South Africa Siyenza Dashboard FY20 Q4 Follow-up

[HoustonJamie]

Timeline: Request Received: 10/19 Cluster lead: Not assigned (Treatment?/DAQ?) Leads review on: 10/26

Request Engagement form submitted by Dieter Schneider (South Africa PCO SI advisor)

OU POC Name: (see ICPI engagement form for additional details)

Technical Area(s): Siyenza (Surge support dashboard)

Preferred Date Request is Completed By: Early December

Type of Engagement: Virtual Support

Primary Requestor(s): (see ICPI engagement form for additional details)

Title of Request: Siyenza Dashboard update

HQ POC Name: N/A

[Yabayabadoo]

Meeting Notes 10/22 Attendees: Dieter S, Mabel Moti, @HoustonJamie , @gsarfaty , @wvp3 , @Yabayabadoo

• SA's request:

  • Aggregate weekly and bi weekly historical Siyenza data to Monthly
  • Update R code to merge historical monthly data with new incoming monthly data

• Outstanding Details:

  • Draft new visuals for additional indicators?
  • Provide analytic questions for additional indicators
  • Data structure of new monthly data extract
  • Aggregation approach for historical data
  • Sustainability/ Maintenance plan

• To Dos

• [x] - Dieter to share meeting discussion with CDC counterparts

• [x] - Dieter to share most recent data extracts, dashboard and code with HQ

• [x] - SA team to determine new data extract structure and decisions on outstanding items

• [x] - HQ Team to determine LOE

[Yabayabadoo]

Updates from Dieter Schlodder, 10/29 Bolded bullets outstanding items 1) Debrief CDC team a. CDC Have been briefed 2) Determine aggregation approach for historical data a. Weekly data will be rolled up to monthly, running from First Saturday to Last Friday for each month 3) Determine dataset structure of new monthly dataset a. Some Indicators will remain on the same weekly schedule for collection, Saturday to Friday b. Monthly indicators should be aligned to the same schedule c. Will remain on the Saturday to Friday Cycle, with collection rolled up from First Saturday to last Friday of the month d. Final checks with C&T Leads pending e. Interagency data format template will be put together to have consistent data structure and naming convention 4) Determine any calculated metrics to consider for R code a. Agencies will follow up with proposed calculations off of indicators for inclusion in R b. We will receive indicators that are collected weekly in monthly format, so no calculation needed here c. Agency data need to be merged into interagency data 5) Confirm scope of request to ICPI (dataset mods only vs. dataset + new visualizations) a. ICPI assistance for Dataset mods needed b. ICPI Assistance for new visuals needed, but will be with heavy collaboration of SI team

Next Steps for @gsarfaty @wvp3 @Yabayabadoo - Determine LOE and share with @HoustonJamie and ICPI Leads

[HoustonJamie]

@gsarfaty @wvp3 @Yabayabadoo - Thanks for gathering the additional info. Can you estimate LOE prior to leads at 3 pm today? If so we can likely get it approved today and i can send the email update to the South Africa team tomorrow.

[Yabayabadoo]

@gsarfaty @nmcarloni Please correct as needed. @HoustonJamie @wvp3 FYI

Met Friday afternoon to assign tasks and determine LOE. See attached document for details

• [x] Rollup historical datasets - @gsarfaty
• [x] Create calculated metrics with dummy dataset - @Yabayabadoo
• [ ] Build dashboard based on graphics - @nmcarloni

Timeline: 1) Submit Initial Draft 11/18 2) Refine based on feedback 11/20 3) Final submission 11/30 Interagency collected_calculated Siyenza indicators_graphics_11.5.20_APV.xlsx

[HoustonJamie]

@gsarfaty @nmcarloni @Yabayabadoo Leads reviewed and approved request on 11/10. I am drafting email to OU team and Chair/PPM to confirm awareness of request.

[gsarfaty]

2020-11-16 mini huddle:

• Code is being stored here.

• [x] @Yabayabadoo will update code that creates the indicator reference doc

• [x] @nmcarloni will share dashboard draft w/Yaa + Gina; then will send it to team for feedback requesting their response by 11/20; In that message, Nicole will also ask: A) for USAID's historic data inclusive of early missed & late missed appointments; B) in new data format, what will the tab be named that will have the raw data; C) do we need to account for partner shifts in the code?

[Yabayabadoo]

2020-11-18 update

Draft dashboard completed on 17/11 @nmcarloni QC Identified the following issues: 1) CDC historical data excluded data before May 11, 2019 2) USAID dataset with historical early and late missed data still outstanding 3) need guidance on accounting for partner shifts

@gsarfaty modified historical data roll up code @Yabayabadoo modified new monthly data code

[gsarfaty]

2020-12-3 check in

For HQ Team

• [x] no need to adjust for partner transitions as this should already be accounted for in the raw data
• [ ] add a time trend tab for cumulative by month graphs
• [x] rollup data to SA team (12/3)
• [ ] send dash to SA team (12/3 DC COB)
• [ ] QC Dash (12/3 DC COB)

For SA Team

• [ ] review rolled up data (including assessment of months w 5wks vs 4)
• [ ] review dashboard
• [ ] send feedback by OOB DC time 12/8

[HoustonJamie]

Draft dashboard completed on 1/14