Research transparency
IELTS Research and Data Methodology
How IELTS International validates practice data, protects learner privacy, and decides which research findings are reliable enough to publish.
Current publication status
We are not currently publishing platform outcome rates, causal improvement claims, or guaranteed score timelines. Earlier summaries did not expose enough information about sample selection, repeat-attempt matching, human validation, uncertainty, and analysis code to make those claims independently reproducible.
Those claims have been withdrawn. They will remain unpublished until the supporting dataset and methodology pass the standards below.
Publication standard
- A frozen, versioned analysis dataset with documented inclusion and exclusion rules.
- Clear separation between practice scores and official IELTS examination results.
- Paired learner-level analysis that does not count repeated essays as independent people.
- Human-examiner validation with sample size, agreement statistics, and limitations.
- Reproducible calculation code, confidence intervals, and correction of missing data.
- Privacy review ensuring all published results are anonymized and aggregated.
What current platform numbers mean
Public activity totals count platform records such as submitted practice essays and registered profiles. They are operational counts, not evidence that a particular study method causes score improvement. The live totals are available from the public statistics endpoint.
Responsible interpretation
AI-generated practice scores are estimates designed for learning feedback. They can differ from an official IELTS result and should not be represented as examiner-awarded scores. Progress depends on language background, starting level, practice conditions, feedback quality, and test-day performance; no fixed timeline can be guaranteed.
Read our editorial standards
See how sources, corrections, authorship, and high-stakes IELTS information are handled.
Editorial policyHow scoring works