Raw readings almost never answer your research question directly. A stopwatch gives you seconds when what you wanted was a rate; a balance gives you grams when what you wanted was a percentage. Processing is the step that closes that gap — and the step where showing your working matters as much as getting the number right.
📘 What you need to know
Processed data goes in its own table, separate from the raw data.
Show one full worked example of every type of calculation you do.
Mean from your repeats, excluding any anomaly you have justified.
Rate = 1 ÷ time when you measured how long something took, or change ÷ time when you measured an amount.
Percentage change = (final − initial) ÷ initial × 100, which lets you compare samples that started differently.
Standard deviation quantifies the spread of your repeats — a mean without it is only half an answer.
Round to the significant figures your least precise raw measurement justifies.
Raw table, calculations, processed table
Keep the two tables apart. The raw table is the evidence; the processed table is the argument. Putting them in one grid makes it impossible for a reader to see which numbers you measured and which you worked out.
The middle box is the one students skip. Without it a reader has to take your processed numbers on trust, and method marks are lost.
Means, and what to do with an anomaly
The mean of your repeats is what you plot. The only real decision is whether every repeat belongs in it.
That is the giveaway for an anomaly pulling a mean: the average sits in a gap where none of your actual readings lie.
The rule for excluding a reading: it must be clearly outside the spread of its own repeats, and you must state a reason. “It did not fit my prediction” is not a reason. “The enzyme was added before the tube reached temperature” is.
Turning a time into a rate
If you measured how long something took, plotting time gets the biology backwards — a longer time means a slower reaction, so your graph slopes the wrong way. Take the reciprocal.
Rate from a time
rate = 1 ÷ time (units: s−1)
Rate from an amount
rate = change in the quantity ÷ time (units: cm3 s−1, mg min−1…)
WORKED EXAMPLE
At 40 °C the three trials gave 60 s, 58 s and 62 s. Calculate the mean time and the rate of reaction, giving the rate to the correct number of significant figures.
Step 1: the mean time(60 + 58 + 62) ÷ 3 = 180 ÷ 3 = 60.0 sStep 2: convert to a raterate = 1 ÷ 60.0 = 0.016666… s−1Step 3: round to match the raw data
The times were measured to 2 significant figures, but the mean is given to 3, so the rate is quoted to 3 s.f.
Mean 60.0 s, rate = 0.0167 s−1Copying 0.016666666 straight off the calculator claims a precision your stopwatch never had.
Temperature / °C (±0.5)
Mean time / s
Rate / s−1
Standard deviation of time / s
10.0
243.0
0.00412
3.0
20.0
151.0
0.00662
2.7
30.0
95.0 (anomaly excluded)
0.0105
1.4
40.0
60.0
0.0167
2.0
50.0
132.0
0.00758
4.0
Notice what the processed table does that the raw one could not: the rate column rises to a clear maximum at 40 °C and falls away either side. That shape is the answer to the research question, and it only appears after processing.
Percentage change
Use it whenever your samples did not start identical — which, with living material, is nearly always. It puts every sample on the same footing by comparing each one to its own starting value.
A bag of dialysis tubing has a mass of 12.40 g before immersion and 14.05 g after. Calculate the percentage change in mass.
Step 1: the change in mass14.05 − 12.40 = +1.65 gStep 2: divide by the starting mass1.65 ÷ 12.40 = 0.13306…Step 3: multiply by 100 and round0.13306… × 100 = 13.306…+13.3% (3 s.f.)Keep the plus sign. A gain and a loss of 13.3% are completely different results, and the sign is the only thing telling them apart.
Showing the spread
Because biological material varies so much, a mean on its own hides the thing a reader most needs to know: did your repeats agree?
Standard deviation is the usual choice. A small SD means the repeats clustered tightly around the mean; a large one means they were scattered.
Calculate the mean first — the standard deviation is measured from it.
Quote the SD in your processed table, and use it to draw error bars on your graph.
Comparing SDs between conditions tells you which measurements you can lean on hardest.
Look back at the table above. The standard deviation at 50 °C is double that at 30 °C — the readings got less consistent as the enzyme started to denature. That is a real biological observation hiding in a spread column.
Significant figures in processed values
Your answer can only be as precise as the least precise measurement that went into it.
Do the whole calculation first and round once, at the end. Rounding partway compounds the error.
Keep every value in a column to the same number of decimal places or significant figures.
Very small or very large processed values are clearer in standard form: 4.12 × 10−3 s−1 rather than 0.00412.
💡 Exam tip
One full worked example for each type of calculation — formula, numbers substituted, answer with units.
State plainly which reading you excluded and why, in the text as well as the table.
Convert times to rates before plotting, so the graph runs the way the biology does.
Always pair a mean with a standard deviation.
Round at the end, once, to match your raw data.
Give processed columns proper headings with units, exactly like raw ones.
Sanity-check every processed value against the raw numbers it came from.
⚠ Common mix-up
Excluding an anomaly with no justification, or worse, without saying you did.
Averaging in a reading that is clearly wrong, giving a mean that matches none of the data.
Plotting time instead of rate, so faster reactions appear lower on the graph.
Copying the full calculator display into the processed table.
Rounding at every step rather than once at the end.
Percentage change divided by the final value instead of the initial one.
Giving means with no measure of spread, so nothing can be said about reliability.
Merging raw and processed data into a single table.
Up next: Interpreting Results — graphing the processed data, describing the trend, explaining it with biology, and judging how far you can trust it.
Want this explained one-to-one?
Book a free session with an experienced IB Biology tutor and get your trickiest topics made simple.
