When you have finished building your study, we recommend testing it before going live.
Testing should be performed in all parts of the study, for example the forms, randomization settings, reports and surveys. It is important to make sure the structure and content of the study are logical, as a clear structure will help to reduce the risk of mistakes when entering real data.
We recommend creating a test record in which you can test all aspects of the study just as in a standard record. You can create up to 5 test records before going live, so it's possible to add other users to the study and ask them to create some test records as well.
Considerations during testing
Are all the formulations of the questions clear?
Data entry users might not interpret the questions the way you intended. If necessary, you can add additional explanations in the 'help text' area when editing questions in the form builder.
Did you choose the correct field type for each question?
When multiple answers are provided for a question, you have to decide which multiple choice option is most appropriate. If you want the participants to answer with only one of the possibilities then it's best to use radio buttons. If you want them to check multiple options then it's best to use checkbox fields. Where you want the answer to contain only numbers, it's advised to use a number field instead of a text field.
Are all dependencies set up correctly?
Does the person entering data understand when to use reports?
To make this clear, you can use the report button in your study form; this will lead to the appropriate report.
Is your randomization appropriately set up?
When you decide to set your study to 'live', you will be given the option to reset your randomization. You can therefore randomize your test record(s) to check if the randomization settings you have chosen are working correctly without concern over influencing the data you will collect later.
We also recommend:
- Creating a data export of your test record(s) to see what the exported data will look like.
- Reviewing the option groups, to make sure the correct values are entered. It is not possible to change these once the study is live.
- If you plan to send surveys, create a test survey.
- If you plan to use automations, test these to make sure that the automations are executed properly.
When you are finished testing your study and want to start collecting data, you can delete all of your test records. You can do this by conducting a "hard delete" (this only can be done if your study has never been set to live). When your study is live it’s not possible to delete records, it is only possible to archive them. We recommend keeping 1 test record in order to test any potential adjustments to your study forms after it is set to live.
If you have any doubts, please contact us.