Point Scoring

This is the primary scoring pipeline — the one that produces the overall test score every candidate sees. It runs on each question's positive and negative multiplier (also known as weight) and rolls up through the same structure a test is built from: questions sit on pages, pages sit in sections, sections make up the test.

How a Question's Points Are Calculated

Every answer is first turned into a success rate, somewhere between −100% and +100%. It's not just right or wrong — a partially correct answer might land at 50%, for instance. That success rate is then multiplied by the question's weight to produce points, but which weight depends entirely on the sign of the success rate:

• Success rate above 0% — multiplied by the positive multiplier. Points = success rate × positive multiplier.
• Success rate below 0% — multiplied by the negative multiplier instead. Points = success rate × negative multiplier.
• Success rate exactly 0% — the points are always 0, full stop. Neither multiplier is applied — there's nothing to multiply.

In other words, a wrong answer only loses points if the answer itself evaluates to a negative success rate — a blank or neutral 0% answer never triggers the negative multiplier, it simply scores 0. See Positive Multiplier and Negative Multiplier for exactly how each one behaves, with worked numbers.

The Roll-Up Formula

• Page score = sum of the points earned by every question on that page
• Section score = sum of the scores of every page in that section
• Test score = sum of the scores of every section in the test

Points Are Absolute, Scores Are Shown as a Percentage

The roll-up above produces raw points, which on their own don't say much — 6.5 points means nothing without knowing what the maximum possible was. In practice, every level is also expressed as a percentage: points earned ÷ maximum possible points at that level, for the page, the section, and the test. The maximum possible is what a candidate would score by answering every question fully correctly — the sum of every question's positive multiplier at that level. A wrong answer's negative multiplier never adds to that maximum; it only ever subtracts from what was actually earned.

Worked Example

A test has 2 sections. Section 1 has 1 page with 3 questions; Section 2 has 2 pages, the first with 2 questions and the second with 3.

Q7 and Q8 both land on a 0% success rate, so both score 0 points — 0% always means 0 points, regardless of the multiplier.

Rolled up: Section 1 / Page 1 scores 1 + 1 − 0.5 = 1.5, so Section 1 scores 1.5 (its only page). Section 2 / Page 1 scores 1 + 3 = 4; Section 2 / Page 2 scores 1 + 0 + 0 = 1; so Section 2 scores 4 + 1 = 5. The test score is 1.5 + 5 = 6.5 points.

As a percentage, take Section 1 / Page 1 as an example: its questions have positive multipliers 1, 2, and (assume) 1, so the maximum possible for that page is 4. The page actually scored 1.5, so it's shown as 1.5 / 4 = 37.5% — not the raw 1.5. The same division happens at the section and test level: whatever points were earned, divided by whatever full marks on every question would have been.

Multipliers Are Copied From the Bank, Not Linked

A question's positive and negative multiplier in the Question Bank are only starting values. The moment you add the question to a test, those values are copied into the test, and from then on the bank and the test are completely independent:

• Adjusting a multiplier inside the test does not change the question's values in the bank.
• Editing the multiplier on the bank question afterward does not retroactively update any test that already added it.

Worked example: Question Q1 has a positive multiplier of 1 in the bank. You add it to Test A — Test A now has its own copy at 1. Later you raise Q1's bank multiplier to 5. Test A still scores Q1 at 1, unchanged, because its copy was made before the bank edit. If you now add Q1 to a new Test B, Test B's copy starts at 5 — the current bank value at the time of adding.

There is one exception: questions drawn from a question pool for randomization, when the pool is set to dynamic scoring (keep each question's original scoring). Those questions aren't part of the test up front — they're picked from the bank when a candidate's session starts, and they bring their current bank multipliers with them. So a bank edit does reach pool-drawn questions in any session that starts after the edit; sessions that already started keep the values they were drawn with.

In practice: if a test that's already in use needs different scoring for a question, adjust the multiplier inside that test directly — editing the bank copy only changes the starting value future tests will pick up.