IB Science Experiment Design Guide | Variables, Methodology & Error Analysis
Struggling with experiment design for your IB science IA? This guide breaks down a universal framework—covering variables, methodology, and error analysis—that works across Biology, Chemistry, and Physics.
What Is the Core Framework for IB Science Experiment Design?
Every IB science investigation—whether it appears in your Internal Assessment (IA), a classroom practical, or a Paper exam question—rests on three pillars: variable classification, methodology, and error analysis. Get these right and the rest of your report falls into place naturally. Get them wrong and even excellent data cannot save your score.
The short answer is this: classify your variables precisely before you touch any equipment, write your methodology as if the reader has never seen your lab, and treat error analysis not as a concluding footnote but as a genuine scientific argument. The sections below break each pillar into actionable steps.
How Do You Correctly Classify Independent, Dependent, and Controlled Variables?
This is where most students lose marks without realising it—not because they don't know the definitions, but because they apply them carelessly.
Defining each variable type
| Variable | Definition | Common mistake |
|---|---|---|
| Independent variable (IV) | The one factor you deliberately change across trials | Being vague: "I changed the temperature" rather than specifying the range and intervals |
| Dependent variable (DV) | What you measure as a result of changing the IV | Confusing the measurement tool with the variable itself (e.g., writing "absorbance reading" when the variable is enzyme activity) |
| Controlled variables | All other factors that could affect the DV, held constant | Listing them without explaining why each must be controlled and how you will control it |
The operationalisation test
Before finalising your variables, apply the operationalisation test: can every variable be measured or set using a specific, real-world procedure? For example:
- Weak: "Temperature is the independent variable."
- Strong: "The temperature of the water bath, measured with a calibrated thermometer (±0.1 °C), is the independent variable, set to five values across a biologically relevant range."
The same principle applies to your dependent variable. State the instrument, its precision, and the unit of measurement.
Why controlled variables need justification
Examiners and moderators read hundreds of IAs. A bullet-point list of controlled variables with no explanation reads as a formality, not as scientific reasoning. For each controlled variable, write a brief justification:
pH is controlled by using a buffer solution prepared to the target pH for each trial, because enzyme activity is sensitive to hydrogen ion concentration—a shift in pH would confound any relationship observed between temperature and reaction rate.
This one sentence shows that you understand the variable's mechanism, not just its name.
How Do You Write a Methodology That Is Truly Reproducible?
Reproducibility is the standard. A stranger—your examiner, a fellow student, a scientist in another country—should be able to replicate your experiment exactly from your written method alone. This is not hyperbole; it is the explicit expectation in IB science assessment.
The reproducibility audit
After drafting your method, perform a reproducibility audit by asking:
- Are all materials listed with concentrations, masses, volumes, or other relevant specifications?
- Are all instruments named, along with their precision (e.g., electronic balance ±0.001 g)?
- Is the sequence of steps numbered and unambiguous?
- Is the number of trials stated, and is it sufficient to allow meaningful statistical treatment of data?
- Are safety and ethical considerations addressed where relevant?
If any answer is "no," revise before moving forward.
Step-by-step versus narrative format
Most IB students default to a numbered list, which is appropriate for procedural clarity. However, the setup and rationale are often better explained in a short paragraph before the numbered steps. This allows you to explain why you designed the experiment as you did—a consideration that earns marks in the Personal Engagement and Exploration criteria.
Matching your method to your research question
Your methodology must be a logical response to your research question. If your research question asks about the rate of a reaction, your method must include a way to measure rate (e.g., time taken to reach a fixed endpoint, or a continuous absorbance reading). If there is a mismatch between question and method, the entire investigation loses coherence.
For subject-specific methodology considerations, the IB Biology HL guide and the IB Chemistry HL guide both contain useful context on the kinds of investigations typical in each discipline.
How Do You Distinguish Between Random and Systematic Errors—and Why Does It Matter?
Error analysis is the section where students most frequently write something superficially correct but scientifically shallow. The distinction between random and systematic errors is not just terminological; it has direct consequences for how you interpret your data and propose improvements.
Definitions and examples
| Error type | Definition | Effect on data | Example |
|---|---|---|---|
| Random error | Unpredictable fluctuations that vary from measurement to measurement | Increases imprecision; data points scatter around the true value | Slight variations in timing when using a manual stopwatch |
| Systematic error | Consistent, repeatable deviation in the same direction | Decreases accuracy; all readings are shifted from the true value | A balance that reads 0.2 g too high on every measurement due to an uncalibrated zero point |
Why the distinction matters for your conclusion
Random errors affect your precision and can be reduced by increasing the number of trials and taking averages. Systematic errors affect your accuracy and cannot be reduced by repeating the same flawed procedure—they require a change in method, instrument, or calibration.
If your data shows a consistent offset from a literature or theoretical value, that is a signal of systematic error. If your repeat trials show a wide spread around a central value, random error is the dominant concern. Labelling these correctly and explaining their origins is what separates a strong evaluation from a generic one.
Writing about errors: the three-part structure
For each significant error, structure your analysis in three parts:
- Identify the error and classify it (random or systematic).
- Explain the cause—what in your method or equipment produced this error?
- Propose a specific, realistic improvement—not simply "be more careful," but a concrete procedural or instrumental change.
The manual timing of colour change introduced a random error, as human reaction time (typically in the range of several hundred milliseconds) varied between trials. This increased the spread of recorded times. A more precise approach would be to use a colorimeter connected to a data logger, which would record the endpoint automatically and eliminate observer-dependent variation.
Notice that this example does not blame human carelessness—it identifies a structural limitation of the method and offers a technical solution.
What Are the Most Common Mistakes in IB Science Experiment Design, and How Do You Avoid Them?
Understanding the pitfalls is as valuable as understanding the framework. Below are the errors that appear most frequently in student work, along with practical remedies.
Vague research questions
A research question that cannot be answered with an experiment is not a scientific question. "How does temperature affect enzymes?" is a topic, not a question. "How does changing the temperature of a reaction mixture affect the initial rate of enzyme-catalysed hydrolysis?" is a researchable question with a measurable DV and a manipulable IV.
Treating controlled variables as an afterthought
Many students write their controlled variables list after designing their method—then struggle to explain how those variables were actually controlled. The better approach is to identify your controlled variables during the design phase, so that your method explicitly includes the steps that control them.
Insufficient trials
A single measurement is not data—it is an observation. IB science assessments expect enough repeated trials to allow you to assess the spread of your results and calculate a meaningful measure of central tendency. The appropriate number of trials depends on your specific investigation; discuss this with your teacher and refer to the current subject guide.
Generic error analysis
"Human error" is not an acceptable error category in IB science. Every error must have a specific cause, a direction of effect, and a proposed improvement. Aim to discuss the two or three most significant sources of error in genuine depth rather than listing five errors superficially.
How Does This Framework Apply Across Biology, Chemistry, and Physics?
The three-pillar framework—variable classification, methodology, error analysis—is the same across all IB sciences. What differs is the content of each pillar.
| Pillar | Biology example | Chemistry example | Physics example |
|---|---|---|---|
| IV operationalisation | Substrate concentration (g dm⁻³), adjusted by serial dilution | Concentration of acid (mol dm⁻³), prepared by dilution from a stock solution | Voltage across a component (V), set using a variable resistor |
| DV operationalisation | Rate of O₂ production (cm³ min⁻¹), measured by gas syringe | Change in absorbance (arbitrary units), measured by spectrophotometer | Current through component (A), measured by ammeter |
| Controlled variable example | Temperature (maintained by water bath); pH (buffer solution) | Temperature; ionic strength of solution | Temperature (to prevent resistance changes); length of wire |
| Typical random error | Variation in biological material between organisms | Slight variation in reading a meniscus | Fluctuation in power supply voltage |
| Typical systematic error | Evaporation changing concentration over time | Impurity in reagents causing a consistent offset | Uncalibrated ammeter reading consistently high |
Subject-specific methodology—the types of apparatus, the standard analytical techniques, the field vs. lab context—varies substantially. For Physics in particular, the IB Physics HL guide provides useful context on the kinds of experimental setups you are expected to handle. Always confirm your experimental design against the official subject guide for your specific course and year.
Bringing It Together: A Pre-Submission Checklist
Before you submit any IB science investigation, work through these questions:
Variables
- [ ] Is the IV stated with a specific range and measurement intervals?
- [ ] Is the DV operationalised with an instrument, precision, and unit?
- [ ] Does each controlled variable include both a justification (why) and a method (how)?
Methodology
- [ ] Could a stranger replicate your experiment from your written description alone?
- [ ] Are all materials specified with relevant quantities and concentrations?
- [ ] Are instruments named with their precision values?
- [ ] Is the number of trials stated and justified?
Error analysis
- [ ] Is each error classified as random or systematic, with an explanation of its cause?
- [ ] Does each proposed improvement address the specific cause of the error?
- [ ] Is the error analysis connected back to the reliability and validity of your conclusion?
Applying this framework consistently—across your IA, your class practicals, and your exam responses—builds the kind of scientific reasoning that IB examiners are looking for. If you want to work through your specific investigation design with someone who has been through the IB process, the mentors at Quick IB offer one-to-one support tailored to exactly these challenges.