IB Chemistry HLInquiry 1 — Exploring and DesigningPaper 3 & IACore skill~12 min read
Controlling Variables
Listing your controlled variables is design. Actually holding them still is a practical skill, and it is the difference between data you can trust and data that only looks tidy. This page is about the three places control really happens: your instruments, your surroundings, and the heat leaking out of your apparatus.
📘 What you need to know
Controlling variables is how you keep an investigation valid — so any change you see really did come from your independent variable.
The main enemy is systematic error, which shifts every result the same way and cannot be averaged out.
Calibration means checking an instrument against a known standard and correcting it if it is off.
pH meters are calibrated with buffer solutions; thermometers with melting ice and boiling water; balances are tared before every mass.
Lab conditions matter: use a water bath for temperature, a draught shield for air currents.
In calorimetry, heat loss is nearly always the largest systematic error — insulate, add a lid, and use a polystyrene cup.
If a variable genuinely cannot be controlled, monitor and record it, then discuss it in your evaluation.
Why this matters more than it sounds
Imagine you are measuring how concentration affects rate, and the lab warms up by four degrees while you work. Your later runs are faster — but not because of concentration. You now have a graph that looks fine and means nothing, and no amount of repeating will reveal the problem, because the problem is not random.
That is the whole argument. Controlling variables protects the link between what you changed and what you measured. Break that link and the experiment stops answering your research question, no matter how neat the results look.
A useful mental test: for each controlled variable, ask “if this drifted while I worked, would my graph still look believable?” If the answer is yes, that variable is dangerous, and it needs a real method holding it steady.
Calibrating your instruments
Every instrument you use is claiming something about itself. Calibration is you checking that claim against something you already know the answer to.
Instrument
Check it against
What you are looking for
pH meter
At least two standard buffers, e.g. pH 4.00 and pH 7.00
Both readings correct; adjust the probe if not
Thermometer or temperature probe
Melting ice (0.0 °C) and boiling distilled water (100.0 °C at standard pressure)
A constant offset, or a stretched scale
Digital balance
Tare to zero before every mass; check with a known mass if one is available
Zero drift, and anything left on the pan
Colorimeter
Zero it with a cuvette of the pure solvent
Absorbance readings that start from a true zero
Volumetric glassware
The tolerance printed on the glass itself
The uncertainty you should be quoting
This is why two fixed points are used rather than one. Ice alone would show the thermometer is 1.2 too high; only the boiling point reveals that its scale is slightly compressed as well.
WORKED EXAMPLE
Correcting readings from a thermometer that is off
A thermometer reads 1.2 °C in melting ice and 100.8 °C in boiling distilled water at standard pressure. During a calorimetry run it reads 24.6 °C before mixing and 31.4 °C at the peak. Find the corrected temperatures and the corrected temperature rise.
Step 1: How far apart are the two fixed points on this scale?100.8 − 1.2 = 99.6 degrees, where there should be 100.0Step 2: Build the correctiontrue = (reading − 1.2) × 100 ÷ 99.6Step 3: Correct both readings(24.6 − 1.2) × 100 ÷ 99.6 = 23.49 → 23.5 °C(31.4 − 1.2) × 100 ÷ 99.6 = 30.32 → 30.3 °CStep 4: The temperature rise
Read: 31.4 − 24.6 = 6.8. Corrected: 30.3 − 23.5 = 6.8
Absolute readings shift by 1.1 °C, but ΔT barely movesthe offset cancels in a subtraction — so for ΔT only the 0.4% stretch matters
That last line is worth remembering. If your dependent variable is a temperature change, a constant offset cancels itself out. If it is an absolute temperature — a boiling point, a melting point, a solubility at a stated temperature — the offset goes straight into your answer, and calibration really matters.
Keeping the lab conditions steady
The room is part of your apparatus, whether you planned it that way or not.
Temperature
In anything kinetic, temperature is usually the most powerful variable in the room — a rise of ten degrees can roughly double a reaction rate. Standing your solutions in a thermostatically controlled water bath for five to ten minutes before mixing is the standard fix, and it works because it brings everything to the same starting point rather than just hoping they match.
Air currents
A draught from a window or an air-conditioning vent will quietly cool a calorimeter and steal your temperature rise. Close windows, move away from the vent, and put a simple draught shield around the apparatus — a card box open at the front is enough.
Pressure and humidity
Usually you can ignore these, with one exception: if you are collecting a gas or using a boiling point as a fixed point, atmospheric pressure genuinely shifts the answer. Record the pressure if your school has a barometer, or at least say you assumed standard pressure.
The simple ones people forget
Use the same balance for all masses, so any small zero error is at least consistent.
Use the same person for anything judged by eye — an endpoint colour, a cross disappearing.
Use the same batch of solution, made up once, rather than mixing fresh each session.
Clean and dry glassware between runs, or the leftovers become an uninvited variable.
Insulating against heat loss
In any calorimetry experiment the biggest systematic error is heat going somewhere you did not want it to. You will never stop it completely. What you can do is slow it down and then account for what is left.
The holes in the lid have to be small. Big enough for the thermometer and stirrer, no bigger — every extra millimetre is another route for warm vapour to leave.
🧩 Standard insulation, and what each part is for
Polystyrene cup instead of a glass beaker. Polystyrene conducts heat far more slowly than glass, so less is lost through the walls.
Cup inside a larger beaker. Gives stability, and the trapped air between them is itself a poor conductor.
Lid with small holes. Cuts the loss by evaporation and by warm air convecting off the surface.
Draught shield around the outside. Stops moving air stripping heat off the apparatus.
Record and extrapolate. Take temperatures at fixed intervals and extrapolate the cooling line back to the moment of mixing, to estimate the rise you would have got with no loss at all.
WORKED EXAMPLE
Choosing the controls that actually matter
Research question: “What is the effect of alcohol chain length (methanol to butan-1-ol) on the enthalpy change of combustion?” Give three controlled variables, the method of control, and why each one matters.
Control 1: mass of water being heated100.0 cm³ measured with the same measuring cylinder each time.
Because q = mcΔT — change m and every energy value shifts.
Control 2: distance from wick to calorimeter baseFix at 5.0 cm using a clamp and a ruler, every run.
Because a lower flame delivers more of its heat into the can and less to the room.
Control 3: air movement around the flameCard draught shield on three sides, windows closed.
Because a flickering flame loses heat sideways and the loss is not the same twice.
Three controls, three methods, three reasonsrank them — heat loss dominates here, so say so rather than treating all controls as equal
When you genuinely cannot control something
Some variables are simply out of reach in a school lab. Room temperature drifts across an afternoon. Atmospheric pressure changes between sessions. The purity of an old bottle of reagent is whatever it is.
The honest response is not to pretend. It is to monitor and record: write the value down alongside each trial, then look at whether it tracks your results, and say what you found in the evaluation. That is a genuinely strong thing to be able to write, and it beats a blanket claim that everything was held constant.
WORKED EXAMPLE
Handling a variable you cannot hold still
A student runs 15 trials across one afternoon. The lab thermometer reads 19.4 °C at the start and 23.1 °C by the end. There is no water bath available. What should they do?
Step 1: Accept it cannot be fixed todayDrift of 23.1 − 19.4 = 3.7 °C across the session.Step 2: Record it against every trial
Add a room-temperature column to the results table, not a single value at the top.
Step 3: Break the pattern between drift and orderRun the concentrations in a shuffled order, not lowest to highest.
So the warming is spread across all values instead of piling onto the last ones.
Step 4: Say what you found
Check whether trials done late look systematically faster, and report it either way.
Monitor, randomise the order, then discuss it honestlyshuffling the run order turns a systematic drift into something closer to random scatter
💡 Exam tip
For every control, write three things: what is controlled, how, and why it matters. Two out of three loses the mark.
Rank your controls. Saying which one has the biggest effect on your results shows insight; treating them all as equal does not.
“Calibrate” means check against a known standard. Zeroing a balance is calibration; wiping it clean is not.
In calorimetry, always name heat loss as the dominant systematic error, then say specifically how you reduced it.
If a control failed, say so. An honest limitation reads far better than a claim the examiner will not believe.
Randomising the order of your runs is a cheap, powerful trick against any condition that drifts with time.
⚠ Common mix-up
Confusing a controlled variable with a control experiment. A controlled variable is one you hold steady; a control experiment is a run with the factor removed.
Thinking more repeats fix a drifting room. A drift is systematic. Averaging keeps it.
Calling “using the same beaker” calibration. That is consistency, not calibration.
Believing insulation removes heat loss. It slows it. You still extrapolate for the rest.
Recording room temperature once at the start and calling it controlled.
Assuming water boils at exactly 100.0 °C. Only at standard pressure, and only if it is pure.
Listing ten controls with no reasons. Three well-justified ones score better than a long unexplained list.
Up next: Inquiry 2 — Collecting and Processing Data — recording results properly, handling uncertainties through a calculation, and turning a table into a graph that says something.
Want this explained one-to-one?
Book a free session with an experienced IB Chemistry tutor and get your trickiest topics made simple.
