GLA 10 common mistakes
GLA 10 · Understand the traps and how to avoid them.
Reading a scenario item as a question about values or ethics and picking the option that sounds most principled, rather than the option the item's competency definition treats as effective practice.
Leadership-style items often reuse moral vocabulary (integrity, inclusion, transparency), so candidates pattern-match to the most virtuous-sounding phrase instead of the behavioural criterion being sampled.
What to do instead
Before reading the options, state in one line what the item is measuring (e.g. delegation, stakeholder management, coaching). Then eliminate any option that sounds admirable but does not actually address that behaviour in the situation described.
Defaulting to the same rating (usually the top or second-highest point) on most self-report or frequency items, producing a flat, undifferentiated profile.
Extremes feel like confidence and differentiation requires effort; candidates also drift toward the self-image they want rather than a remembered pattern of behaviour.
What to do instead
Force a distribution: before starting, commit to using the full scale. For each item, recall one specific incident from the last six months that supports the rating; if no incident comes to mind, the honest rating is lower than instinct suggests.
Overlooking the operative qualifier in the stem, such as 'first action', 'most likely', 'least effective', or 'in the long term', and choosing the generally best-sounding response.
Under time pressure the eye jumps from the scenario to the options, and the qualifier is the shortest, least salient part of the stem.
What to do instead
Physically mark the qualifier before reading any option, and restate the stem to yourself as a constraint sentence ('the FIRST thing this person should do is...'). Re-check every shortlisted option against that exact constraint, not against generic good practice.
Answering from the perspective of the current job level rather than the level the assessment targets, so responses describe what the candidate does now rather than what the target role requires.
People answer from lived habit; the current role is concrete and recent, while the target level is abstract.
What to do instead
Write the target level's three or four defining shifts (for example, from doing to directing, from team to function) at the top of the scratch sheet, and run each response through the question: 'is this the behaviour of someone already operating at that level?'
Treating forced-choice pairs by picking the statement that feels more personally flattering, or splitting between both statements when the format does not allow it.
Forced-choice formats feel unfair because neither statement is fully true, and the instinct is to correct the test rather than choose the more descriptive option.
What to do instead
Accept that both statements may be partly true and select the one that is more often true of your actual behaviour at work, even if the other is more attractive. Never spend time trying to infer which trait the pair is measuring; consistency matters more than decoding.
Burying the recommendation in the middle or end of a written response after background, context, and restated theory.
Academic writing habits reward building an argument before the conclusion, and length is mistaken for depth under word pressure.
What to do instead
Use a fixed three-part structure: recommendation first in one sentence, then the two or three pieces of case evidence that justify it, then the main risk and mitigation. Cut any sentence that restates the scenario.
Memorising framework names, stage models, and vocabulary instead of rehearsing how each would change an action in a case.
Prior exam experience rewards recall, and labelled models are the most visible, most easily revised content in any preparation material.
What to do instead
For each framework, write one sentence of the form 'a leader using this would do X rather than Y in situation Z'. Practise only in that applied form; discard any study note that cannot be converted into a described behaviour.
Revisiting and changing earlier answers after moving on, or changing answers to make the profile look more consistent.
Later items create doubt, and candidates assume a polished, coherent-looking profile scores higher than an honest uneven one.
What to do instead
Adopt a single-pass rule: answer once, flag only items left genuinely blank, and do not reopen answered items. If an item feels inconsistent with an earlier one, that is legitimate data, not an error to be smoothed out.
Sinking large blocks of time into one difficult scenario and leaving later items, including the shorter diagnostic sections, rushed or blank.
Perfectionism plus the assumption that every item deserves equal deliberation; hard items feel disproportionately important because they consume attention.
What to do instead
Set a hard per-item time cap before starting (total time divided by item count) and a visible clock. When the cap hits, select the best current option, mark it, and move on; return only if time remains at the end.
Completing the assessment in fragmented sessions, late at night, on a phone, or while multitasking, and then reading the feedback report as a verdict on ability rather than a pattern to act on.
These instruments are often taken unsupervised, so scheduling is left to convenience; and feedback is read once, emotionally, without linking it back to specific observed behaviour.
What to do instead
Book one uninterrupted block in a quiet setting on a full-size screen, at a time of day you are normally alert. When the report arrives, convert the two lowest themes into one observable behaviour change each and a named person who will notice it within 30 days.