is consistent, but only just
“the result is consistent with the accepted value, though the margin is small”
Notice what the second row buys you. You have not just admitted a problem — you have proved one exists, using two numbers you already worked out. That gives your evaluation something real to chase, instead of a list of things that might in principle have gone wrong.
What does not belong in a conclusion
New explanations. If the chemistry was worth saying, it was worth saying in the interpretation.
Improvements. They belong in the evaluation, and repeating them here wastes words.
Apologies. “Unfortunately my results were not very good” tells the reader nothing they can use.
Claims your data cannot carry. Four data points do not establish a law.
Vague comparisons. “Close to the literature value” is a feeling. 0.5% is a fact.
WORKED EXAMPLE
A student writes: “In conclusion the experiment went well and my answer was quite close to the real value, which proves that concentration affects the rate of reaction. Next time I would use better equipment.” Identify four faults.
Fault 1: no number anywhereThere is no result, no uncertainty and no literature value, so nothing here can be checked.Fault 2: “quite close” is not a comparisonAccuracy is quantified with a percentage error, not described with an adjective.Fault 3: it answers a textbook question, not this one“Concentration affects rate” was already known. The research question asked about a specific reaction over a specific range.Fault 4: the improvement is in the wrong sectionAnd “better equipment” is too vague to be an improvement anywhere.a better opening sentence“The initial rate of reaction was found to be directly proportional to the concentration of hydrochloric acid over the range 0.50 to 2.50 mol dm−3, with a gradient of 0.00999 dm3 mol−1 s−1.”
WORKED EXAMPLE
Student A found the enthalpy of neutralisation of HCl with NaOH to be −57 ± 2 kJ mol−1, a percentage uncertainty of 3.2%. The literature value is −57.3 kJ mol−1. Write the accuracy part of the conclusion.
Work out the percentage error|57 − 57.3| ÷ 57.3 × 100 = 0.52%Compare the two percentagesThe error of 0.5% is far smaller than the uncertainty of 3.2%, so the difference is well within what the equipment alone could produce.Check the range directly−57 ± 2 gives −55 to −59, which contains −57.3the result agrees with the accepted valueWritten out“The experimental value of −57 ± 2 kJ mol−1 gives a percentage error of 0.5% against the accepted value of −57.3 kJ mol−1. Since this is much smaller than the propagated uncertainty of 3.2%, the result is consistent with the literature value within experimental limits.”
WORKED EXAMPLE
Student B ran the same experiment and obtained −52.3 ± 2 kJ mol−1, also with a 3.2% uncertainty. Write their accuracy paragraph and say what it tells them.
Percentage error(57.3 − 52.3) ÷ 57.3 × 100 = 8.7%Compare8.7% against an uncertainty of 3.2%. The gap is nearly three times bigger than the equipment can account for.Check the range−52.3 ± 2 gives −50.3 to −54.3, and −57.3 is outside ita systematic error is presentWhich direction?Every value was less exothermic than it should be, which is exactly what heat escaping the cup would do. The conclusion states this; the evaluation is where it gets chased down.Note that B’s results could still be beautifully precise. Precision and accuracy are separate questions, and this is a result that answers them differently.
💡 Exam tip
Start with a sentence that answers the research question, naming the same chemicals and range.
Put the final value and its uncertainty in that first sentence or the one after.
State clearly whether the hypothesis was supported or not.
Give the literature value with a citation, then the percentage error.
Compare percentage error with percentage uncertainty and say what follows from it.
Keep it to a short paragraph. Length is not what is being marked.
⚠️ Common mix-up
Concluding general chemistry instead of the result of your own experiment.
Saying “close to” the literature value without ever calculating how close.
Dropping the uncertainty from the final answer, which makes accuracy impossible to judge.
Rewriting the hypothesis afterwards so that it agrees with the data.
Putting improvements in the conclusion, then repeating them in the evaluation.
Treating a refuted hypothesis as a failed experiment. It is a finding.
Up next: Evaluating — you have just shown, with two percentages, that something in the method was pushing your results one way. The next page is about finding out what, and saying what you would do about it.
Want this explained one-to-one?
Book a free session with an experienced IB Chemistry tutor and get your trickiest topics made simple.
