Select to view content in your preferred language

Closing repeat block after record entry

79
4
13 hours ago
DavidSJohnston
Emerging Contributor

I am a relatively new user of Survey123, though I successfully deployed a survey using repeats earlier this year. This survey was for information about each group of people encountered while hiking a trail, such as the number of humans/dogs and whether they were camping. Each encounter was recorded in a repeat, with the appearance set to "minimal" so that the label and a plus sign appeared initially. The surveyor would create a new repeat record each time they encountered a new group, which would happen intermittently during the hike. 

I found that one of the most frequent causes of user confusion and error was what to do after entering all the data for a record in a repeat. There is an intuitive urge to do the equivalent of pressing "Enter" to complete the entry and move on to the next thing. If the next task is to immediately start another record for a new encounter in the same repeat, then it is intuitive enough to press the plus sign at the bottom of the repeat block and start the next record. But if they do not immediately encounter another group, it is disconcerting for them to leave the first record "open".

If they press the plus sign to start a new record, but don't have another encounter before the end of the hike, that blank record (which has required fields) prevents them from submitting the survey. Some people were able to figure out to go back and use the trash can to delete the blank record. But others were flummoxed and the best they could do was save the survey in Drafts and contact me about what to do.  

I encourage people to use the triangle button at the top of the repeat block to collapse the repeat block after each encounter. But then when they have the next encounter, the plus sign is no longer visible and they have to reopen the block using the triangle and scroll down to the end of the previous record to find the plus sign to start a new record. But this was also less than intuitive and I got negative feedback on the whole process.

I'm in the process of deploying a new version of this which will have multiple repeat blocks, in addition to Encounters, such as campsite conditions, trail issues etc. Users will need to move among different repeat blocks depending on what type of condition they run across and need to create a record for. I anticipate this "how do I finish the record and close the block?" confusion to be even more of an issue.

Ideally there would be a way to confirm the completion of each record in a repeat, which could be set to return the repeat block to its initial appearance state: closed, with the label and a plus sign. Pressing the plus sign would open the block and start a new record, just as with the first record. But I have searched the internet, the S123 documentation and this forum without finding a way to do this or otherwise control the behavior after a repeat record is completed.

Is there a setting, tool, calculation, incantation or workaround that could be used to give the user a more positive way of completing the record and, ideally, closing the block and returning to the initial appearance, or at least approximate that? Any suggestions you might have would be very welcome.

Thanks,

Dave

0 Kudos
4 Replies
MobiusSnake
MVP Regular Contributor

This isn't really advice regarding repeats in Survey123, but have you looked into using QuickCapture as an alternative?  Overall the functionality is more basic than S123 but I find it works great when I'm looking to capture a series of isolated incidents over a span of time (like when someone's walking over a distance) and the details of each incident are relatively simple (a couple questions and a photo, for example).  Provided your form doesn't have a lot of additional questions outside repeats, it sounds like it could be a good fit.

You can configure a QuickCapture project to have multiple buttons, each button represents a feature type - e.g. one button could be "Encounter", one could be "Trail Damage", one could be "Campsite Litter", and so on.  Pressing the button opens input for text/numeric input (if required) and/or a photo (if required).  There's no need to navigate between repeats, it gives you locations of each incident as separate point features, and "closing" is intuitive as you submit each incident in isolation.

0 Kudos
DavidSJohnston
Emerging Contributor

Thank you! I am not familiar with Quick Capture but will look into it. But we are pretty committed to Survey123 and I suspect we will need its full capabilities for the current project as well as planned further ones. Other than this repeat annoyance S123 is working perfectly for our needs and users are happy with it. In order to keep taking advantage of all of its capabilities I really want make it work, preferably with at least some mitigation of this issue.

Dave

0 Kudos
Philip-Wilson
Esri Contributor

Hi @DavidSJohnston,

Currently there is no way to control the apearance of a repeat (expanded/collapsed) via calculations or settings controlled via the data entry in the survey form. This can only be done on form load using the minimal appearance, and by user interaction of the collapse button.

One option could be to add a note at top of repeat telling the user to delete the repeat record if no more groups are encountered. Additionally you could add an initial question at the start of the repeat, which asks the user if they want want to enter a new record or continue, and has a yes/no, with the default as no, and based on this you can set all the following questions to be not-required and hidden, until they click yes. This would allow submitting the survey as there would be no required questions in that open record, however will create a blank record in the repeat table when submitted. This would require clean up after submission.

Hope this helps.

Regards,

Phil.

0 Kudos
DavidSJohnston
Emerging Contributor

Hi Phillip, yes that helps. At least I know to not bother trying to pursue a solution that does want I want!

I'll explore the workarounds you suggested, which might have some value. I appreciate your envisioning things that would at least reduce the incidence of confusion or bad data. 

Dave