Assessment summary

What this report is — and the decision it supports

Individual reports answer questions about candidates; the assessment summary answers questions about the assessment itself. Was it designed around the right competencies? Did the tasks resemble the actual work? Did scores spread enough to distinguish candidates, and did evaluators apply the rubric consistently? The summary presents all of this in stakeholder-friendly language, with the technical detail confined to an appendix.

The decision it supports is trust and continuation: a client deciding whether to keep working with this process, leadership deciding whether the method is sound enough to expand to other roles, or a program owner deciding what to adjust before the next round. It describes the process honestly — including where scores clustered or evaluators diverged — because a summary that only flatters the design is not decision support.

When to reach for it

For consultancies and agencies, the natural moment is the client review: the engagement wraps, and the client wants to understand what was measured and how the pool performed before acting on the shortlist. The summary is the document you present, and its method appendix answers the 'how do we know this is rigorous?' question before it is asked.

In-house, it belongs at the end of every hiring round and at the start of the next. Reviewing the score distribution and calibration notes shows whether tasks separated candidates well and whether evaluators need a calibration session — the two adjustments that most improve an assessment between rounds. It is also the right document when a new stakeholder inherits the program and needs to understand it quickly.

• Presenting results and method to a client at the end of an engagement.
• Reviewing a hiring round with leadership before expanding the program.
• Deciding what to tune — tasks, rubrics, or evaluator calibration — before the next cohort.
• Onboarding a new stakeholder who needs the whole picture in one read.

What's inside, section by section

The summary reads from design to results to method: first what the assessment set out to measure, then how the pool performed, then the detail that lets a careful reader verify both.

• Competency map and task sequence: the competencies from the role blueprint and the work-sample tasks that made each one observable, in the order candidates met them.
• Score distribution overview: how the candidate pool performed across competencies and sections — where candidates spread, where they clustered.
• Evaluator calibration notes: how consistently evaluators applied the rubric, and where reviews diverged and were reconciled.
• Method and rubric appendix: the rubrics, weights, and scoring approach in full, for readers who want to check the work.

How it's produced

The summary is generated from the assessment's own structure and records — the role blueprint, the task sequence, the rubrics, and the scores and notes evaluators entered during review. Nobody compiles it by hand, which means it reflects the assessment as it actually ran, not as anyone remembers it running.

The scoring behind the distribution view is the same transparent math used across SkillCort: evaluator scores average within a criterion, criteria combine by rubric weights, and tasks roll up by skill weight times section weight. The appendix states this method in plain language so a stakeholder can follow any number in the report back to its source.

How to read it well

Start with the competency map and ask the only question that matters at that level: are these the things the role actually needs? Everything downstream — tasks, scores, distributions — inherits its meaning from that mapping, and it is where stakeholder feedback is most valuable.

In the distribution view, resist reading a high-scoring pool as a strong pool or a low-scoring one as weak. Distributions describe the pool against this rubric; a tight cluster near the top may mean effective pre-screening, or a task that stopped discriminating. The calibration notes are the tiebreaker — consistent evaluators and a compressed spread point to the task; divergent evaluators point to the rubric. Neither is a verdict on the candidates.