Random order and shuffle options
Reduce position bias without messing up the logic.
2 minutes read
On this page
Reduce position effects without breaking paths.
Purpose
Vary exposure while keeping answers linked to the correct options.
When to use it
When comparing alternatives whose position may influence choices.
Before you start
- Contiguous blocks for a group
- Reviewed logic rules
Example
Randomize concepts A and B while keeping None as the last option.
Step by step
- Choose the range. In the builder, place Concept A and Concept B next to each other in the list. Put shared instructions before them and final questions after them. Check that each concept makes sense independently of the other answer. Concept A · Concept B.
- Define random group. Under Logic, open Random order and select the two adjacent blocks. Click Create group and check the listed names. If the button stays disabled, check that at least two adjacent blocks are selected outside other groups. No dependence on the previous question.
- Shuffle options. Select the choice or screener question and enable answer shuffling. If None must remain last, enable the last-option setting. Check its position in the authored list before testing. Shuffle · Pin last if needed.
- Compare preview assignments. Click Study preview and check Assigned order. Use Restart same participant: the order should stay the same. Click New participant to simulate another assignment and follow the questions; two people may receive the same order by chance. Preview does not create sessions or take quota slots. To validate real collection, publish and start distinct participations · Each real participation may create a response and consume quota.
Validate before collecting responses
In the builder, Study preview applies draft groups, rules and screening without creating sessions or taking quota slots. For Figma tasks, websites and interviews, it shows Simulated answer to test logic: the author supplies the outcome; this does not demonstrate a real interaction.
Preview in new tab lets you check each block on desktop or mobile, including interactive Figma navigation and a simulated interview conversation. This block preview does not apply study skip rules.
To validate real websites, recording, voice and AI follow-up, publish and use Open link in new tab in Share. The published link opens a real participation at /r/[token]: it may create responses and consume quota. Keep this participation separate when analyzing collection.
How to read the result
- Stable order within each session
- Less dependence on first position
Cautions and limitations
- Do not create jumps incompatible with random-group boundaries.
- Option identity is preserved, not its screen position.
Continue with the interactive guide
Related guides
Is something missing? Write to contato@paineltap.com.br.