Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Data visualization in R


Published on

A brief PechaKucha talk highlighting contemporary data viz philosophy in R. The grammar of graphics is described, contrasts between base graphics and ggplot2 tools are described, and strategies to avoid overplotting are proposed.

Published in: Data & Analytics
  • Be the first to comment

  • Be the first to like this

Data visualization in R

  1. 1. contemporary data viz in R @cjlortie
  2. 2. philosophy of graphics clean and simple
  3. 3. grammar of graphics
  4. 4. layers
  5. 5. primacy of layers (simple first, then complex)
  6. 6. data are mapped to aesthetic attributes and objects
  7. 7. data are always mapped but explicit control is more powerful
  8. 8. data first then statistics
  9. 9. you define purpose not the data visualization
  10. 10. basic graphics grid graphics lattice and lattice extra ggplot2
  11. 11. base <- edit only on top grids <- provide structure lattice <- trellis graphics (details/conditions) ggplot2 <- grid/trellis but model approach
  12. 12. implications for workflow model suggests structures aesthetics always important building takes forethought
  13. 13. plot versus qplot
  14. 14. qplot automatically maps data attributes
  15. 15. geoms are the types of objects you map data unto
  16. 16. qplot takes care of the layers for you
  17. 17. ggplot() requires you to add a layer with +
  18. 18. tip: assign ggplot to object then add
  19. 19. beware overplotting
  20. 20. faceting jitter alpha