IB Biology Internal Assessment Guide

The International Baccalaureate Biology Internal Assessment (IA) is a crucial component of your final grade, accounting for 20% (SL) or 24% (HL) of your score. It is your opportunity to step into the role of a scientist, designing, executing, and communicating an original investigation. Mastering this task is not just about earning marks; it is about developing the analytical, critical thinking, and methodological skills that are fundamental to scientific inquiry. A well-executed IA demonstrates your ability to apply classroom knowledge to a real-world question, transforming theoretical understanding into practical competence.

Formulating a Clear Research Question and Identifying Variables

Your entire investigation hinges on a strong, focused research question. An effective question is specific, testable, and grounded in biological theory. It should clearly imply a cause-and-effect relationship. Avoid overly broad questions like “How does light affect plants?” Instead, aim for precision: “What is the effect of varying wavelengths of light (450nm, 550nm, 650nm) on the rate of oxygen production in Elodea canadensis, measured over a 10-minute interval?” This version specifies the independent variable (wavelength), the dependent variable (rate of oxygen production), the organism, and a key measurement condition.

This leads directly to identifying variables. The independent variable is the factor you deliberately manipulate (e.g., light wavelength, substrate concentration, temperature). The dependent variable is the factor you measure as the outcome (e.g., rate of reaction, growth, frequency). Controlled variables are all other conditions you must keep constant to ensure a fair test (e.g., species of plant, volume of solution, pH, duration of exposure). Defining these explicitly is the foundation of a valid experiment. Your research question should make both the independent and dependent variables immediately obvious to the reader.

Designing a Methodology for Reliable Data Collection

A robust methodology is your blueprint for generating reliable and sufficient data. It must be detailed enough for another student to replicate your experiment exactly. Describe your apparatus and materials with precision. The procedure should be a step-by-step narrative in the passive voice (e.g., “Five discs were cut from a spinach leaf using a cork borer”). Crucially, you must justify your choices. Why did you choose a sample size of 10? To reduce the impact of random error. Why did you repeat the experiment three times? To ensure repeatability and allow for statistical analysis.

Prioritize safety and ethical considerations, especially when working with living organisms. When collecting data, focus on accuracy and precision. Use appropriate measuring instruments (e.g., a colorimeter instead of subjective color comparison) and record raw data in organized, clearly labeled tables. Include units for every measurement. Plan to collect sufficient relevant data; for an experiment with a continuous independent variable like concentration, this means testing at least five different, appropriately spaced values, not just "high" and "low." This range allows you to identify trends and relationships effectively.

Applying Appropriate Statistical Analysis and Presenting Data

Raw data is meaningless without analysis. You must process your data to identify trends and test for significance. Begin with calculated means and measures of spread, such as standard deviation, which shows the variability within your data sets. Present your processed data clearly in tables and, most importantly, in graphical form. Choose the correct graph type: a line graph for continuous data (e.g., rate vs. concentration), a bar chart for categorial independent variables (e.g., enzyme type). Ensure all graphs are fully labeled with titles, scaled axes, and units.

Statistical testing moves your analysis from observation to inference. For comparing the means of two sets of data, you will typically use a t-test. For assessing the relationship or correlation between two continuous variables, linear regression or a Pearson’s correlation test is appropriate. For comparing observed data against an expected distribution (e.g., genetic crosses), a chi-squared test is used. You must state the null hypothesis, perform the calculation (showing one worked example in an appendix is excellent practice), state the critical value and degrees of freedom, and then clearly accept or reject the null hypothesis based on your calculated value. Simply stating a p-value is not enough; you must interpret what it means for your biological question.

Structuring the Written Report and Addressing Assessment Criteria

Your written report is the vehicle for your scientific story. Structure it to align perfectly with the IB assessment criteria (Design, Data Analysis, Conclusion, Evaluation). Use clear subheadings.

1. Introduction: Start with biological context, cite relevant theory, and logically lead to your focused research question and hypothesis.
2. Methodology: Write in paragraphs, not a bullet-point list. Include subsections for materials, variables, procedure, and safety/ethics.
3. Data Collection and Processing: Present raw and processed data tables. Show sample calculations. Include graphs with clear trend lines (not dot-to-dot).
4. Conclusion: Restate your findings, explicitly linking processed data and statistical results back to your hypothesis and the underlying biological theory. Explain the why behind the trends.
5. Evaluation: This is a critical section for high marks. Discuss the strengths and weaknesses of your methodology. Comment on the reliability (consistency) and validity (whether you measured what you intended) of your data. Suggest specific, realistic improvements for each weakness identified.

Developing a Critical Evaluation and Meaningful Improvements

The Evaluation section separates good IAs from excellent ones. Avoid vague statements like “human error was a problem.” Instead, identify specific systematic errors (flaws in the method that consistently skew results in one direction, e.g., a calibrated thermometer) and random errors (unpredictable variations, e.g., slight differences in measuring volumes). Quantify uncertainty where possible.

For each weakness, propose a specific, biologically sound improvement and, crucially, explain how this modification would enhance the reliability or validity of the investigation. For example: “Weakness: The enzyme solution was prepared once and used for all trials, potentially degrading over time. Improvement: A fresh aliquot of enzyme solution should be prepared for each individual trial from a stock solution kept on ice. This would minimize degradation, increasing the reliability of the measured reaction rates across trials.” This shows deep, critical reflection on your own work.

Common Pitfalls

Pitfall 1: A Vague or Untestable Research Question. A question like "How does pollution affect bacteria?" is doomed. Without defining "pollution" (a specific heavy metal concentration) and "affect" (growth rate measured by turbidity), you cannot design a valid experiment. Correction: Invest significant time refining your question. Ensure it is singular, focused, and contains your two key variables.

Pitfall 2: Insufficient or Low-Quality Data. Testing only two concentrations or having a sample size of n=3 provides an inadequate basis for analysis. Similarly, using subjective measures (e.g., "cloudiness" rated 1-5) introduces major bias. Correction: Plan for a wide range of at least five values for continuous variables and a minimum sample size of 5-10 for replicates. Use objective, quantitative measurement tools (photometers, data loggers, precise timers).

Pitfall 3: Graphic Misrepresentation and Missing Statistics. Creating a bar chart for continuous data or forgetting error bars obscures your results. Failing to perform a statistical test means you cannot objectively support your conclusion. Correction: Match your graph type to your data. Always include error bars (e.g., ±SD) on graphs. Select and correctly apply one appropriate statistical test to provide evidence for your claims.

Pitfall 4: Superficial Evaluation. Writing "the experiment could be more accurate" is worthless. It does not identify the source of inaccuracy or propose a targeted solution. Correction: Structure your evaluation around specific methodological steps. For each limitation, describe its impact on your results and provide a detailed, practical improvement that directly addresses it, explaining the expected benefit.

Summary

• Foundation First: A precise, testable research question with clearly defined independent, dependent, and controlled variables is the non-negotiable starting point for a successful IA.
• Methodology is Key: A replicable, detailed, and justified methodology that prioritizes the collection of sufficient, relevant, and precise data is essential for generating meaningful results.
• Analyze, Don't Just Describe: You must process raw data (means, standard deviation), present it effectively in graphs, and apply appropriate statistical tests to objectively interpret your findings and draw conclusions.
• Write to the Criteria: Structure your report explicitly around the IB assessment criteria (Design, Data Analysis, Conclusion, Evaluation) to ensure you address all areas examiners are scoring.
• Evaluate Critically: A high-scoring evaluation specifically identifies systematic and random errors, assesses reliability and validity, and pairs every weakness with a realistic, well-explained improvement.
• Show Your Work: Demonstrate your personal engagement and understanding by including annotated research, sample calculations, and a clear narrative that explains your scientific reasoning at every stage.