Hi! I am piloting a snacking-research project and face some troubles. That is, even if participant one replies “No” to having desired/consumed a snack, follow-up questions for snack categories that were selected in previous responses will be displayed sometimes. The way the criteria are set up, this is not possible. I am wondering what I can do to avoid this from happening, any suggestions?
Hi @c.frohs
Welcome to the forum. I know setting up the criteria and the flow can be confusing sometimes. Can you tell me how did you set up the flow? Provide as much details as possible, like “Question X asks for X, and has options A and B. If A chosen, I have configured it to go to Section 1, otherwise Section 2”, and so on.
Thanks
Hi Mohammad!
Yes, so there is a baseline survey, participants are randomly assigned to either a tailored or non-tailored condition. Depending on that, a daily EMA is sent out: PP in the non-tailored condition will receive 6 pre-scheduled notifications/day, with 2 sections i have set up in criteria only for them (by the button “start” the clicked in eligibility survey). They are asked whether they have a) felt the desire to snack and b) eaten a snack. Each time if they say YES, a multiple-choice menu of snack-categories appears, they choose which apply and via criteria will be asked for this snack how many portions were desired/eaten and how strong their craving was. Filler questions will be asked only if they said NO to both craving and eating.
The tailored conditon has 2 seperate sections. To make it short: They receive their 6 notifications at tailored times (via an uploaded CSV file) and are asked craving/consumption specifically for the snacks they have indicated as their favourites (via criteria from baseline survey where they indicated those 3 top fav snacks). They can also select to haven desired/consumed another snack from a list (same as non-tailored) and are then asked the same follow-up questions.
I think we (my colleagues and me) have set up the criteria quite strictly and most works, just the remembering started to happen after a few piloting sessions. Sorry it is a lot, perhaps it might help if you have access to the survey?
Thanks for your help!
Hi @c.frohs,
Thanks for the detailed explanation, that’s very helpful.
What you’re seeing is expected behavior with criteria in Avicenna. By default, criteria can reference the most recent answer across sessions, not only the current one. This can cause follow-up questions to appear based on snack selections made in previous EMAs, even when participants answer “No” in the current session.
To prevent this, you’ll want to update your criteria so that dependent questions are conditioned on the full response chain from the current session (for example, checking both the craving or consumption question and the snack selection together).
We have a step-by-step guide that explains this behavior and how to adjust your criteria accordingly: Force Criteria to Use Only the Answers from the Current Session
Please check this and let me know if it helps prevent the issue.
And don’t hesitate to let me know if you have any questions or need help.
Thanks!