OLSAT Level E common mistakes
OLSAT Level E · Understand the traps and how to avoid them.
Answering a passage-based comprehension item from general knowledge about the topic instead of from what the passage actually says - e.g. picking the option that is true in real life but never stated or implied in the text.
Familiar subject matter triggers fast, automatic associations; the brain 'fills in' plausible details, and the student recognises a true statement and stops reading rather than verifying it against the text.
What to do instead
Force a text-location step before marking any option: point to the sentence (or the exact words) in the passage that supports the choice. If no sentence supports it, cross it out no matter how true it sounds. Practise with short passages where the correct answer is deliberately different from the real-world fact.
Skipping or misreading the qualifier in the question stem - treating 'Which is NOT supported?', 'All of the following EXCEPT', or 'the best' as if it said 'Which is supported?' - and then selecting a statement that is factually accurate about the passage.
Under time pressure the eye lands on the content words (passage topic) and glides past small capitalised or negative words; the correct-sounding option feels rewarding, so the student stops there.
What to do instead
Underline or circle the qualifier before reading the options, and restate the stem in positive form in the margin ('find the one that is FALSE'). After choosing, re-read the stem once to confirm the choice answers the question as written, not the question the student assumed.
Solving verbal analogies by matching topic or category rather than relationship - e.g. for 'kitten is to cat' choosing 'puppy is to dog' when the tested relationship is young/adult, but choosing 'cat is to animal' because the words feel related, or accepting an option where the relationship runs in the wrong direction.
Word association is a strong, fast habit from everyday reading; two words from the same semantic field feel 'matched', and the direction of the relationship (which item comes first) is easy to lose.
What to do instead
Write the relationship as a full sentence with a specific bridge, e.g. 'A kitten is a young cat.' Insert each option pair into the same sentence and read it aloud; reject any pair where the sentence becomes false or must be reworded, and test the pair in reverse order to check direction.
In verbal classification ('choose the word that does not belong'), accepting the first shared feature noticed between two or three words and eliminating on that basis, without a rule that four of the five options actually satisfy.
Pairwise similarity is salient and satisfying; once two words seem linked, the student assumes the group is defined and stops checking the remaining options against a single consistent rule.
What to do instead
State a candidate rule, then test it against all five options and count: exactly four must fit and exactly one must fail. If two or more options break the rule, discard the rule and search for a stricter or different one (word class, part/whole, size order, number of syllables, position in a sequence).
On pictorial matrix and classification items, checking the missing cell against only one direction - the row or the column - and choosing an option that completes the row while breaking the column pattern.
The eye scans left-to-right by habit, so the row rule is noticed first and feels sufficient; the vertical relationship requires a deliberate second pass that students rarely make unprompted.
What to do instead
Require two confirmations per answer: state the row rule aloud, then state the column rule aloud, and only mark an option that satisfies both. Physically trace the column with a finger or pencil before committing.
On pictorial series and pattern-continuation items, generalising a rule from the first two or three pictures and applying it to the whole series, without verifying the rule against every consecutive pair already shown.
Pattern-hunting is rewarding once a rule is found, and checking the remaining figures feels redundant; the mind prefers closure to verification.
What to do instead
After forming a hypothesis, step through the entire visible sequence one pair at a time saying 'this changes to this because…'. If any consecutive pair contradicts the rule, abandon the hypothesis and look for a two-step or alternating rule.
Confusing a reflection with a rotation - selecting the option that looks 'turned' but is actually a mirror image of the original figure, or vice versa.
Mirror images and rotations of asymmetric figures are visually similar at a glance, and students compare overall shape rather than tracking a specific feature's position.
What to do instead
Pick one asymmetric landmark on the figure (a notch, a dot, an off-centre tab) and track only its position and facing through the transformation. If the landmark's clockwise order relative to a second feature reverses, the figure has been flipped, not rotated.
In figural series where elements change in two independent ways (e.g. a shading rule and a positional rule), locking onto one changing feature and ignoring the other, then choosing an option that satisfies only the noticed rule.
Attention narrows onto the most visually obvious change (usually colour or size), and once one rule is confirmed the student treats the item as solved.
What to do instead
Before looking at options, list every attribute that varies - shape, size, shading, orientation, number, position - and write a one-line rule for each. The chosen option must satisfy all listed rules; any option failing even one is out.
Assuming a single constant difference or constant multiplier in a number series and extending it, when the series actually uses alternating operations, interleaved sequences, or growing differences.
Arithmetic and geometric progressions are the patterns students meet most often in school, so the first difference is taken as the rule; a rule that fits the first three terms can still fail later.
What to do instead
Always compute the full difference chain (first differences, then second differences if the first are not constant) and separately check the odd-positioned and even-positioned terms as two sub-series. Confirm the rule reproduces every given term before extending it.
In quantitative word problems, solving for a convenient intermediate quantity (a unit rate, a subtotal, a difference) and selecting that number as the answer, or dropping the requested unit/format when transferring the result to the answer choices.
Multi-step problems create a strong sense of completion at the first clean number produced, and the final question sentence is often read before the working rather than after it.
What to do instead
Underline what is being asked before starting, label every intermediate result with its unit and a name ('cost per box', not just 12), and after computing re-read the final question sentence and write a full-sentence answer with units; then match that sentence to an option.