Skip to main content
Most tests on Equip are graded automatically the moment a candidate submits. A few need human judgment, and you grade them from the assessment results page.

​
Which Tests Need Manual Grading

Manual grading applies only where judgment is required, such as watching a video answer or comparing a design. Every other test and question type is auto-graded or unscored. See How Tests Are Graded for the full list.
Until every test in a submission is graded, the test shows an ungraded label and the candidate’s Overall Score also stays ungraded. Use the Show Ungraded filter on the results dashboard to find submissions that still need your attention.

​
Grading a Submission

The path is the same for every manually graded test.
1

Open the assessment results

Click the assessment title on your Assessments Dashboard.
2

Open the test

Click the candidate’s score for the test you want to grade.
3

Award points

Review the response, then click the pencil icon in the top right corner. Repeat until no questions remain ungraded.

​
Video Response

Play the candidate’s video before you award points. Ungraded questions have a yellow outline and are marked Not Yet Graded, and the Filters dropdown can show only those.

​
CSS Challenge

Match the Target Design on the right against the candidate’s Code Preview on the left, using the mobile, tablet, and desktop icons to check each breakpoint. Click Submitted Code to read the CSS the candidate wrote.

​
Open-Ended Quiz Questions

Open the quiz marked ungraded, then read or listen to each answer marked Not Yet Graded.

​
Overriding Scores

The pencil icon stays available on every question, so you can revise points at any time, even after a test has been fully graded. The test score and the candidate’s Overall Score recompute automatically. This also applies to AI-graded responses: if you disagree with the AI’s evaluation, edit the score and your value wins.
Score changes affect how candidates rank against each other. If you regrade one candidate’s answer, review whether the same standard should apply to other submissions so comparisons stay fair.