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.

Using visual aids effectively

963 views

Published on

Using effective visual aids is important for getting across your message when describing data. This can be in a presentation, poster or paper. This talk goes through some basic design tips that can help your visual aids look professional and work effectively.

Written for the Enabling Excellence ETN. https://eetraining.wordpress.com/

Published in: Science
  • Be the first to comment

Using visual aids effectively

  1. 1. Visual Aids Phil Ewels Using visual aids effectively phil.ewels@scilifelab.se
  2. 2. What is a visual aid? Figures Talk slides Posters
  3. 3. What is a Figures Talk slides Posters • Helps your audience understand • Often simple • Interesting » Humour / Different / Interactive good visual aid?
  4. 4. What is a Figures Talk slides Posters good visual aid? • Target audience / Content focus • Good impressions / Visual aids • Data visualisation / Workshop
  5. 5. Planning • Target audience » One size does not fit all » Be sympathetic - put yourselves in their shoes » Aim for the lowest common denominator • Content planning » Plan from the top down, not from the details up » Write down an outline before you start » Think about an “elevator pitch”
  6. 6. Elevator Pitch distillation of idea all work core concept
  7. 7. Elevator Pitch distillation of idea all work core concept • Paper abstracts • Graphical ToCs • Talk poster / focus • Coffee chat
  8. 8. Best / Worst Examples
  9. 9. Phil Ewels - Challenging samples for NGS / 20 Sample Setup 9 SciLifeLab ID Library Prep Starting amount Sequenced Reads P1102_101 Manual 1000 ng 38,995,594 P1102_102 Manual 1000 ng 37,663,274 P1102_103 Manual 1000 ng 39,666,722 P1102_104 Manual 500 ng 35,332,272 P1102_105 Manual 200 ng 40,568,034 P1102_106 Manual 50 ng 47,044,650 P2011_1005 NeoPrep Run 1 25 ng 93,316,971 P2011_1006 NeoPrep Run 1 25 ng 115,648,988 P2011_1007 NeoPrep Run 1 25 ng 118,489,187 P2013_1004 NeoPrep Run 2 25 ng 72,128,476 P2013_1005 NeoPrep Run 2 25 ng 62,774,142
  10. 10. Phil Ewels - Challenging samples for NGS / 20 Sample Setup 10 Sample Library Prep Starting amount (ng) Sequenced Reads (M) 1 Manual 2 Manual 3 Manual 4 Manual 5 Manual 6 Manual 7 NeoPrep Run 1 8 NeoPrep Run 1 9 NeoPrep Run 1 10 NeoPrep Run 2 11 NeoPrep Run 2 25 25 25 25 25 50 200 500 1000 1000 1000 0 30 60 90 120
  11. 11. Visual Design
  12. 12. Introduction • Don’t underes-mate the impact of your first few slides • Fonts and visual presenta-on immediately set the tone for your audience • Anchor your work in the context of your audience’s work • Go slow - everyone will thank you for it • This includes not using too much content • Try not to read every bullet point from the screen - talk around your slides instead • Don’t put all of your bullets up at once, the audience will read them instead of listening to you • Now is the perfect -me to use a visual aid NO
  13. 13. Introduction • Don’t underestimate the impact of your first few slides • Fonts and visual presentation immediately set the tone for your audience • Anchor your work in the context of your audience’s work • Go slow - everyone will thank you for it • This includes not using too much content • Try not to read every bullet point from the screen - talk around your slides instead • Don’t put all of your bullets up at once, the audience will read them instead of listening to you • Now is the perfect time to use a visual aid
  14. 14. Visual Design • Visual design is important • Visual design is easy » Clear message » Focussed » Easy to read and interpret » Honest and true reflection of the data • Fonts. Colours. Layout.
  15. 15. Fonts Body text Figures Presentations
  16. 16. Sans-serif Fonts WARNING serif fonts WARNING serif fonts
  17. 17. Fonts • Pick a font and stick to it • Avoid MS defaults » https://www.google.com/fonts • Make use of font weights Open sans Lato Roboto (Arial) Bold Medium Regular Light Thin Hairline Cambria Calibri
  18. 18. Every time you use Comic Sans, Faye will punch this adorable little bunny. comic sans criminal.com
  19. 19. Colour Palettes
  20. 20. Google Material Design Guidelines Set of guidelines about design Aimed for app developers Includes some nice colour palettes Lots of good stuff about design theory https://www.google.com/design/spec/style/color.html
  21. 21. https://coolors.co
  22. 22. Data Visualisation
  23. 23. Choosing a plot • What type of graph best represents the argument that you’re trying to make • Which data are necessary Relationship Comparison Composition Distribution
  24. 24. What are you trying to show? Distribution Relationship / Comparison / Composition / Distribution
  25. 25. What are you trying to show? Relationship Relationship / Comparison / Composition / Distribution
  26. 26. What are you trying to show? Composition Relationship / Comparison / Composition / Distribution
  27. 27. What are you trying to show? Comparison Relationship / Comparison / Composition / Distribution
  28. 28. What are you trying to show?
  29. 29. What are you trying to show?
  30. 30. What are you trying to show? • Distributions are not relevant • Only the median value is needed • Trends difficult to compare • Easier comparison
  31. 31. Making Comparisons • Fast, accurate judgement » Length » Slope • Medium judgement » Colour » Patterns (eg. grouping) • Slow, inaccurate judgement » Angle » Area » Text
  32. 32. Making Comparisons • Length & Slope » Bar / column charts » Box plots » Line graphs • Colour » Heatmaps • Patterns » Scatter plots • Angle » Pie charts • Area » Venn diagrams • Text » Tables » Plot labels
  33. 33. Making Comparisons 1 2 3 4 5 6 7
  34. 34. Making Comparisons 1 2 3 4 5 6 7 0 0.5 1 1.5 2 1 2 3 4 5 6 7
  35. 35. Making Comparisons 1 2 3 4 5 6 7 • Angles are bad for comparison • Legend is disassociated from plot • Requires colour link for series identification 0 0.5 1 1.5 2 1 2 3 4 5 6 7
  36. 36. Making Comparisons
  37. 37. Making Comparisons
  38. 38. Keep it simple
  39. 39. Keep it simple 0.0 0.5 1.0 1.5 2.0 1 2 3 4 5 6 7 Series1 Series2 • Use of 3D was arbitrary • No need for colour and texture
  40. 40. Aspect Ratio • Elements that need accurate aspect ratios: » Images » Text » Anything circular » Axes with comparable units
  41. 41. Aspect Ratio
  42. 42. Aspect Ratio
  43. 43. Aspect Ratio
  44. 44. -6 -4 -2 0 2 4 6 8 10 12 14 -6 -4 -2 0 2 4 6 8 10 12 14 log10FPKM log10 FPKM Aspect Ratio -6 -4 -2 0 2 4 6 8 10 12 14 6 -4 -2 0 2 4 6 8 10 12 14 log10FPKM log10 FPKM
  45. 45. Features of a good plot • Minimalistic • Suitable plot type • Big and clear • Attractive (avoid defaults) • The test - can you draw it from memory? 0 0.5 1 1.5 2 2.5 1 2 3 4 5 6 7 0.0 0.5 1.0 1.5 2.0 2.5 1 2 3 4 5 6 7
  46. 46. Using a graph • Hold audience focus » Talk through your data » Don’t show everything at once • Use layering for complex plots » Progressively add data
  47. 47. Colour
  48. 48. Colour • Colour can be used to: » Highlight specific data » Group categories of data » Encode quantitative values • The more selective you are with colour, the greater its effect 0.0 0.2 0.5 0.7 0.9 1 2 3 4 5 6 0.0 0.2 0.5 0.7 0.9 1 2 3 4 5 6
  49. 49. Colour Scales • Represent quantitative data • Sequential • Divergent • Categorical
  50. 50. Colour Sensitivity
  51. 51. Colour Sensitivity Humans do not perceive the spectrum evenly Spectrum is a bad colour scheme with which to encode data
  52. 52. Bad Defaults Jet Luminescence afmhot Luminescence YlOrRd Luminescence not even not even
  53. 53. Scale types Categorical Sequential
  54. 54. Scale types Sequential (should be categorical) Which category is this?
  55. 55. Colour Blindness • Common in Northern European men • Colour schemes such as Magenta – Green designed to be colour blind friendly 22% chance at least one colour blind 3 NE male reviewers
  56. 56. Colour Blindness Normal Vision Protanopia
  57. 57. Colour Blindness Normal Vision Protanopia
  58. 58. Color Brewer http://colorbrewer2.org Sequential Divergent Categorical
  59. 59. Summary
  60. 60. Summary • Decide your key points early » Build around target audience • Show only what you need to » Bullets, not prose » Suitable graphs » Avoid defaults • Use data as a visual aid colorbrewer2.org google.com/fonts flaticon.com coolors.co Adobe Illustrator Inkscape (free)
  61. 61. Credits Course written by Phil Ewels. Some material developed whilst working at the Babraham Institute in Cambridge, UK. Now working at the National Genomics Infrastructure, part of the Science for Life Laboratory in Stockholm, Sweden. Find more at http://phil.ewels.co.uk

×