The IB Chemistry internal assessment is officially called the scientific investigation. It is an open-ended investigation in which you formulate a research question, collect and analyse quantitative data, and submit an individual written report of no more than 3,000 words. For both SL and HL, it contributes 20% of the final Chemistry grade and is assessed using four equally weighted criteria worth 6 marks each, giving 24 marks in total.
This IB Chemistry IA guide explains the requirements applying in 2026, from selecting a feasible topic to checking the final report. It follows the Chemistry course first assessed in 2025, so older advice based on Personal Engagement, Exploration and Communication criteria is no longer current.
IB Chemistry IA requirements for 2026
The IB describes the scientific investigation as an open-ended task in which a student gathers and analyses data to answer their own formulated research question. The official subject brief assigns approximately 10 hours to the task, although schools determine their own schedules and internal deadlines.
The central requirements are:
- Weighting: 20% of the final Chemistry result at SL and HL
- Maximum report length: 3,000 words
- Assessment: 24 marks across four criteria
- Final product: an individual written report
- Research focus: a student-formulated question answered through data
- Marking: initially marked by the teacher and subject to IB moderation
Charts, diagrams, data tables, equations, formulas, calculations, citations, references, the bibliography and headers are excluded from the word count. The report should begin with the investigation title, candidate code, candidate codes of group members where applicable, and word count. A cover page and contents page are not required.
The current course permits limited collaboration. Students may work in small groups and, where appropriate, use similar methodologies, but each student must investigate a different independent or dependent variable and collect unique data. The submitted report must remain the student's own work.
The four IB Chemistry IA criteria
Each criterion contributes 6 marks, so planning, analysis, conclusion and evaluation deserve comparable attention.
| Criterion | Marks | What it assesses |
|---|---|---|
| Research design | 6 | How effectively the research question, scientific context and reproducible methodology are communicated |
| Data analysis | 6 | How clearly and accurately data are recorded, processed and presented, including appropriate treatment of uncertainties |
| Conclusion | 6 | How successfully the research question is answered using the analysis and accepted scientific context |
| Evaluation | 6 | How effectively methodological weaknesses are explained and realistic improvements are proposed |
| Total | 24 | Scaled to form the IA's 20% contribution |
A useful starting point is RevisionDojo's current IB Chemistry IA criteria guide. Avoid relying on older exemplars without checking their date because reports assessed under the former five-criterion framework do not model the current rubric exactly.
Step 1: Choose a focused, feasible topic
A strong topic produces measurable variation, allows meaningful chemical analysis and can be investigated safely with available equipment. Complexity alone does not earn marks. A controlled titration or rate experiment with sufficient data is usually more defensible than an ambitious investigation that cannot be reproduced reliably.
Begin by identifying a chemical relationship that interests you, then assess:
- Which variable can be changed systematically?
- Which outcome can be measured quantitatively?
- Can important variables be controlled?
- Is the equipment sufficiently precise?
- Can enough repeated measurements be collected?
- Are the chemicals and procedures safe and environmentally responsible?
- Is there accepted scientific literature against which results can be compared?
The RevisionDojo Chemistry IA ideas collection can help generate starting points, but an idea must be adapted to your laboratory, interests and available time. Copying a published question and method weakens authenticity and may create academic integrity concerns.
Step 2: Formulate the research question
A useful experimental research question normally identifies the independent variable, dependent variable and chemical system. It may also specify the measurement method or relevant conditions when these define the investigation's scope.
For example:
How does temperature from 298 K to 338 K affect the rate constant for the reaction between sodium thiosulfate and hydrochloric acid, determined using colorimetry under constant reactant concentrations?
This question establishes a numerical range, measurable outcome, system and analytical approach. By contrast, “How does temperature affect reaction rate?” is too broad because the reaction, range, measurement and controlled conditions are undefined.
Correlation-based, database and model-based questions may require different wording. In every case, the question must be narrow enough to answer with the data you can realistically obtain. RevisionDojo's guide to designing an effective science IA experiment provides further planning prompts.
Step 3: Build the research design
The background section should explain only the chemistry needed to understand your decisions and later interpret the results. Relevant equations, molecular explanations and predicted relationships are useful; a long summary of an entire syllabus topic is not.
Your methodology must allow another trained student to reproduce the investigation. Include:
- reagent identities, concentrations, quantities and preparation methods
- apparatus capacities, precision and instrumental uncertainties
- the selected independent-variable range and intervals
- repeated trials or sampling frequency
- the method used to measure the dependent variable
- controlled variables, how they were controlled and why this matters
- the planned method of processing the data
- safety, ethical and environmental considerations
Justify consequential choices. If five temperatures are selected, explain why the range should reveal a trend without causing evaporation or another interfering process. If colorimetry is used instead of a visual endpoint, explain how it improves objectivity or sampling frequency.
Conduct a pilot study before final data collection. A pilot can reveal that changes are too small for the apparatus, the range is unsafe, or the endpoint is inconsistent. Revising the procedure at this stage is good experimental practice, not evidence that the original idea failed.
Step 4: Collect sufficient, honest data
Create raw-data tables before entering the laboratory. Column headings should contain the measured quantity and unit, while instrumental uncertainty should be stated clearly. Record all trials at the precision supported by the instrument and add relevant qualitative observations, such as unexpected precipitation, colour changes or gas leakage.
Collect enough data to establish a defensible relationship. There is no universal IB rule requiring a particular number of independent-variable values or repeats; the appropriate quantity depends on the method and expected variation. As practical guidance, several well-spaced values and repeated measurements are normally necessary to reveal trends and estimate variability.
Do not delete an anomalous value merely because it disrupts the graph. Retain it, investigate plausible causes and explain any decision to exclude it from a calculation. Fabricating, altering or selectively hiding results is academic misconduct.
Step 5: Analyse data and uncertainties
Data analysis is not simply presenting a spreadsheet. It should transform measurements into evidence that answers the research question. The reader must be able to follow each stage from raw results to processed values and final interpretation.
Include, where appropriate:
- clearly labelled raw and processed data tables
- one representative calculation
- correct units, decimal places and significant figures
- treatment and propagation of measurement uncertainties
- means and measures of spread
- graphs with labelled axes, units and suitable error bars
- justified best-fit models rather than automatically chosen trendlines
- discussion of patterns, scatter and anomalous results
- statistical tests where they genuinely help answer the question
The correct uncertainty treatment depends on the mathematical operation and investigation. Avoid attaching error bars or calculating percentage uncertainty mechanically without explaining what they indicate. RevisionDojo's Chemistry IA data analysis guide can support this stage, while the broader IB Chemistry resources help review underlying concepts.
Step 6: Write a justified conclusion
Open the conclusion with a direct answer to the research question. Then support it using key processed values, trends, uncertainty ranges, graph parameters or statistical results already presented in the analysis.
A high-quality conclusion also compares the outcome with accepted scientific context. This might include a published value, established model or theoretical prediction from a credible textbook, database or peer-reviewed paper. Cite the source precisely and discuss agreement quantitatively where possible rather than stating that the result was simply “close.”
Uncertainty matters to interpretation. If two values overlap within their uncertainty ranges, the evidence for a meaningful difference may be limited. Likewise, a high coefficient of determination does not by itself prove causation or validate the method.
Step 7: Evaluate the investigation
Evaluation requires more than listing “human error.” Identify specific methodological weaknesses, explain how each affected the data or conclusion, and propose an improvement that directly addresses the cause.
| Weakness | Likely impact | Targeted improvement |
|---|---|---|
| Reaction temperature changed during each run | Rate constants may not represent the stated temperatures | Use a thermostatically controlled water bath and allow solutions to equilibrate before mixing |
| Visual endpoint depended on the observer | Increased random variation in measured reaction time | Use a colorimeter with a defined absorbance threshold and automated logging |
| Narrow concentration range | Limits confidence in the proposed relationship outside that range | Extend the range after pilot testing while keeping other conditions controlled |
Distinguish random uncertainty, which contributes to scatter, from systematic error, which shifts measurements consistently. Also consider limitations in scope, assumptions, sample representativeness and the validity of the selected model. The RevisionDojo evaluation guide offers a useful framework for turning weaknesses into evidence-based improvements.
Step 8: Edit and submit the final report
Organize the report around the four criteria rather than treating formatting as an afterthought. Headings such as Research Design, Data Analysis, Conclusion and Evaluation make the logic easy to follow, although the IB does not require one fixed template.
Before submission, check that:
- the report remains within 3,000 words
- the research question is identical throughout
- tables and figures are numbered and referenced in the text
- symbols, units and significant figures are consistent
- every borrowed idea, value and method is cited
- the bibliography contains every cited source
- essential reasoning is in the report rather than hidden in supplementary material
- the conclusion answers the question using analysed data
- each evaluation point links weakness, impact and improvement
Use the current Chemistry IA checklist for a final audit. You can also examine Chemistry IA exemplars and use the RevisionDojo Chemistry IA Grader for criterion-based feedback, but your teacher remains the authority on school deadlines, permitted feedback and submission procedures.
Academic integrity and responsible AI use
Your investigation, data processing and final writing must represent your own work. All external theory, methods, datasets, images and literature values must be acknowledged using a consistent referencing system.
The IB does not impose a general ban on AI tools, but it does not regard AI-generated content as the student's own. If AI-generated text, images or graphs are included or adapted, the tool must be acknowledged in the text and bibliography according to IB and school requirements. Jojo AI can support brainstorming, conceptual review or rubric-based reflection, but it must not replace your scientific decisions, fabricate data or write unacknowledged assessed content.
Conclusion
The strongest Chemistry scientific investigations are focused, reproducible and analytically rigorous. Begin with a question that can be answered using reliable quantitative evidence, design the method around that question, and plan uncertainty treatment before collecting data. Then use the conclusion to answer the question and the evaluation to explain how the method shaped the confidence and scope of that answer.
RevisionDojo can support the process through its current IA guide, exemplars, IA Feedback and Jojo AI. Use the Chemistry IA checklist and rubric-based grader near submission, while keeping every scientific decision and final sentence authentically your own.
Sources and referenced URLs
- Official IB Chemistry subject brief
- Official IB Diploma Programme Chemistry overview
- Official IB Chemistry curriculum updates
- Official IB academic integrity policy
- Official IB statement on artificial intelligence in assessment
- RevisionDojo IB Chemistry IA criteria guide
- RevisionDojo Chemistry IA ideas
- RevisionDojo science IA experiment design guide
- RevisionDojo Chemistry IA data analysis guide
- RevisionDojo IB Chemistry resources
- RevisionDojo IA evaluation guide
- RevisionDojo Chemistry IA checklist
- RevisionDojo Chemistry IA exemplars
- RevisionDojo Chemistry IA Grader