IA idea · Survey-based investigations
Who answers online surveys? Measuring non-response bias
Research question
When the same questions are sent as an optional online form and asked in person to a random sample, do the two sets of answers differ significantly, and what does that suggest about who chooses to respond?
Make it yours, then agree it with your teacher.
Next step: plan this IA free — the planner opens with this question filled in. For feedback on your own draft, the IA package (€39 once) gives you 5 AI feedback credits and every annotated exemplar in full.
Why it makes a good exploration
Non-response bias affects every real poll. Comparing a self-selected sample with a random one lets you measure it rather than just mention it.
The mathematics you'll need
- Random versus self-selected samples
- Comparing proportions and means between samples
- Chi-squared or t-tests with conditions
- Response rates and their effect
- Weighting to correct a sample (new)
The statistics, step by step
Worked with every number shown, with what examiners look for and the common mistakes: t-test: two-sample and paired. Then run the same steps on your own data in Analyse my data, or start from the statistics workflow.
Where the data comes from
Two collection methods for the same harmless questions; anonymity, consent and school approval.
- Desmos graphing calculator — Free graphing and regression (y₁ ~ ax₁ + b) — fit models to your data and show residuals.
Real data for this idea
- Primary source: Census at School New Zealand random sampler. Draw a random sample of past responses as a comparison for your own survey, not a substitute for it. Licence: Check the terms on the site.
- Also useful: Census at School USA (American Statistical Association): Census at School USA. The US random sampler gives the same kind of comparison sample. Licence: Check the terms on the site.
Census at School responses are voluntary, so they are not a random sample of all students: use them as a comparison, not as the population. Links checked October 2026; we link to the data and never copy it. More free data sets.
A possible outline
- Design the questions and two collection methods.
- Collect both samples.
- Compare answers statistically.
- Try weighting the online sample.
- Reflect on implications for school surveys.
A first step, worked
The heart of this exploration is one comparison: the answers of people who chose to reply against the answers of a random sample. Here it is on invented numbers.
- Online form: 120 of 400 students reply (a 30% response rate) and 75 of them, 62.5%, answer yes. In person: a random sample of 50, of whom 23, 46%, answer yes.
- Put the answers in a 2 × 2 table (method against yes/no). The expected counts are 69.2, 50.8, 28.8 and 21.2: all above 5, so the chi-squared test can be used.
- χ² ≈ 3.94 with 1 degree of freedom, p ≈ 0.047. At the 5% level the two methods give significantly different answers.
- Then the real question: who replied online, and why might they answer differently? That reflection is where the marks for Criterion D are.
The numbers are invented to show the method. Collect your own two samples, with consent and your school's approval.
Pitfalls that cost marks
- Differences in wording between methods.
- Small in-person sample.
- Questions people feel pressured to answer in person.
Showing personal engagement
- Use a real school question.
- Predict which group will be more positive.
- Advise the school on its next survey.
See Criterion C: personal engagement for what examiners look for.
Which course is it for?
| Course | Fit | Maths to lean on |
|---|---|---|
| AA SL | Not a natural fit | The core technique sits in the AI course or at HL; an AA SL student could use it only as clearly explained new mathematics. |
| AA HL | Not a natural fit | The mathematics is mainly from the AI course; at AA HL the exploration would need an AA-level approach (calculus, proof or probability theory) to reach the top of Criterion E. |
| AI SL | Good fit | Random versus self-selected samples; Comparing proportions and means between samples |
| AI HL | Good fit | Random versus self-selected samples; Comparing proportions and means between samples |
Level: Solid. Needs some independent work beyond class examples. See how the IA differs between AA and AI, SL and HL.
How this idea reaches the top bands
Reflection (D)
For this idea, start with the weak points: differences in wording between methods; and small in-person sample. Say how each one affects your answer.
Use of mathematics (E)
SL: random versus self-selected samples; comparing proportions and means between samples; chi-squared or t-tests with conditions; response rates and their effect; weighting to correct a sample (new), each explained in your own words and carried out correctly.
How surveys explorations reach the top bands in general: Survey-based investigations IA ideas. Guides: A · B · C · D · E · self-grade sheet.
Taking it further
Estimate how large non-response bias would need to be to reverse a published school-survey result.
Extending it for HL
Add an HL technique the question genuinely needs: what an HL surveys exploration adds.
See a complete IA, marked
Do students who sleep less react more slowly? (AI SL) is a complete surveys exploration, marked on every criterion.
Frequently asked questions
Which IB Maths course is this IA idea for?
It works for AI SL and AI HL. We rate it Solid: it needs some independent work beyond class examples.
What mathematics does this IA use?
Random versus self-selected samples, comparing proportions and means between samples, chi-squared or t-tests with conditions, response rates and their effect and weighting to correct a sample (new).
Where can I get data for this IA?
Two collection methods for the same harmless questions; anonymity, consent and school approval. Checked sources and tools: Census at School New Zealand, Census at School USA (American Statistical Association) and Desmos graphing calculator.
Turn this idea into your IA
The planner opens with “Who answers online surveys? Measuring non-response bias” filled in: the question, the maths and the data sources.
Your plan is kept in this browser. A free account keeps it with your drafts on any device: Join free · Sign in
Also: free IA self-check · A–E checklist (PDF) · have a draft? AI feedback or a review with Pete (€100)
Writing it up: how to write a Maths IA, step by step · structure and format · mistakes that cost marks · IA statistics, step by step
Similar ideas
- How big should a school poll be? Sampling from a known populationAI SLAI HLAA SLSolid
- Can you average a 1–5 rating? Analysing Likert-scale survey answersAI SLAI HLSolid
- Does the wording change the answer? A split-sample survey experimentAI SLAI HLAA SLSolid
- Do we know how long we spend on our phones? Estimates versus screen-time dataAI SLAA SLAI HLAccessible
All surveys ideas · AI SL ideas · AI HL ideas · All 239 IA ideas