IB Chemistry HLInquiry 2 — Collecting and Processing DataPaper 3 & IACore skill~11 min read
Collecting Data
This is the doing part — the bit that feels like real chemistry. But the marks here are not for being busy at the bench. They are for what ends up written down: numbers recorded to the right precision, in a table built before you started, alongside the things you noticed with your own eyes.
📘 What you need to know
Raw data is what you directly measured. No calculations, no averages — those come later, in a separate table.
Design your raw data table before the lesson, so you are filling it in rather than inventing it under pressure.
Units and uncertainties go in the column headers, never repeated next to every number.
Record every value to the precision your instrument actually gives, including trailing zeros.
Aim for five values of the independent variable and three trials at each.
Qualitative observations — colours, fizzing, precipitates, smells, soot — are data, and they often explain your numbers later.
When something goes wrong, write down what happened and what you did. Never quietly erase it.
What actually counts as raw data
Raw data is the numbers you read off an instrument. Nothing else. A burette reading is raw data; a titre is not, because you subtracted to get it. A stopwatch time is raw data; a rate is not, because you divided.
That distinction matters because your first table has to show the examiner what you did, not what you worked out. If you only record titres, nobody can check your subtraction, and you have quietly hidden half your measurements.
The test: could someone recreate your calculations from this table alone? If a number in it came out of a calculator rather than off an instrument, it belongs in the processed table instead.
Build the table before you start
A table designed at the bench, mid-experiment, is always a mess. Draw it the night before, with every column headed and every uncertainty already written in, and collecting data becomes filling in blanks.
Strictly, the titre column is already processed data. It is allowed here because it is a simple subtraction, but the two readings it came from must still be visible.
Record to the precision the instrument gives you
This is one of the easiest places to drop marks, and one of the easiest to fix. If a balance shows 5.00 g, you write 5.00 g. Writing 5 g throws away information the instrument gave you for free, and it tells the examiner you do not understand what those zeros mean.
Reading between the marks is normal and expected. On a burette marked every 0.1 cm3 you judge to the nearest half-division, which is why the last digit is always a 0 or a 5.
Digital displays are the easy case — copy exactly what it says. Analogue scales need you to estimate the last digit yourself, and that estimate is the whole reason your uncertainty is half a division rather than a whole one.
Qualitative observations are data too
Students treat these as an afterthought and then wonder what to write in their evaluation. Your observations are the evidence that explains why the numbers came out as they did.
What you notice
What it might be telling you
Where it becomes useful
A colour change that creeps in slowly
The endpoint is hard to judge
Explaining scattered titres in your evaluation
Fizzing that dies away before you expect
One reactant has run out
Explaining why a rate curve flattens early
Black soot on the base of a calorimeter
Incomplete combustion
Explaining an enthalpy value far below the literature one
A precipitate forming when none was planned
A side reaction, or an impure reagent
Justifying an anomalous result
The flask feeling warm to the touch
An exothermic step you did not account for
Explaining a drifting rate through a run
Write observations as you go, not from memory. “The solution went pale pink after one drop and stayed pink” is a useful record. “It changed colour” written up two weeks later is not.
When something goes wrong mid-experiment
Experiments misbehave. Reacting to that sensibly, at the bench, is a genuine scientific skill — and it is assessed.
🧩 What to do when a run goes wrong
The reaction is far too fast or too slow. Adjust the concentration or the temperature so it becomes measurable — then record what you changed and why.
One trial looks nothing like the others. Keep it in your table, then run an extra trial so you still have three that agree.
You cannot judge the endpoint. Note that it was gradual. It becomes a real limitation to discuss rather than a vague apology.
You spill, overshoot, or lose some product. Discard that run, say so, and repeat it. Do not process a run you know was wrong.
Anything unexpected at all. Write it in your notes at the time. You will not remember it later, and it is usually the thing that explains your results.
WORKED EXAMPLE
Finding the faults in a raw data table
A student hands in a table headed “Results”, with one column labelled “Titre” containing the values 21.4, 21.35, 21.3, and a final column labelled “Average = 21.35 cm3“. Identify four problems.
Fault 1: the title says nothing
“Results” could head any table in any subject. Name the chemicals and the measurement.
Fault 2: the burette readings are missingOnly the titres are shown, so the subtraction cannot be checked.Fault 3: inconsistent decimal places21.4 and 21.3 should be 21.40 and 21.30 — the burette gives 2 d.p.Fault 4: a calculated value in the raw table
The average is processed data. It belongs in the second table, with the working shown.
No title, no raw readings, ragged precision, calculations in the wrong tablealso missing: units and uncertainty in the headers
WORKED EXAMPLE
Which readings are written down wrongly?
A student records: burette 18.2 cm3; balance (2 d.p.) 3.4 g; thermometer marked every 1 °C reading 22 °C; gas syringe marked every 1 cm3 reading 46.5 cm3. Correct any that are wrong.
Burette: 18.2 cm³Wrong → should be 18.20 cm³ (2 d.p., last digit 0 or 5)Balance: 3.4 gWrong → should be 3.40 g, matching the displayThermometer: 22 °CWrong → should be 22.0 or 22.5 °C, estimated to half a divisionGas syringe: 46.5 cm³Correct — half of the 1 cm³ division is exactly rightThree wrong, one right — and the fix is always the trailing digitif it feels like writing a pointless zero, that zero is doing real work
WORKED EXAMPLE
Dealing with an odd trial while you are still at the bench
Three trials of a gas collection at the same concentration give 18.5, 18.7 and 24.2 cm3 in 30 s. The third run happened just after the bung was pushed in late. What should the student do now, and what must they not do?
Step 1: Notice it during the session, not afterwards
24.2 sits well outside the other two, and there is a known cause.
Step 2: Record the observation immediately“Trial 3: bung fitted late, gas escaped at the start.”Step 3: Run a fourth trialGives 18.6 cm³ → now three trials agree.Step 4: Keep the odd one in the table
It stays in the raw data, flagged, and you justify excluding it when you average.
Repeat it, flag it, keep it — never delete itan excluded result with a stated reason reads far better than a table that is suspiciously perfect
💡 Exam tip
Give every table a title that names the chemicals and the measurement, not just “Results”.
Units and uncertainties go in the header only. Never write “cm3” after every number.
Keep the decimal places consistent down a whole column.
Show both burette readings, and both masses if you weighed by difference.
Record qualitative observations in the same session, in a labelled section — not scribbled in a margin.
If you change anything mid-experiment, write down what you changed and why. That is evidence of good practice, not an admission of failure.
⚠ Common mix-up
Putting averages in the raw data table. An average is calculated, so it belongs in the processed table.
Recording only the titre. Both readings, always.
Dropping trailing zeros. 5 g and 5.00 g claim very different things about your balance.
Writing units next to every value instead of once in the header.
Treating observations as decoration. They are the evidence you will need in the evaluation.
Deleting an anomalous trial on the spot. Keep it, flag it, and justify excluding it later.
Quoting a value with more decimal places than the instrument can give. A measuring cylinder cannot read 25.00 cm3.
Up next: Processing Data — turning those raw readings into a final answer, and carrying the uncertainty through every step of the calculation with it.
Want this explained one-to-one?
Book a free session with an experienced IB Chemistry tutor and get your trickiest topics made simple.
