Reasoning and judgement
Scenario responses and “what would you do and why”, where thinking matters as much as the conclusion.
A prompt and a blank space for the participant's own words. Also called essay questions, they measure reasoning no tick box can.
A long answer question asks the participant to compose an extended response in their own words, a paragraph or several, rather than select or supply a short fixed answer. It belongs to the family of open-ended, or constructed-response, questions, which also covers essay questions: formats where the participant constructs the answer instead of choosing it. This is the opposite end of the spectrum from multiple choice, and it measures something different.
Objective questions test whether someone knows a fact; open-ended questions test what they can do with it. When you need to see reasoning, judgement or the quality of a written explanation, there is no substitute. The value lies precisely in the answer you didn't anticipate: a participant can't guess their way through, or hide shallow understanding behind a lucky selection.
That richness comes at a cost: open-ended answers can't be graded by exact matching, because there is no single correct string. They require human judgement, applied consistently. What makes essay-style marking fair is a clear marking guide, written before you read the responses and applied the same way to every one. That is what the grading tools are built around.
A colleague reports a near-miss involving a reversing vehicle in the yard. Describe the steps you would take, and explain why each one matters.
Graded against a marking guide rather than a fixed answer. For example: acknowledges and records the near-miss, secures the area, escalates to a supervisor, and explains the reasoning behind each step.
Reach for it when the answer is reasoning or expression, not a fact to be matched.
Scenario responses and “what would you do and why”, where thinking matters as much as the conclusion.
When you need to see how clearly someone can explain an idea or justify a decision.
Open-ended prompts reveal whether a participant can transfer what they know to an unfamiliar case.
If the answer is a single fact, an objective type gives instant, consistent results at scale.
Multiple choice questions →Long answer is the one format TestSetGo does not grade automatically: a fair judgement is a human one. When an attempt is submitted, the objective questions score instantly and the long answers drop into a manual grading queue. The dashboard shows how many responses are still waiting, so nothing is quietly missed.
Each response sits beside its marking guide, frozen at the moment the run opened, so every participant is judged against the same criteria even if you revise the guide later. Enter points, add a comment, and move on with save-and-next.
Every grade is recorded with an audit trail: who marked what, and when. You can override a score, and the change is logged rather than silently replacing the original. Grade one question across all participants for consistency, or one participant across all questions.
An open-ended, or constructed-response, question asks the participant to compose an answer in their own words rather than choose from options. Long answer and essay questions are the main examples, measuring reasoning and expression that objective formats can't.
No. TestSetGo grades them manually, because fair judgement of an open response is a human task. They enter a grading queue beside a frozen marking guide, while the objective questions on the same attempt score instantly.
Write a marking guide before you read the responses and apply it the same way to every one. TestSetGo freezes the guide when the run opens, offers save-and-next through the queue, and lets you grade one question across all participants at a time.
Yes. You can override a score, and the change is recorded in the audit trail rather than silently replacing the original, so the marking history stays transparent and defensible.
Mix open-ended questions with auto-graded ones, and mark the written responses fairly against a frozen guide.