Controlling IB Biology IA variables means doing more than listing what you change, measure and keep constant. You must show that your investigation isolates a meaningful relationship, explain how other biological factors could affect the result, and document procedures precisely enough for another researcher to reproduce your work.
Under the current course, first assessed in 2025, the internal assessment is called the scientific investigation. It contributes 20% of the final Biology grade, has a maximum report length of 3,000 words, and is assessed using four criteria worth 24 marks in total. Research design is worth 6 marks, so careful treatment of variables directly supports one quarter of the available IA marks.
What are the three types of IB Biology IA variables?
In a controlled experiment, variables have three distinct roles.
| Variable type | Definition | Catalase investigation example |
|---|---|---|
| Independent variable | The factor deliberately changed by the researcher | Hydrogen peroxide concentration in mol dm⁻³ |
| Dependent variable | The response measured or calculated | Initial rate of oxygen production in cm³ s⁻¹ |
| Controlled variable | A factor kept sufficiently constant because it could otherwise affect the dependent variable | Temperature, pH, catalase concentration and total reaction volume |
Your research question should normally identify the independent and dependent variables, including how the dependent variable will be measured. The current Biology guide also allows investigations involving two correlated variables, as may occur in database or fieldwork studies. In that situation, it is more accurate to describe predictor and response variables than to claim that you manipulated a factor.
Independent variable
The independent variable is the factor whose effect you are testing. It needs a justified range, appropriate intervals or categories, and a reliable method of manipulation or measurement.
For example, moving a lamp to different distances from pondweed does not directly establish light intensity. Distance is what you manipulate, but actual light intensity should ideally be measured with a light sensor because ambient light, lamp output and the inverse-square relationship may complicate the treatment. A pilot study can help determine a range that produces measurable differences without damaging the organism or causing every response to plateau.
There is no universal IB rule requiring exactly five independent-variable levels. Using several well-chosen levels is often sensible because it reveals trends more convincingly than a simple two-group comparison, but the appropriate number depends on the biological question, method and available time. What the official guidance emphasizes is justifying the range and quantity of measurements.
Dependent variable
The dependent variable is the biological response you measure or derive from measurements. Define it operationally, which means specifying exactly how it will be quantified.
For example, “enzyme activity” is too vague. A stronger definition would be “initial catalase activity, calculated from the gradient of oxygen volume against time during the first 30 seconds, in cm³ s⁻¹.” This tells the examiner what was recorded, how the result was processed and which units will appear in the analysis.
Choose an outcome that is sensitive enough to detect change. Counting pondweed bubbles is less reliable than measuring oxygen volume because bubble sizes vary, while percentage change in potato mass may be more comparable than absolute mass change when samples have slightly different starting masses.
Controlled variables
A controlled variable is a biological, chemical or environmental factor that could influence the dependent variable independently of the treatment. Controlling it reduces alternative explanations for the observed pattern.
If temperature rises during a catalase investigation, reaction rate may increase because molecules have greater kinetic energy rather than because hydrogen peroxide concentration changed. If potato cylinders differ in diameter, their surface-area-to-volume ratios may affect the rate of osmosis. Relevant controls therefore follow from biological reasoning, not from producing the longest possible list.
Controlled variables are not the same as a control group
These terms are often confused.
| Term | Purpose | Example |
|---|---|---|
| Controlled variable | A factor maintained consistently across treatments | Keeping catalase reactions at the same temperature |
| Control treatment or group | A reference condition used for comparison | A catalase mixture containing no hydrogen peroxide |
| Negative control | Should produce no response, helping reveal contamination or unintended effects | A boiled-enzyme treatment expected to show little or no catalase activity |
| Positive control | Should produce a known response, demonstrating that the method can detect an effect | A standard treatment known to produce measurable oxygen |
A control treatment is not automatically required in every IA. It should be included when it answers a methodological question, establishes a baseline or confirms that the procedure works. Adding a nominal control with no clear purpose does not strengthen the investigation.
How variable control supports Research design marks
The current Research design criterion assesses how effectively you communicate the purpose and practice of the methodology. Relevant methodological considerations include selecting methods for measuring variables, choosing the range and quantity of measurements, identifying control variables, and considering safety, ethical and environmental issues.
At the strongest level, the research question is placed in a specific biological context, methodological choices are explained, and the method can be reproduced. A variable table helps, but it cannot replace a detailed procedure. The examiner needs to see how the controls operate in practice.
A convincing controlled-variable table contains four elements:
| Controlled variable | Why it matters biologically | How it will be controlled | Evidence or precision |
|---|---|---|---|
| Temperature | Catalase activity is temperature-dependent because temperature affects collision frequency and enzyme structure | Equilibrate solutions in a thermostatically controlled water bath before mixing and keep the reaction vessel in the bath | Record temperature before each trial with a thermometer of stated uncertainty |
| pH | Changes in hydrogen-ion concentration can alter bonding and the enzyme's active-site structure | Prepare every treatment using the same batch and volume of buffer | State buffer pH, concentration and measured pH if checked |
| Catalase concentration | More enzyme molecules provide more available active sites | Homogenize one stock preparation and transfer the same measured volume to every trial | State stock preparation, aliquot volume and apparatus uncertainty |
| Reaction time | Longer collection periods can produce greater oxygen volumes even when rate is unchanged | Begin timing at mixing and collect readings at identical intervals | Use the same timing procedure and report stopwatch resolution |
Only report a tolerance such as 25.0 ± 0.5°C if your apparatus and monitoring genuinely achieved it. Unsupported precision makes a method less credible, not more scientific.
How to write a reproducible variables section
Use the following process when planning your IA.
- Write the causal or correlational claim clearly. Identify what changes, what is measured and the biological system involved.
- Define both main variables operationally. Include units, instruments, calculation methods and timing.
- Trace possible alternative causes. Consider organism age, genotype, size, health, temperature, pH, light, concentration, exposure time and sampling location where relevant.
- Prioritize variables that could materially affect the outcome. There is no official requirement for a fixed number of controlled variables.
- Explain why each control matters. Connect it to biological theory rather than writing only “for fairness.”
- State exactly how control will be achieved and checked. Include quantities, equipment, calibration, randomization or tolerances where appropriate.
- Integrate controls into the method. A researcher should not need to infer how your table translates into experimental steps.
The RevisionDojo guide to designing effective science IA experiments can help you test whether the whole design is feasible. The Biology scientific investigation essentials guide is also useful for checking how variables fit into the complete report.
What to do when a variable cannot be held constant
Biological systems contain natural variation, so perfect control is rarely possible. The scientifically honest response is to reduce, monitor or distribute that variation rather than claiming it disappeared.
You can:
- use organisms of similar age, size or developmental stage
- take samples from the same stock or source population
- randomly allocate specimens to treatment groups
- randomize treatment order to reduce time-related bias
- use blocking when location or batch could influence results
- measure a potential confounder and report its observed range
- conduct equal numbers of replicates under each condition
- acknowledge remaining variation during evaluation
A confounding variable is associated with both the explanatory variable and the outcome, making their relationship difficult to interpret. In fieldwork, for example, an apparent relationship between distance from a path and plant abundance might be confounded by soil moisture if moisture also changes with distance. Recording soil moisture, using stratified sampling or redesigning the transect could make the conclusion more defensible.
Common variable-control mistakes
Listing equipment as controlled variables
“Use the same colorimeter” describes procedural consistency, not the biological variable itself. The relevant controlled factors may be wavelength, cuvette orientation, calibration procedure or sample volume. Explain the factor that can affect the reading, then show how equipment standardization manages it.
Saying only that a variable was kept the same
“Temperature was controlled” is not reproducible. State the target temperature, apparatus, equilibration procedure, monitoring method and realistic variation. The same principle applies to pH, light intensity and solution volume.
Confusing repeats with controls
Replicates improve your estimate of variation and increase reliability, but they do not prevent systematic confounding. Repeating an experiment five times at steadily increasing room temperatures gives more data without isolating the independent variable.
Controlling variables that should be measured
Do not call a factor controlled if it was merely assumed constant. If ambient temperature or light could fluctuate, record it. Your evaluation can then assess whether the observed variation was large enough to influence the result.
Using an ambiguous research question
Questions such as “How does salt affect plants?” do not identify the treatment, response, organism or measurement period. The RevisionDojo sample Biology IA investigation and Biology IA exemplar library can help you compare broad questions with operationally defined ones without copying another student's design.
A practical final check
Before collecting data, ask another student to reconstruct your investigation using only your plan. If they cannot determine the treatment values, measurement timing, sample allocation or control procedures, the design is not yet reproducible. A small pilot can reveal uncontrolled heating, unsuitable ranges, ceiling effects and measurements that are too imprecise.
After completing the IA, use RevisionDojo's Biology Questionbank to practise interpreting experimental variables in unfamiliar contexts. For exam preparation, the IB Biology past paper video solutions show how to identify variables and evaluate experimental designs in data-based questions.
Conclusion
Strong control of IB Biology IA variables begins with a precisely defined independent variable, an operationally measurable dependent variable and a selective set of biologically relevant controlled variables. For Research design, explain why each control matters, how it was implemented, how consistency was checked and what variation remained.
Do not treat a variable table as a formality. It is evidence that your method can isolate the relationship in the research question and generate sufficient relevant data. RevisionDojo's IA exemplars and Jojo AI can help you review clarity, while the Biology Questionbank and past paper video solutions provide focused practice with experimental design.
Sources and referenced URLs
- IB Biology subject brief, first assessment 2025
- IB Biology curriculum updates
- IB Biology guide, first assessment 2025
- MIT OpenCourseWare experimental design guidance
- US National Library of Medicine guide to dependent, independent and confounding variables
- RevisionDojo guide to effective Science IA experiments
- RevisionDojo Biology scientific investigation essentials
- RevisionDojo sample Biology IA investigation
- RevisionDojo Biology IA exemplar library
- RevisionDojo IB Biology Questionbank
- RevisionDojo IB Biology past paper video solutions
