NNAT3 Level C common mistakes
NNAT3 Level C · Understand the traps and how to avoid them.
On a Pattern Completion item, the candidate checks only the missing patch and ignores the surrounding frame, then picks the option that looks plausible in isolation.
Young candidates scan the answer options first and compare them to each other rather than to the incomplete picture, so the frame's colours, line directions, and spacing never enter the comparison.
What to do instead
Train a fixed order: look at the hole, then trace the lines entering the hole from all four sides, then eliminate options that break any traced line. Have the child point at the exact edge where each rejected option fails before moving on.
The candidate accepts a Pattern Completion option that matches the pattern's content but is rotated or mirrored relative to the frame.
Level C patterns often use repeating motifs where a flipped copy still looks familiar, and children match on overall resemblance instead of orientation of specific elements.
What to do instead
Practise with two identical motifs, one rotated, and ask which way a named detail (a stripe, a dot, a corner) points. Require the child to state the orientation of one detail in the chosen option before committing.
On a Reasoning by Analogy item, the candidate works only from A to B and never checks that the same change applied to C yields the chosen answer.
The relationship is usually obvious in one direction, so the child stops after finding any option that shares a feature with B rather than verifying the full A-to-B rule survives the C-to-D step.
What to do instead
Enforce a spoken two-sentence check: 'A changed to B by ___, so C changes to D by ___'. If the child cannot say both sentences with the same rule, the option is rejected.
The candidate applies a rule that involves only one attribute, such as colour, when the analogy also changes shape, count, or position.
Single-attribute matching is faster and feels sufficient, and distractors are deliberately built to reward it.
What to do instead
Before choosing, have the child list every attribute that differs between A and B: colour, size, number, orientation, fill, position. Then confirm the chosen D changes the same set and nothing extra.
On a Serial Reasoning item, the candidate identifies a repeating cycle but does not check whether the cycle length divides evenly into the position being asked for.
Children spot 'it goes red, blue, green' and extend by eye from the last visible frame instead of counting the step number, which fails when the sequence skips a frame or the cycle is longer than expected.
What to do instead
Teach counting the step index and using the cycle length: if the pattern repeats every three, the answer at step ten matches step one. Have the child write the step numbers under each frame before answering.
The candidate assumes a Serial Reasoning sequence alternates between two states when it actually accumulates, for example adding one shape each step or rotating by a growing amount.
Alternation is the first pattern children learn, so it is over-applied; accumulating sequences look irregular at the first two frames and get misread.
What to do instead
Compare consecutive frames and ask 'is the change the same each time, or is the change itself getting bigger?' Practise mixed sets where the child must label each sequence as repeat, grow, or rotate before solving.
On a Spatial Visualization item, the candidate mentally rotates the whole figure as a rigid block when the item requires folding, unfolding, or reflecting only part of the figure.
Whole-object rotation is the easiest mental operation and children default to it, so they miss that a fold creates a mirror relationship across the fold line rather than a rotation.
What to do instead
Have the child physically fold paper and punch or colour it, then open it to see the mirrored result. Bridge to the item by naming the fold line and checking that each mark has a mirror twin at equal distance on the other side.
The candidate loses track of left and right, or of the viewer's perspective versus the figure's own perspective, when a Spatial Visualization item involves a turn.
Egocentric left/right is not fully consolidated at this age, and items that rotate a figure invite the child to keep using their own body as the reference frame.
What to do instead
Anchor with a concrete cue: put a dot on the figure's 'nose' or marked corner and track where that marker ends up after the turn. Ask 'where does the marked corner point now?' rather than 'which way is left?'
The candidate answers in the order the options are printed, stopping at the first option that seems reasonable instead of eliminating all wrong options first.
Under time pressure, satisficing feels efficient, and well-designed distractors are placed early to capture exactly this behaviour.
What to do instead
Make elimination the default: the child must cross out at least two options with a stated reason before selecting. Time practice sets generously at first so the habit forms before speed is added.
The candidate rushes the last items or guesses on unfamiliar item types because the test is untimed and pacing was never practised.
NNAT3 is administered without a strict per-item clock, so children either dawdle and lose focus or panic when a novel item type appears, and fatigue near the end produces careless errors across all four domains.
What to do instead
Run short timed practice blocks of mixed item types to build stamina and familiarity with all four formats, and rehearse a reset routine: on an unfamiliar item, name the domain, state the rule aloud, then eliminate. Keep sessions short and frequent rather than long and rare.