The 2026 Clerical Officer Scoring Error: What Happened and What It Means
More than 12,500 Clerical Officer applicants were given incorrect scores and rankings in 2026. What went wrong, what publicjobs own review found, and what it means for candidates.
In April 2026, more than 12,500 people who had applied for the Clerical Officer competition were told that the scores and rankings they had been given were wrong.
publicjobs published an internal review of what happened. It is worth reading if you are sitting any competition, because it explains exactly which part of the process failed, and it is unusually candid about it.
What happened
Applicants received their scores and their position on the order of merit in the normal way. Those scores were incorrect.
The cause was in the scenario questions. Two of the six scenario questions had not updated as intended. When candidates’ responses were scored, the scoring key applied was the one written for the intended updated questions, not the questions candidates had actually answered.
So a set of answers was marked against a key for different questions. Everyone affected received a score and a ranking generated on that basis.
In late April, applicants were notified with an apology and an explanation, and revised scores and rankings were issued the following day.
What the review found
The internal review, reported in July 2026, attributed the failure to a combination of communication failures, system problems, and inadequate testing of the online application process before it went live.
The review’s own assessment was that the error seemed avoidable, and that it was likely to have had a negative effect on the reputation and credibility of publicjobs as a professional and trusted recruitment body.
That is publicjobs describing its own process, not a critic describing it.
You can read the coverage of the review at RTÉ.
What it actually meant for candidates
Two very different experiences, depending on which way the correction went.
For candidates who had been marked lower than they should have been, the revision was good news, and some moved up materially.
For candidates who had been marked higher than they had actually scored, the revised ranking was worse than the one they had been living with for weeks. Some had already made decisions on the strength of the first number.
This is the part worth sitting with. A ranking is not just information, it is something people plan around. They decide whether to accept other work, whether to keep applying elsewhere, whether to hold out for a particular location. Issuing a wrong ranking and then correcting it is not a neutral event, even when the correction is handled quickly.
What it means for you
Three practical takeaways, and one thing not to take away.
Your ranking is a data point, not a certainty. Keep applying to other competitions rather than reorganising your plans around a single number. That was sound advice before this happened and it is better advice now.
Read the messages you are sent, properly. The correction was communicated by email and through candidate accounts. Candidates who only skim notifications are the ones who find out late. Check the account itself rather than relying on your inbox, because notifications get filtered.
Errors do get corrected. The relevant precedent here is that when a marking problem has been identified, candidates have been contacted and scores amended. That is not a guarantee of anything, but it is the pattern.
And the thing not to take away: this is not a reason to assume that a score you disagree with is wrong. Candidates who are convinced a test was faulty are usually mistaken, and confirmed marking failures like this one are rare enough to be national news when they happen. If you did not get the mark you wanted, the far more likely explanation is the ordinary one.
The wider point about format
The reason this failure was possible is that the assessment is a specific instrument with specific mechanics: scenario questions, marked against a calibrated key, generating a ranking. It is not a general aptitude test and it is not the UK version.
That is also the reason candidates who prepare on generic material tend to underperform on it. The format has particular scoring behaviour, and if you have never seen how these questions are actually marked, you are working it out under time pressure on the day.
FAQ
How many people were affected by the Clerical Officer 2026 scoring error? More than 12,500 applicants were issued incorrect scores and rankings.
What caused it? Two of the six scenario questions had not updated as intended, and responses were scored against the key written for the intended questions rather than the ones candidates answered.
When were candidates told? In late April 2026, with an apology and an explanation. Revised scores and rankings followed the next day.
Did everyone’s ranking go up? No. Candidates who had been marked lower than they should have been moved up, and candidates who had been marked higher moved down.
Did publicjobs accept fault? Its internal review found the error seemed avoidable and was likely to damage its reputation and credibility, and attributed it to communication failures, system problems and inadequate testing.
Does this affect competitions I sit now? There is no indication that it does. The review’s value to a current candidate is what it says about how the assessment is built and marked.
Prepare for the actual instrument
The assessment is scored on mechanics most candidates never see before the day. The situational judgement test guide covers how the ranked-answer format is actually marked, and the publicjobs application process guide walks through the stages in order.
You can try the free taster and rank a real-format scenario yourself, with no card required.