nepious.Research Sources and Practice Method
Inspect the official DET sources behind Nepious’s task-format guidance and understand where local practice differs from official scoring.
Official references, original teaching material
Nepious uses public official publications to understand task families, timings and broad assessment criteria. Our prompts, passages, answers and learning advice are original. We do not publish leaked or recalled live exam questions.
- Official test structure
Current families and frequencies; approximately one hour; adaptive length.
- 2026 scoring guide for teachers
Timing, rubric dimensions, score scale, worked official scoring examples.
- Official question guide (2025)
Task controls, replay limits, preparation and response timing.
- How items are scored
Objective comparisons and open-response grading.
- Interactive Writing
Five-minute answer followed by three-minute follow-up; initial response remains visible.
- Speaking Sample
Thirty-second preparation and up to three-minute response; official sample is video.
- Interactive Reading: Identify the Idea
Shared seven- or eight-minute timer and passage-wide interpretation.
- Updated Interactive Listening
Scenario completion, conversational turns and summary timing; updated July 2025.
- Test rules and requirements
Room, equipment, assistance and software restrictions.
- Official free practice test
Use this for a better external estimate and a comparison with your local practice.
- Technical manual, July 2025
Item-response calibration and proprietary production scorers.
- Official audio accent guidance
Current confirmation that audio prompts include General American and British English.
- Assessing Listening, June 2025
Listening stimulus design, synthetic/human voice input and approximate input pace. Use the newer help page for current accent coverage.
How conflicts are handled
Newer official task descriptions take priority when older publications differ. For example, the current help page identifies both American and British accents, while an older listening paper discusses American pronunciation. Interactive set counts can also vary across publications; the local full simulation uses two reading and two listening sets and clearly identifies its fixed format.
Learning feedback is not official grading
Answer keys, word-alignment credit and evidence-token overlap give practical feedback. Open responses use guided rubric review. Items are not calibrated against an official response sample, and the displayed practice band is not a measured prediction interval.