Upgrade to Pro — share decks privately, control downloads, hide ads and more …

OHA Visualizing Data

Aaron
August 04, 2022

OHA Visualizing Data

Visualizing Data
Part of the OHA/SIEI Analytics Building Blocks Summer Series 2022

Aaron Chafetz and Tim Essam

Aaron

August 04, 2022
Tweet

More Decks by Aaron

Other Decks in Technology

Transcript

  1. Visualizing Data Part 1 - Aug 1 | Part 2

    - Aug 4 OHA/SIEI Analytics Building Blocks Summer Series 2022
  2. SESSION PRESENTERS 2 AARON Chafetz Senior Economist TIM Essam Senior

    Data Scientist US Agency for International Development Office of HIV/AIDS | Strategic Information Branch
  3. BLUF • Data visualization is not graphic design • Less

    is usually better, or if using a lot of data you MUST have visual hierarchy • Good data visualization takes time, lots of it • Need to be always guided by a central question 4
  4. Learning Objectives 5 • Understand the benefits of visualizing and

    exploring data • Be able to identify problems with a visualization • Feel comfortable with data visualization principles • Apply data viz principles to work products
  5. 8 How many 4ʼs are in the visual? 8 8

    2 7 5 4 6 1 5 6 4 6 8 2 4 8 8 1 4 5 8 1 8 8 0 4 3 8 4 7 6 7 3 6 0 7 0 1 8 5
  6. 9 How many 4ʼs are in the visual? 8 8

    2 7 5 4 6 1 5 6 4 6 8 2 4 8 8 1 4 5 8 1 8 8 0 4 3 8 4 7 6 7 3 6 0 7 0 1 8 5
  7. 10 How many 4ʼs are in the visual? 8 8

    2 7 5 4 6 1 5 6 4 6 8 2 4 8 8 1 4 5 8 1 8 8 0 4 3 8 4 7 6 7 3 6 0 7 0 1 8 5 These are 4s.
  8. 11 Covid response hampered by population data glitches - FT/Oliver

    Barnes & John Burn-Murdoch These choices arenʼt random; there are good reasons graphics teams do what they do
  9. 12 How many 4ʼs are in the visual? 8 8

    2 7 5 4 6 1 5 6 4 6 8 2 4 8 8 1 4 5 8 1 8 8 0 4 3 8 4 7 6 7 3 6 0 7 0 1 8 5 There are 6 fours in the visual below.
  10. 15 x1 x2 x3 x4 y1 y2 y3 y4 10

    10 10 8 8.04 9.14 7.46 6.58 8 8 8 8 6.95 8.14 6.77 5.76 13 13 13 8 7.58 8.74 12.74 7.71 9 9 9 8 8.81 8.77 7.11 8.84 11 11 11 8 8.33 9.26 7.81 8.47 14 14 14 8 9.96 8.1 8.84 7.04 6 6 6 8 7.24 6.13 6.08 5.25 4 4 4 19 4.26 3.1 5.39 12.5 12 12 12 8 10.84 9.13 8.15 5.56 7 7 7 8 4.82 7.26 6.42 7.91 5 5 5 8 5.68 4.74 5.73 6.89 Average 9.0 9.0 9.0 9.0 7.5 7.5 7.5 7.5 Variance 11.0 11.0 11.0 11.0 4.1 4.1 4.1 4.1 St. Dev. 3.3 3.3 3.3 3.3 2.0 2.0 2.0 2.0 Anscombeʼs Quartet
  11. 16 x1 x2 x3 x4 y1 y2 y3 y4 10

    10 10 8 8.04 9.14 7.46 6.58 8 8 8 8 6.95 8.14 6.77 5.76 13 13 13 8 7.58 8.74 12.74 7.71 9 9 9 8 8.81 8.77 7.11 8.84 11 11 11 8 8.33 9.26 7.81 8.47 14 14 14 8 9.96 8.1 8.84 7.04 6 6 6 8 7.24 6.13 6.08 5.25 4 4 4 19 4.26 3.1 5.39 12.5 12 12 12 8 10.84 9.13 8.15 5.56 7 7 7 8 4.82 7.26 6.42 7.91 5 5 5 8 5.68 4.74 5.73 6.89 Average 9.0 9.0 9.0 9.0 7.5 7.5 7.5 7.5 Variance 11.0 11.0 11.0 11.0 4.1 4.1 4.1 4.1 St. Dev. 3.3 3.3 3.3 3.3 2.0 2.0 2.0 2.0
  12. 17

  13. 18

  14. 19 Stop aggregating away the signal in your data -

    Visualization by Shan Carter, Zan Armstrong, Mike Freeman, and Ian Johnson
  15. 20 Stop aggregating away the signal in your data -

    Visualization by Shan Carter, Zan Armstrong, Mike Freeman, and Ian Johnson
  16. 21 Stop aggregating away the signal in your data -

    Visualization by Shan Carter, Zan Armstrong, Mike Freeman, and Ian Johnson
  17. 23 What about now? Partner achievement at PSNU level Overall

    partner achievement Partner (PSNU level) achievement capped at 110%
  18. 29 We canʼt teach you everything in three hours, but

    we can give you the ʻwhyʼ and tools you need to create a solid foundation to build from. Final Thoughts
  19. 38

  20. 42 Homework • Download and review OHA Style Guide (link)

    • Come prepared to discuss how you can apply data viz principles from the OHA Style Guide • Extra Credit: Try remaking slide 33 with this dataset
  21. 43 Recap • Visualizations are not presentations • Do less

    to do more • Leverage design principles to create more effective graphics • BE INTENTIONAL
  22. Data Viz Cheat Sheet 52 Style Guide Colors Wiki Adobe

    Color colorbrewer viz-palette Click any box for a link to the resource OHA COLORS
  23. 53 Click any box for a link to the resource

    MS Palettes Excel Hacks Excel Data Viz Open Source Tools R & Rstudio SI Custom Package
  24. 54 Click any box for a link to the resource

    Tableau Palettes Slides Template Tableau Template Tableau Learning Collaborative Tableau Tutorials Tableau Containers
  25. You will read this first 91 And then you will

    read this Then this one And you will read this last
  26. 100 PEPFAR Annual Report to Congress 2022 Key Message? “PEPFAR

    has prioritized, and made significant progress toward, transitioning a substantial majority of our funding by agency to local partners”
  27. 108 PEPFAR Annual Report to Congress 2022 Sorted Colors Clutter

    Title Labels Axes Achieved 70% goal Short of 70% goal Local partner share of budget 70% goal
  28. 109 “PEPFAR has prioritized, and made significant progress toward, transitioning

    a substantial majority of our funding by agency to local partners”
  29. 110 PEPFAR Annual Report to Congress 2022 Local partner share

    of budget IN FY21, RWANDA, SOUTH AFRICA, ETHIOPIA & UKRAINE ALLOCATED OVER 70% OF BUDGET TO LOCAL PARTNERS The remaining 24 operating units fell short of the 70% goal. Achieved 70% goal Short of 70% goal 70% goal “PEPFAR has prioritized, and made significant progress toward, transitioning a substantial majority of our funding by agency to local partners”
  30. 112

  31. Recapping our Bottom Line • Data visualization is not graphic

    design • Less is usually better, or if using a lot of data you MUST have visual hierarchy • Good data visualization takes time, lots of it • Need to be always guided by a central question 113
  32. 117 Notes and Attributions • Gestalt Principles - https://www.interaction-design.org/literature/topics/gestalt-principles •

    1.3.3 Gestalt rules from Kieran Healy's Data Visualization - https://socviz.co/lookatdata.html • DatasauRus Package - https://cran.r-project.org/web/packages/datasauRus/vignettes/Datasaurus.html • What Questions to Ask When Creating Charts - https://blog.datawrapper.de/better-charts/ • How Weʼve Learned Data Viz, and Why You May Want To Do It Differently - https://medium.com/nightingale/how-weve-learned-data-viz-and-why-you-may-want-to-do-it-differently-ec1267bd39b2 • PEPFAR Annual Report to Congress - https://www.state.gov/wp-content/uploads/2022/05/PEPFAR2022.pdf • Graphic design has rules, and they work … - https://twitter.com/MR_RO_BO_T/status/1533517961377587201 • #RotateTheDamnPlot - https://twitter.com/ikashnitsky/status/1521960898440613889?s=20&t=rxuGq6l-O8BMdYDG-zzw9A • bar charts and dot plots and line graphs, oh my! - https://www.storytellingwithdata.com/blog/bar-charts-and-dot-plots-and-line-graphs-oh-my • Examples from: Three Simple Flexible tools for Empowered Data Visualization - https://www.youtube.com/watch?v=W02ZlvulHSY • OHA Style Guide - https://issuu.com/achafetz/docs/oha_styleguide • Better Visualiztions - Jon Schwabish • Covid response hampered by population data glitches - Oliver Barnes & John Burn-Murdoch - https://www.ft.com/content/125fbaf8-175a-4e2e-852a-9995ca5176b2