What can be changed after launching a survey?
It is common you are tempted to make changes or upgrades to an experiment after it has been launched. In most cases, such an approach is not recommended as it can negatively affect your results’ robustness due to biased results occurring from exposing your respondents to different questions/scenarios before and after the changes. For this reason, we distinguish below between actions that can be changed and actions that should not be changed after you have already launched the experiment.
Acceptable changes after the experiment is launched
- Formatting the question text
- Adjusting question settings that do not affect the nature of the questions and options (e.g. grammatical formatting, show go back button, force respondents to stay on a question for a specific time, possible number of characters, adjust response quality management, change the width of confidence intervals, add translations, etc)
- Changing survey panel
Not recommended changes after the experiment is launched
- Adding/removing attributes and levels
- Adding/removing question options