Name what happened before deciding what to change.
Work through problem definitions, evidence, alternatives, implementation and review. Ten practice answers do not measure a student’s general ability.
A correct count describes ten selected tasks in this original bank. It is not a grade, population rank or verified curriculum achievement. Every count opens every explanation and all three complete guides.
Use the explanation to locate a step
The ten items in this group sample parts of a problem-solving process using fictional school and everyday situations. Your count says how many selected answers matched the stated task or model. It cannot tell whether a difference came from reading, attention, calculation or an unfamiliar term. Read the actual explanation before choosing a next exercise. There is no “weak student” or “gifted problem-solver” label to infer from this small practice record.
Describe an observation instead of a person
Write what happened, where or when it happened, and what count or record supports the description. “Three notices had the wrong room number” can be checked; “the editor is hopeless” adds a judgment without locating the error. A precise definition also stays separate from a proposed fix. Buying equipment or changing a schedule may become options later, but neither action by itself establishes what went wrong in the first place.
Keep explanations open while gathering evidence
List a few plausible explanations and what you would expect to observe under each. A record that fits one explanation may fit another too. Collecting two variables does not automatically identify a cause, and asking “why” repeatedly does not make the resulting story true. The bank sometimes supplies an explicit small model so a deduction is possible. Real situations may contain additional causes, measurement errors and constraints outside that model.
Change a different lever, not just a name
Two versions of “meet on Monday” are not two distinct alternatives. Moving the day and shortening the duration change different parts of a plan, so they can be compared separately. Before selecting, note mandatory limits such as available time or an agreed room. There is no need to generate an endless list: the point is to avoid treating the first attractive idea as the only possible response before any comparison has occurred.
Trace a fresh case without jumping to blame
Imagine a school display has four labels that disagree with its current catalogue. Start by noting which labels differ and what each says. Possible next checks include the catalogue version and the label-printing date. Those checks may separate an old-file explanation from a copying error, but no culprit is established yet. Only after the relevant evidence is checked should the group compare fixes, such as correcting individual labels or reprinting a verified set.
Make the trial observable
An action, an owner and a day make a trial plan easier to check than “we should improve things”. They still do not prove execution: the record must show that the action happened and the agreed observation was made. Approval is a planning event, while a completed trial is an execution event. Keep its outcome separate again, because a trial can run exactly as planned and still fail to meet its goal.
Reserve an opportunity to check
If a forty-minute plan allocates eight minutes to setup and seventeen to making, fifteen remain for checking or other stated tasks. That arithmetic shows available time only. It does not guarantee that every error will be noticed or that a grade will improve. If setup takes longer, revise the remaining-time calculation. Planning and review are useful topics to practise because their assumptions and records can be made explicit.
Build a small fictional case of your own
Write a short invented problem, two possible explanations and two distinct responses. State what evidence would help choose, then what you would record after a trial. Do not change a real school process merely because a quiz option sounded sensible, and keep real pupil records out of the exercise. Public guides contain methods rather than private results; a retake repeats known items and cannot by itself establish transfer to new problems.