administration

TOEFL iBT 2026: Original fixed practice form 1 Source-only review draft. This is ordinary English iBT practice for the January 21,2026 redesign, excluding the Australia version. It is not an ETS product or a calibrated adaptive simulation. Use the learner questions from the canonical host's question view and a separate response sheet. test01.tex contains all four host fields, including answers; do not expose its answer view while assessing. Keep facilitator-scripts.txt and scoring files hidden. The questions-only reviewer file includes hidden aural scripts so independent reviewers can solve all items before seeing keys; do not hand that combined file to a learner. Reading: Questions 1–50,30 minutes. Use 21 minutes for 1–30 and 9 for 31–50. Move freely within the current reading module, close it when its time ends, and do not return afterward. The fixed module split is editorial, with no score-based route. Thirty letter gaps, ten daily-life questions, and ten academic-passage questions contribute 50 teachingpoints. Each underscore denotes one missing letter. Write missing letters only. Listening: Questions 51–97,29 minutes. Use 12 minutes for 51–72 and 17 for 73–97. Move forward only. A qualified facilitator reads each original script once in clear, intelligible English, keeping it out of sight and using distinct voices or role cues for the two speakers in conversations. Do not speak labels such as "Student A". Preview the hidden schedule and practice the delivery before beginning. Do not paraphrase, coach, replay, or slow a prompt to expose an answer. Learners may take notes on permitted paper. After a grouped stimulus, learners answer its questions in order without hearing it again. The exchange questions offer visible possible replies; the question utterance itself is spoken only. At a provisional 150 words/minute, all original listening stimulus content takes approximately 10 minutes, leaving about 19 minutes for response selection and transitions. This is a text-derived estimate, not measured audio. The first module has 22 positions and the second 25; a trial administrator must verify both module schedules against the exact 12/17 limits. Use the time remaining in a module for its responses, not transcript reading. If administration cannot fit, log the departure and do not call the result a standard timed listening practice. See timing-and-media.json. No recorded assets or listening quality checks have been completed. This draft supports a live facilitator adaptation only; recorded/native-audio readiness remains unmet. Writing: Questions 98–109,23 minutes. Close the sentence-building task after 6 minutes, the email after 7, and the discussion after 10. For 98–107, use selected chunks once with their internal order unchanged, fixed printed boundaries, and capitalized first words where needed. Some inventories include an unused chunk. Supply a complete sentence or a clear ordered chunk sequence. No alternative MC answers are invented. Use a separate sheet or a plain text editor for 108–109, with spelling/grammar aids, dictionary, prediction, and external assistance disabled. A100 word discussion is guidance for development, not an automatic scoring threshold. No email minimum is imposed. Speaking: Questions 110–120,approximately 8 minutes including prompt/setup intervals. No preparation. Read each repeat prompt once, then immediately allow 8,8,10,10,10,12,12 seconds respectively. Stop at the limit and proceed. The four interview questions are spoken once, with 45 seconds after each. One response attempt. Prompts and targets remain hidden until afterward. Record the actual responses where available and use the current task-specific practice rubrics; without hearing actual speech, do not rate intelligibility, pronunciation, rhythm, pace, or fluency from a transcript. The printed examples are illustrations and have no acoustic rating. No scheduled break. Section work totals 90 minutes excluding directions, movement, equipment checks, and prompt setup; allow about 2 hours. Use approved-style note paper and an ordinary response sheet. Do not use dictionaries, outside material, grammar assistance, or prior access to solutions. This exercise does not reproduce secure software, adaptive routing, live pretest identities, or official media. All 107 machine positions receive practice feedback, unlike a live form's undisclosed scored/pretest allocation. The 13 constructed responses receive separate 0–5 feedback. Maximum teaching points:R50,L47,W20,S55,total 172. Record section results and task needs separately. Never convert these totals to ETS 1–6,0–120,CEFR,pass/fail,admissions predictions,or a native-test percentile. Reading a hidden listening transcript changes the skill being assessed and must be labelled a reading exercise.