01
Intent
Read the commitment
Decide whether 今天先这样，明天再看 closes the issue, rejects it, or pauses it without fixing the next step. The useful skill is not translating each word; it is separating what has been said from what has actually been promised.
Free · five-minute conversation screen
What breaks first when Chinese gets real?
Five decisions reveal whether your next gain is intent, clarification, listening repair, tone, or transfer. No signup to see the result, no trick questions, and no inflated language level.
Diagnostic
Question 1 of 5
A colleague says 今天先这样，明天再看. What is the safest reading?
Conversation skill, not a badge
What this Mandarin diagnostic actually measures.
A Chinese level test usually asks how much language you can recognise: vocabulary, grammar, characters, listening, or reading. A conversation can still fail after you know every word on the screen. You may read 今天先这样 correctly but mistake a temporary pause for a firm promise. You may understand a request but answer too bluntly, guess a missed detail, or freeze when the person changes the time. This free diagnostic looks for that narrower gap between knowing Chinese and keeping an exchange workable.
The screen contains five decisions, not five trick questions. Each one represents a job that appears in an ordinary conversation: infer what is actually settled, ask for a missing condition, repair partial understanding, disagree without ending cooperation, and move the same language to a changed situation. The result is a starting route. It is not an HSK or CEFR estimate, an official proficiency rating, or evidence that five multiple-choice answers can measure speaking fluency.
Five decisions
Where a real Chinese exchange can break.
02
Response
Clarify the condition
Choose a reply that checks whether a deadline is today or tomorrow while keeping the request alive. A strong answer narrows the missing information instead of refusing, blaming, or handing the decision back.
03
Listening repair
Repair one missed detail
When one detail disappears in a fast reply, decide what to do next. Asking for the missing name, number, or time is better evidence of conversational control than guessing or abandoning the whole exchange.
04
Tone
Disagree and preserve the thread
Identify the response that acknowledges another person's point before naming a concrete constraint. The test checks whether disagreement still gives the other person a usable reason and a path forward.
05
Transfer
Move the plan
Ask a friend to change plans without making the change sound like rejection. The important evidence is whether the opening creates room for an explanation and a new time, not whether it repeats a memorised line.
Read the result honestly
A route label, not a proficiency level.
80–100 · Confident starting point
The five decisions did not reveal an obvious first break. Start with changed conversation tasks, because a short screen cannot prove that the same choices will hold under natural speed or original output.
60 · Working starting point
One or two decisions need attention. Begin with the first missed skill, practise one repair, and then meet the same conversational job with different people, times, and wording.
0–40 · Foundation starting point
Several decisions are not stable yet. Start with a complete model exchange, notice the purpose of each move, and remove support before attempting a changed prompt.
Before you start
Questions this five-minute screen should answer.
Is this a Chinese level test?
No. It does not assign HSK, CEFR, ACTFL, or course-credit levels. Those claims require broader evidence across language, tasks, modes, and conditions. This screen only routes five conversation decisions inside ChineseNuance.
Does it test speaking or listening?
It does not record speech or play an audio test. One item checks whether you choose a useful repair after missing a spoken detail. Actual listening and speaking evidence must come from later tasks with audio, original output, changed prompts, and delayed return.
Do I need an account?
No. All five questions and the result are free without signing in. An account is requested only if you choose to save the result and connect it to a continuing learning route.
Why not just test vocabulary?
Vocabulary recognition matters, but it cannot show whether you read a soft commitment safely, ask for a missing condition, or keep a disagreement workable. This diagnostic isolates those interaction decisions instead of turning a word count into a conversation claim.
What to do after the score.
Use the score as a routing hint, not a badge. Every item is worth twenty points because there are five decisions; the number is deliberately simple and does not imply psychometric precision. The labels confident, working, and foundation describe where this course starts, not your global identity as a Chinese learner.
The next trustworthy check is performance: understand a new exchange, produce your own response, repair the highest-impact problem, and try the same purpose again after the wording changes. ChineseNuance is AI-only. Software stores the evidence, selects the next repair, and schedules return practice; there is no hidden teacher, live interview, manual rating, or human certification behind this screen.
Method sources
- ACTFL Proficiency Guidelines 2024
Defines interpretive, interpersonal, and presentational modes and the evidence expected in proficiency judgments.
- ACTFL Assessments
Explains which formal assessments can provide official ACTFL ratings.
- Chinese Testing Service · HSK test structure
Shows the listening, reading, writing, item, duration, and score structure of the official HSK tests.