Read the exact selected task before using a row. These descriptions are original, formative and task-specific. They are not QCAA A–E descriptors or numerical cut scores. Feedback targets one next scientific job; no total hides a difference between an arithmetic error, inference error and unobserved action. Accept a scientifically equivalent reasoning route and accurate calculator-supported working.
E / RE1 / RE2 · first-stage dimensions
Elicited component
Supported now for this task
One more check
Model next with different ordinary inputs
Q1 · prior height change and condition
Both Day 0-to-Day 3 increases correct, mm and three-day period stated; assigned changed condition correctly named.
Correct idea with one unit/time/condition or subtraction omission.
Calls final height growth, mixes start/end or labels a measured outcome as the changed variable.
Q2 · reasoned prediction
Ranking or qualified uncertainty uses the prior record, says expectation not result and a relevant uncertainty; equal prior means permit tie/uncertainty.
Prediction has support but uncertainty or before-outcome status is unclear.
Invents Day 6 measurement/alive count, demands certainty or claims a universal treatment effect.
Q3 · investigable paper plan
A precise question, changed condition, height/alive measures, at least two relevant controls, consistent schedule/reference and risk/permission/review point; result not observed.
A workable plan misses a named component or uses “keep fair” without specifics.
Changes multiple key features unknowingly, claims the proposed plan was run, or treats a written request as permission.
E / RE1 / RE2 · later-stage dimensions
Elicited component
Supported now
One more check
Model next
Q1 · new comparison
Correct six-day increases and relative Day 6 means, unit/period intact.
One accurate calculation but incomplete comparison or period.
Uses Day 3 as start while claiming Day 0 increase, or confuses mean with each individual.
Q2 · evidence review
Refers to captured first prediction and justifies actual keep/revise/qualified update using new record; does not rewrite history.
Accurate new answer but no link to first reasoning.
Claims the outcome was known first or requires an edit despite a justified keep.
Q3 · growth/survival boundary
Both alive counts use original four and Day 6; separates height increase from alive/health/future survival and retains mean sample rule.
Count accurate but a denominator/time/limit is missing.
Uses survivors-only denominator or assumes no growth proves death/height proves health.
Q4 · method and next enquiry
Specific unrecorded/control/reference problem, relevant further question and improved rule; supplied versus planned versus actual explicit.
Sensible question but vague method or action status.
Says record proves cause, invents controlled conditions or credits unperformed equipment/test action.
F / RF1 / RF2 · first-stage dimensions
Elicited component
Supported now
One more check
Model next
Q1 · two measures
Both eight-day height increases correct; counts out of original six/time and mean definition retained.
Correct measures with a unit/period/count omission.
Final height called increase, survivor denominator substituted or missing value invented.
Q2 · complete decision rule
Separate pass/fail for both requirements on both options; decision follows first reader priority and two exact details. Neither/both/tie handled honestly.
Correct choice but one requirement table/detail/priority unshown.
Drops a failed requirement, combines A/B into a fictional option, or chooses larger final height without full rule.
Q3 · knowledge, limit and next evidence
Explains how distinct growth/survival measures inform fictional proposed choice; relevant limit/question, no real permission/impact/test claim.
Useful next question but knowledge link or proposed status vague.
Calls a fictional committee decision an enacted local action or universal biological rule.
F / RF1 / RF2 · later-stage dimensions
Elicited component
Supported now
One more check
Model next
Q1 · new-cohort record
Both new increases and counts correct, original sample, mm and its own eight-day period clear.
New values right but cohort/time distinction absent.
Concatenates cohorts into sixteen days or reuses earlier outcomes as new.
Q2 · later complete decision
Both minima and later reader priority applied to new cohort; table and two supporting details intact.
Correct option but incomplete evidence/table/priority.
Ignores later priority or silently lowers a minimum.
Q3 · actual reasoning update
Captured first decision referred to; actual keep/revise justified by new source/reader and distinct-cohort explanation.
New decision right with a vague change reason.
Claims a revision not documented or overwrites first work with hindsight.
Q4 · purposeful later note
Useful conditional decision for stated later reader; distinct growth/survival reason, uncertainty and feasible reviewed next enquiry.
Useful wording missing one task component.
Gives a universal care command, asserted ecological impact or permission/measurement not supplied.
Ordinary H01/H02 source tasks · separate evidence
Day 6: learner identifies a named actual contribution in FACT1 or FACT3 and explains why collaboration or building on earlier work mattered to the specific scientific problem; dates/status and one source limit remain. Naming two people or an invented team alone leaves the H01 reasoning unelicited. Ordinary Day 8: learner explains source-supported problem → scientific knowledge → response/decision, identifies local unknowns and a reviewed next step. A slogan, personal preference or invented committee act alone does not establish authentic H02 investigation. These are selected source-based opportunities, not complete achievements or observed collaboration/impact.
Unelicited components and action distinctions
Fresh tasks leave H01 research/contributions, authentic H02 community action, full U01 breadth, other Science understanding descriptions, I02 actual conduct, I03 equipment generation/precision, field permissions and practical outcome evidence unelicited. Do not give a zero mark for an action not asked. Actual ruler use in an optional ordinary paper-model job can support the specific equipment-use observation; helper ruler use alone leaves learner operation unobserved. An exact child-directed instruction supports the child's method idea, not their motor operation. Reading supplied plant heights does not generate data. Actual own guide making/editing can support communication; a helper-authored model does not show the learner wrote or revised it. A spoken/signed/AAC answer may communicate scientific reasoning without pretending it is an enacted experiment or written-language sample.