Android Client

Namibia Specific

Reveal Data Dictionary

Namibia 2019 Electronic IRS Form Prototype.xlsx

Emailed Information 

From: Sameen Babur 
Sent: Tuesday, July 9, 2019 12:47 PM
To: Annie Martin <acmartin@akros.com>
Cc: Anne Liu <aliu@clintonhealthaccess.org>; Vivek Agrawal <vagrawal@clintonhealthaccess.org>
Subject: Updated Namibia Form

Hi Annie,

Please see attached updated Namibia form as an ODK prototype tested by the program. I am happy to talk through this on a call if needed. Apologies for the lengthy email but wanted to spell it out as much as possible as the final touches really depend on how we decide mop-ups will work, and I would not blame you if you still have questions after reading this J

Summary of changes:

1. Removed the mop-up vs. first visit question – only because I was not sure how Reveal is expected to trigger a mop-up, there was some talk of using the date or something, but can definitely include this back
2. Reporting of # of sprayable structures and # of unsprayed structures now include additional disaggregations by structure type (in addition to entering the total #)
3. More validation rules included
4. Household health number fields have been removed
5. Note: the household spray status is attempting to just represent the Reveal data model whereby that represents the spray status/coloring of the point and dictates the main skip logic of whether to continue asking questions or not

I have also taken out the complex mop-up flow logic – either they can spray and fill out the form, or they cannot and they just pick a reason why as they no  longer have to worry about double counting structures given that they are opening the same record to register mop-up info, and not submitting a whole separate form from start to finish. But obviously this still need confirmation on your side to get this right.

Key indicators and dependent fields:

1. Coverage – numerator would be total_sprayed
2. Refusal rate – numerator would be total_unspray where unspray_reason == Refusal
3. Locked rate – numerator would be total_unspray where unspray_reason == Locked

So we need to make sure that Refusal and Locked indicators are updating with mop-up visits. One option is to display all questions including their previous answer for everything under groups spray_count, unspray_count, and unspray_types and just have them review and update the integers in all those fields. Another option would be to have separate data elements that speak to # sprayed on that particular mop-up visit and # remaining unsprayed following the mop-up visit and store those separately as DiSARM did. So DiSARM had a spray_count_1 + spray_count_2 for coverage. For Refusal we would use unspray_1 and unspray_2 and just use unspray_1 for Refusal Rate after First Visit and unspray_2 for Refusal Rate after Mop-up. (See green/purple/orange in the ppt).

Looking forward to finalized the form and indicator calculations once and for all!

Sameen

Namibia DiSARM Form Flow.pptx