<\/p>\n
Cross-cutting IA mistakes (the ones every student should fix)<\/h2>\n1. Starting without a tight question or brief<\/h3>\n
A vague research question or an overly broad brief is the fastest route to a meandering IA. Whether it\u2019s a science experiment, a math exploration or an economics commentary, unclear scope makes everything else harder: you waste time collecting irrelevant data, you struggle to analyze, and your evaluation becomes shallow.<\/p>\n
2. Ignoring the assessment criteria<\/h3>\n
There\u2019s a difference between doing interesting work and doing work that hits the mark scheme. Students often focus on what feels exciting rather than what the criteria reward. Learn the wording of the criteria\u2014\u2018personal engagement,\u2019 \u2018methodology,\u2019 \u2018analysis,\u2019 \u2018evaluation\u2019\u2014and map each section of your IA to specific criterion points.<\/p>\n
3. Poor planning and time management<\/h3>\n
Procrastination turns careful reflection into rushed appendices, and rushed work shows. IAs are iterative: pilot, collect, reflect, revise. Leaving things until the last minute kills the feedback loop that turns an okay IA into an excellent one.<\/p>\n
4. Weak or superficial analysis<\/h3>\n
Many students present data but stop at description. What lifts an IA is deep, subject-appropriate analysis: statistical tests and error discussion for sciences, formal mathematical reasoning for math, historiographical evaluation for history, and so on. Don\u2019t just state trends\u2014explain why they matter, what limits them, and what they reveal about your question.<\/p>\n
5. Bad or missing citations and academic honesty slip-ups<\/h3>\n
Plagiarism\u2014intentional or accidental\u2014can nullify an IA. Always cite sources, label primary and secondary data, and keep a clear log of contributions for group work. Use consistent referencing and include raw data in appendices.<\/p>\n
Sciences (Biology, Chemistry, Physics, ESS): experimental precision wins<\/h2>\nBiggest mistakes<\/h3>\n\n- Weak operational definitions: Students use fuzzy terms like \u201cincrease\u201d without measurable units or clear independent\/dependent variables.<\/li>\n
- Insufficient trials or poor controls: Too few repeats, or forgetting to control key variables, which weakens claims.<\/li>\n
- Overlooking uncertainty and error analysis: Reporting numbers without discussing measurement uncertainty or systematic error.<\/li>\n<\/ul>\n
How to fix them<\/h3>\n\n- Write an operational research question: e.g., \u201cHow does changing X by 10% affect Y measured in [unit] at [temperature]?\u201d. The clearer the procedure, the easier the analysis.<\/li>\n
- Design a pilot test early: A short pilot reveals scale issues, measurement sensitivity and outliers before you collect full data.<\/li>\n
- Include a dedicated error analysis section: simple propagation of uncertainty, repeatability, and discussion of systematic biases go a long way.<\/li>\n<\/ul>\n
Example: Instead of \u201cHow does light affect plant growth?\u201d, a stronger question is \u201cHow does the daily light duration (4 h, 8 h, 12 h) affect the average biomass (g) of Arabidopsis seedlings over four weeks under constant temperature?\u201d. This frames variables, units and timeframe explicitly.<\/p>\n
Mathematics: the exploration is not a homework question<\/h2>\nCommon pitfalls<\/h3>\n\n- Choosing a topic with no depth: If your entire IA is a standard classroom problem, it will lack exploration and originality.<\/li>\n
- Insufficient mathematical justification: Steps may be shown but logical justification and linkage to the research question can be absent.<\/li>\n
- Poor communication: Using dense notation without explanation or failing to interpret results in plain language.<\/li>\n<\/ul>\n
Practical fixes<\/h3>\n\n- Pick a focused question that invites genuine mathematical inquiry: modelling, generalizing a pattern, or investigating limits and rates of change.<\/li>\n
- Show the mathematics clearly and interpret it\u2014after every derivation, write one sentence about what the result means in the context of your question.<\/li>\n
- Use technology wisely: include graphs, spreadsheets, or CAS outputs but explain what they add and how they influence your conclusion.<\/li>\n<\/ul>\n
Economics, Business and Social Sciences<\/h2>\nMost frequent missteps<\/h3>\n\n- Lack of real data or over-reliance on secondary summaries without critique.<\/li>\n
- Poor linkage between theory and evidence\u2014students describe a model but don\u2019t use evidence to support or challenge it.<\/li>\n
- Ignoring evaluation: what are the assumptions and limits of your analysis?<\/li>\n<\/ul>\n
How to strengthen an IA in these subjects<\/h3>\n\n- Start with a crisp real-world context: a specific market, policy or firm and a clear research question.<\/li>\n
- Use appropriate diagrams and label them\u2014supply brief explanations that tie them to your evidence.<\/li>\n
- Assess assumptions: ask what would change if a core assumption fails and write that into your evaluation.<\/li>\n<\/ul>\n
Humanities and Languages (History, Language A, Language B)<\/h2>\nWhere students stumble<\/h3>\n\n- Summary instead of analysis: especially common in language commentaries and certain history tasks.<\/li>\n
- Poorly framed evidence: quotes without explanation, or sources used without situating perspective\/context.<\/li>\n
- Weak engagement with rubric descriptors like \u201cassessment of significance\u201d or \u201ccritical evaluation\u201d.<\/li>\n<\/ul>\n
Better habits to adopt<\/h3>\n\n- Always follow a paragraph structure: claim, evidence, analysis, link back to the question.<\/li>\n
- Situate each piece of evidence: who said it, when, why does their perspective matter?<\/li>\n
- For language tasks, balance style and function\u2014analyse language features and discuss effect, not just presence.<\/li>\n<\/ul>\n
Arts and Performance (Visual Arts, Music, Theatre)<\/h2>\nCommon IA weaknesses<\/h3>\n\n- Surface-level process journals: images without reflective commentary or deliberate technique development.<\/li>\n
- Missing documentation of artistic intent and how it evolved through experimentation.<\/li>\n
- Inconsistent linking between practice and critical\/contextual research.<\/li>\n<\/ul>\n
How to raise your arts IA<\/h3>\n\n- Keep a process log that records decisions, alternatives tried, failures and next steps\u2014these notes are evidence of personal engagement.<\/li>\n
- Include high-quality images or audio extracts (labelled and referenced in appendices), and explain what each piece taught you.<\/li>\n
- Tie your practical work to critical sources and show how ideas translated into method.<\/li>\n<\/ul>\n
Computer Science and Design IAs<\/h2>\nTypical problems<\/h3>\n\n- Scope creep: projects become too big to complete well and end up unfinished or poorly tested.<\/li>\n
- Insufficient testing and documentation: code that \u201cworks\u201d for the student but lacks clear requirements, test cases, or user documentation.<\/li>\n
- Ignoring ethical or security concerns in the evaluation.<\/li>\n<\/ul>\n
Fixes that actually work<\/h3>\n\n- Define minimal viable product (MVP) and advanced features; deliver the MVP well before adding extras.<\/li>\n
- Write user stories, include test cases and show sample inputs\/outputs in appendices.<\/li>\n
- Reflect on privacy, fairness, and maintainability as part of your evaluation.<\/li>\n<\/ul>\n
At-a-glance: Common mistakes and quick fixes<\/h2>\n\n\n\nSubject<\/th>\n Biggest IA Mistake<\/th>\n Fast Fix<\/th>\n<\/tr>\n<\/thead>\n \n\nSciences<\/td>\n Vague variables and no error analysis<\/td>\n Refine question with units; add uncertainty discussion<\/td>\n<\/tr>\n \nMathematics<\/td>\n Shallow exploration<\/td>\n Deepen with generalization or proof; interpret results<\/td>\n<\/tr>\n \nEconomics<\/td>\n Theory not linked to evidence<\/td>\n Use real data and explain deviations from the model<\/td>\n<\/tr>\n \nHistory \/ Languages<\/td>\n Summary instead of analysis<\/td>\n Adopt claim-evidence-analysis structure per paragraph<\/td>\n<\/tr>\n \nArts<\/td>\n Poor process documentation<\/td>\n Keep dated process notes and explain choices<\/td>\n<\/tr>\n \nComputer Science \/ Design<\/td>\n Unclear scope and testing<\/td>\n Define MVP, supply test logs and user manual<\/td>\n<\/tr>\n<\/tbody>\n<\/table><\/div>\nPractical IA workflow you can apply this week<\/h2>\n
This is a compact, repeatable workflow that keeps you honest and aligned with assessment criteria.<\/p>\n
Step 1 \u2014 Define and narrow (Days 1\u20133)<\/h3>\n\n- Draft a one-sentence research question or brief that names variables, units, subject and scope.<\/li>\n
- Spell out exactly what success looks like for the IA and map sections to criteria.<\/li>\n<\/ul>\n
Step 2 \u2014 Pilot and plan (Days 4\u20138)<\/h3>\n\n- Run a small-scale pilot to check feasibility and refine methods; note time and material needs.<\/li>\n
- Create a simple Gantt or checklist of tasks with deadlines and feedback checkpoints.<\/li>\n<\/ul>\n
Step 3 \u2014 Collect and document (Days 9\u201320)<\/h3>\n
- \n
- Weak operational definitions: Students use fuzzy terms like \u201cincrease\u201d without measurable units or clear independent\/dependent variables.<\/li>\n
- Insufficient trials or poor controls: Too few repeats, or forgetting to control key variables, which weakens claims.<\/li>\n
- Overlooking uncertainty and error analysis: Reporting numbers without discussing measurement uncertainty or systematic error.<\/li>\n<\/ul>\n
How to fix them<\/h3>\n
- \n
- Write an operational research question: e.g., \u201cHow does changing X by 10% affect Y measured in [unit] at [temperature]?\u201d. The clearer the procedure, the easier the analysis.<\/li>\n
- Design a pilot test early: A short pilot reveals scale issues, measurement sensitivity and outliers before you collect full data.<\/li>\n
- Include a dedicated error analysis section: simple propagation of uncertainty, repeatability, and discussion of systematic biases go a long way.<\/li>\n<\/ul>\n
Example: Instead of \u201cHow does light affect plant growth?\u201d, a stronger question is \u201cHow does the daily light duration (4 h, 8 h, 12 h) affect the average biomass (g) of Arabidopsis seedlings over four weeks under constant temperature?\u201d. This frames variables, units and timeframe explicitly.<\/p>\n
Mathematics: the exploration is not a homework question<\/h2>\n
Common pitfalls<\/h3>\n
- \n
- Choosing a topic with no depth: If your entire IA is a standard classroom problem, it will lack exploration and originality.<\/li>\n
- Insufficient mathematical justification: Steps may be shown but logical justification and linkage to the research question can be absent.<\/li>\n
- Poor communication: Using dense notation without explanation or failing to interpret results in plain language.<\/li>\n<\/ul>\n
Practical fixes<\/h3>\n
- \n
- Pick a focused question that invites genuine mathematical inquiry: modelling, generalizing a pattern, or investigating limits and rates of change.<\/li>\n
- Show the mathematics clearly and interpret it\u2014after every derivation, write one sentence about what the result means in the context of your question.<\/li>\n
- Use technology wisely: include graphs, spreadsheets, or CAS outputs but explain what they add and how they influence your conclusion.<\/li>\n<\/ul>\n
Economics, Business and Social Sciences<\/h2>\n
Most frequent missteps<\/h3>\n
- \n
- Lack of real data or over-reliance on secondary summaries without critique.<\/li>\n
- Poor linkage between theory and evidence\u2014students describe a model but don\u2019t use evidence to support or challenge it.<\/li>\n
- Ignoring evaluation: what are the assumptions and limits of your analysis?<\/li>\n<\/ul>\n
How to strengthen an IA in these subjects<\/h3>\n
- \n
- Start with a crisp real-world context: a specific market, policy or firm and a clear research question.<\/li>\n
- Use appropriate diagrams and label them\u2014supply brief explanations that tie them to your evidence.<\/li>\n
- Assess assumptions: ask what would change if a core assumption fails and write that into your evaluation.<\/li>\n<\/ul>\n
Humanities and Languages (History, Language A, Language B)<\/h2>\n
Where students stumble<\/h3>\n
- \n
- Summary instead of analysis: especially common in language commentaries and certain history tasks.<\/li>\n
- Poorly framed evidence: quotes without explanation, or sources used without situating perspective\/context.<\/li>\n
- Weak engagement with rubric descriptors like \u201cassessment of significance\u201d or \u201ccritical evaluation\u201d.<\/li>\n<\/ul>\n
Better habits to adopt<\/h3>\n
- \n
- Always follow a paragraph structure: claim, evidence, analysis, link back to the question.<\/li>\n
- Situate each piece of evidence: who said it, when, why does their perspective matter?<\/li>\n
- For language tasks, balance style and function\u2014analyse language features and discuss effect, not just presence.<\/li>\n<\/ul>\n
Arts and Performance (Visual Arts, Music, Theatre)<\/h2>\n
Common IA weaknesses<\/h3>\n
- \n
- Surface-level process journals: images without reflective commentary or deliberate technique development.<\/li>\n
- Missing documentation of artistic intent and how it evolved through experimentation.<\/li>\n
- Inconsistent linking between practice and critical\/contextual research.<\/li>\n<\/ul>\n
How to raise your arts IA<\/h3>\n
- \n
- Keep a process log that records decisions, alternatives tried, failures and next steps\u2014these notes are evidence of personal engagement.<\/li>\n
- Include high-quality images or audio extracts (labelled and referenced in appendices), and explain what each piece taught you.<\/li>\n
- Tie your practical work to critical sources and show how ideas translated into method.<\/li>\n<\/ul>\n
Computer Science and Design IAs<\/h2>\n
Typical problems<\/h3>\n
- \n
- Scope creep: projects become too big to complete well and end up unfinished or poorly tested.<\/li>\n
- Insufficient testing and documentation: code that \u201cworks\u201d for the student but lacks clear requirements, test cases, or user documentation.<\/li>\n
- Ignoring ethical or security concerns in the evaluation.<\/li>\n<\/ul>\n
Fixes that actually work<\/h3>\n
- \n
- Define minimal viable product (MVP) and advanced features; deliver the MVP well before adding extras.<\/li>\n
- Write user stories, include test cases and show sample inputs\/outputs in appendices.<\/li>\n
- Reflect on privacy, fairness, and maintainability as part of your evaluation.<\/li>\n<\/ul>\n
At-a-glance: Common mistakes and quick fixes<\/h2>\n
\n\n
\n \nSubject<\/th>\n Biggest IA Mistake<\/th>\n Fast Fix<\/th>\n<\/tr>\n<\/thead>\n \n Sciences<\/td>\n Vague variables and no error analysis<\/td>\n Refine question with units; add uncertainty discussion<\/td>\n<\/tr>\n \n Mathematics<\/td>\n Shallow exploration<\/td>\n Deepen with generalization or proof; interpret results<\/td>\n<\/tr>\n \n Economics<\/td>\n Theory not linked to evidence<\/td>\n Use real data and explain deviations from the model<\/td>\n<\/tr>\n \n History \/ Languages<\/td>\n Summary instead of analysis<\/td>\n Adopt claim-evidence-analysis structure per paragraph<\/td>\n<\/tr>\n \n Arts<\/td>\n Poor process documentation<\/td>\n Keep dated process notes and explain choices<\/td>\n<\/tr>\n \n Computer Science \/ Design<\/td>\n Unclear scope and testing<\/td>\n Define MVP, supply test logs and user manual<\/td>\n<\/tr>\n<\/tbody>\n<\/table><\/div>\n Practical IA workflow you can apply this week<\/h2>\n
This is a compact, repeatable workflow that keeps you honest and aligned with assessment criteria.<\/p>\n
Step 1 \u2014 Define and narrow (Days 1\u20133)<\/h3>\n
- \n
- Draft a one-sentence research question or brief that names variables, units, subject and scope.<\/li>\n
- Spell out exactly what success looks like for the IA and map sections to criteria.<\/li>\n<\/ul>\n
Step 2 \u2014 Pilot and plan (Days 4\u20138)<\/h3>\n
- \n
- Run a small-scale pilot to check feasibility and refine methods; note time and material needs.<\/li>\n
- Create a simple Gantt or checklist of tasks with deadlines and feedback checkpoints.<\/li>\n<\/ul>\n
Step 3 \u2014 Collect and document (Days 9\u201320)<\/h3>\n