Dimensions

An overall test score answers one question: did the candidate do well? A dimension answers a more specific one: did they do well at what? Instead of a single number, results break down into named competencies — Verbal Reasoning, Attention to Detail, Communication — so a mid-range overall performer can turn out to be excellent at one thing and weak at another, information a single total can never show.

Why More Than One Kind of Dimension

Every dimension name has to come from somewhere consistent, and every question has to decide how its score reaches that dimension. Those are two different problems, which is why dimensions involve three pieces rather than one:

• Scoring Framework — defines the vocabulary once, centrally, so every question and every report uses the exact same dimension names instead of a different contributor inventing Verbal and another inventing verbal-reasoning.
• Associated Dimensions — the simple case: a question's one overall success rate is also credited to one or more dimensions. Most questions that need dimensions use this.
• Scored Dimensions — the case where one response has to be judged differently per competency — an essay strong on structure but weak on vocabulary can't share a single success rate, so each dimension gets scored independently instead.

Which One Do You Need?

Start with the Scoring Framework regardless — it's the prerequisite for the other two, since a dimension has to be defined before anything can be assigned to it. From there: if a question's existing success rate can simply be credited toward a competency as-is, use Associated Dimensions. If a single response genuinely deserves different scores on different competencies — usually true only for open-ended, reviewer- or AI-evaluated questions — use Scored Dimensions instead.

Define your dimension vocabulary in the Scoring Framework before assigning it anywhere — changing a dimension's name later means updating every question that already references the old one.

Good to know