$30 off During Our Annual Pro Sale. View Details »

#26 La data visualisation pour la data science (et les data scientists)

#26 La data visualisation pour la data science (et les data scientists)

Le métier de data scientist comporte bien des facettes, mais repose avant tout sur une bonne appréhension des données. Qu’elles soient brutes ou déjà raffinées, simples ou complexes, ``small’’ ou ``big’’ (ou ``smart’’) , la compréhension de ces ``data’’ passe par une phase d’exploration visuelle. Construire les bons outils de visualisation est alors une étape importante pour deviner les premières caractéristiques de ces données, tester les premières relations, esquisser les premiers patterns, ... Ne pas tromper – ne pas tromper soi-même –, nécessite d’avoir de bons réflexes et de bonnes méthodes afin de voir – peut-être – « ce que l’on ne s’attendait pas à voir » pour reprendre la célèbre phrase de Tuckey. Représenter ou « donner à voir aux autres », est ensuite, un autre challenge, nécessitant d’autres outils, d’autres méthodes, et posant de nouvelles questions que nous aborderons sur la base d’exemples et de contre-exemples patiemment glanés lors de cours, conférences et dans la jungle des publications (web, journaux, TV, pubs…) qui nous assaille quotidiennement.

Christophe Bontemps, ingénieur de recherche (économétrie) à Toulouse School of Economics (INRA), nous propose un tour d'horizon pratique des méthodes et des graphiques qui permettent de visualiser efficacement différents types de données. Il faudra pour cela suivre quelques règles, puisées dans la littérature (statistique, graphisme, psychologie)… pour mieux s'en affranchir, parfois !

Toulouse Data Science

January 16, 2018
Tweet

More Decks by Toulouse Data Science

Other Decks in Technology

Transcript

  1. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    Data Visualization for Data
    Scientists
    Christophe Bontemps
    Toulouse School of Economics, INRA
    @Xtophe_Bontemps

    View Slide

  2. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    MERCI À dream-team DE TDS

    View Slide

  3. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE “VISUAL PERCEPTION” OF A GRAPHIC
    What do you see ?
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )

    View Slide

  4. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE “VISUAL PERCEPTION” OF A GRAPHIC
    And here, what do you see ?
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some other points (N = 500 )

    View Slide

  5. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    “VISUAL PERCEPTION” AS A STATISTICAL TEST
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some other points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )

    View Slide

  6. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    “VISUAL PERCEPTION” AS A STATISTICAL TEST
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some other points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    “ The human eye acts is a broad feature detector and general
    statistical test”. Buja et al. (2009)

    View Slide

  7. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    “VISUAL PERCEPTION” AS A STATISTICAL TEST
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some other points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    “ The human eye acts is a broad feature detector and general
    statistical test”. Buja et al. (2009)
    Test : H0 : {There is "nothing" } = {No relation}

    View Slide

  8. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    “VISUAL PERCEPTION” AS A STATISTICAL TEST
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some other points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    0
    10
    20
    30
    40
    0 10 20 30 40
    Some X variable
    Some Y variable
    Some points (N = 500 )
    “ The human eye acts is a broad feature detector and general
    statistical test”. Buja et al. (2009)
    Test : H0 : {There is "nothing" } = {No relation}
    H1 : { There is "something" } = {There is some relation
    (Correlation, linearity, heterogeneity, groups..) }

    View Slide

  9. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- WHAT IS DATA VISUALISATION ? -]
    For Tukey (1977) “The greatest value of a picture is when it
    forces us to notice what we never expected to see”

    View Slide

  10. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- WHAT IS DATA VISUALISATION ? -]
    For Tukey (1977) “The greatest value of a picture is when it
    forces us to notice what we never expected to see”
    Cleveland (1994) says that “graphical methods and techniques
    are powerful tools for showing the structure of data”

    View Slide

  11. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- WHAT IS DATA VISUALISATION ? -]
    For Tukey (1977) “The greatest value of a picture is when it
    forces us to notice what we never expected to see”
    Cleveland (1994) says that “graphical methods and techniques
    are powerful tools for showing the structure of data”
    Bertin (1970) (translated in Bertin (1983)) defines it as a
    "visual language" and, as such, with a semiology, i.e. with a
    theory of the functions of signs and symbols.

    View Slide

  12. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- WHAT IS DATA VISUALISATION ? -]
    For Tukey (1977) “The greatest value of a picture is when it
    forces us to notice what we never expected to see”
    Cleveland (1994) says that “graphical methods and techniques
    are powerful tools for showing the structure of data”
    Bertin (1970) (translated in Bertin (1983)) defines it as a
    "visual language" and, as such, with a semiology, i.e. with a
    theory of the functions of signs and symbols.
    Tufte (2001) “Graphics are instruments for reasoning about
    quantitative information”

    View Slide

  13. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GOALS OF DATA VISUALISATION
    Data visualisation serves at least two main purposes
    Data exploration
    Graphs as visual tests, comparisons → short time to built
    and to read

    View Slide

  14. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GOALS OF DATA VISUALISATION
    Data visualisation serves at least two main purposes
    Data exploration
    Graphs as visual tests, comparisons → short time to built
    and to read
    Data representation
    Summaries, comparisons, storytelling → long time to
    build, short time to read

    View Slide

  15. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GOALS OF DATA VISUALISATION
    Data visualisation serves at least two main purposes
    Data exploration
    Graphs as visual tests, comparisons → short time to built
    and to read
    Data representation
    Summaries, comparisons, storytelling → long time to
    build, short time to read
    The problem is that :
    “ Communicating implies simplification
    data exploration implies exhaustivity”

    View Slide

  16. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "DATA TO VIZ"

    View Slide

  17. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "DATA TO VIZ"

    View Slide

  18. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "DATA TO VIZ"

    View Slide

  19. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "DATA TO VIZ"

    View Slide

  20. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "DATA TO VIZ"

    View Slide

  21. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "VIZ TO DATA"

    View Slide

  22. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "VIZ TO DATA"

    View Slide

  23. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "VIZ TO DATA"

    View Slide

  24. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE BIG PICTURE : FROM "VIZ TO DATA"

    View Slide

  25. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- DECODING VISUAL INFORMATION -]

    View Slide

  26. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GRAPHICS IMMEDIATE TO UNDERSTAND
    FIGURE – Where do people run in Paris
    source : (N. Yau)
    http://flowingdata.com/2014/02/05/where-people-run/

    View Slide

  27. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GRAPHICS IMMEDIATE TO UNDERSTAND
    FIGURE – Climate forecast uncertainty (S. Planton)

    View Slide

  28. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GRAPHICS THAT NEED EXPLANATIONS :
    FIGURE – How people spend their days (NYT).

    View Slide

  29. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- MISTAKES ! -]
    Source : New Yorker

    View Slide

  30. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES
    FIGURE – (source Wikipedia)

    View Slide

  31. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES : TRUNCATED GRAPH (ZERO AXIS)
    FIGURE – (source Wikipedia)

    View Slide

  32. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES
    FIGURE – Are you looking at the right thing ?

    View Slide

  33. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES : LIMITED SCOPE
    FIGURE – Are you looking at the right thing ?
    from Flowing data

    View Slide

  34. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES
    FIGURE – Percentage of student starting and completing a “step” in
    MOOC
    Source : My 2017’ students (Jan & Mohamed), but also recent researchers’
    presentations

    View Slide

  35. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES : PIE CHARTS

    View Slide

  36. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES : PIE CHARTS

    View Slide

  37. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES : PIE CHARTS
    Source https://twitter.com/freakonometrics/status/
    612742330160951296

    View Slide

  38. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES : STACKED BARS

    View Slide

  39. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL MISTAKES : LEGITIMATE COMPARISON
    IMPOSSIBLE !
    FIGURE – from Dix and Ellis (1998) example

    View Slide

  40. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    VERY USUAL “Mistake”...
    FIGURE – Major Cause of Worker Disability (1975-2010) (J. Schwabish,
    2014).

    View Slide

  41. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    LEGITIMATE QUESTIONS :
    What is the highest value observed ? When was it ?

    View Slide

  42. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    LEGITIMATE QUESTIONS :
    What is the highest value observed ? When was it ?
    In 2010, what is the major cause of disability ?

    View Slide

  43. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    LEGITIMATE QUESTIONS :
    What is the highest value observed ? When was it ?
    In 2010, what is the major cause of disability ?
    Is cancer (red curve) increasing over the period ?

    View Slide

  44. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    LEGITIMATE QUESTIONS :
    What is the highest value observed ? When was it ?
    In 2010, what is the major cause of disability ?
    Is cancer (red curve) increasing over the period ?
    In the recent years, which causes have increased
    (decreased) the most ?

    View Slide

  45. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    LEGITIMATE QUESTIONS :
    What is the highest value observed ? When was it ?
    In 2010, what is the major cause of disability ?
    Is cancer (red curve) increasing over the period ?
    In the recent years, which causes have increased
    (decreased) the most ?
    · · ·

    View Slide

  46. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    LEGITIMATE QUESTIONS :
    What is the highest value observed ? When was it ?
    In 2010, what is the major cause of disability ?
    Is cancer (red curve) increasing over the period ?
    In the recent years, which causes have increased
    (decreased) the most ?
    · · ·
    You do not remember a damn thing of this graph !

    View Slide

  47. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    VERY USUAL “Mistake” : SMALL MULTIPLES
    FIGURE – Major Cause of Disability - 1975-2010 (J. Schwabish, 2014).

    View Slide

  48. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    VERY USUAL “Mistake” II...

    View Slide

  49. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    VERY USUAL “Mistake” II...
    What do you see here ?

    View Slide

  50. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    VERY USUAL “Mistake” II...
    What do you see here ?

    View Slide

  51. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    VERY USUAL “Mistake” II...
    What do you see here ?
    Easier to see the trend of each curve...

    View Slide

  52. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    VERY USUAL “Mistake” II...
    What do you see here ?
    Easier to see the trend of each curve...
    Idea shared by Gelman (2004) and Munzner (2014)

    View Slide

  53. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL “Mistakes” : RADAR PLOTS
    How to represent many characteristics on a single graph ?
    http://data.visualisation.free.fr/Blog/
    Why-you-should-never-use-radar-plots.nb.html

    View Slide

  54. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    USUAL “Mistakes” : BEWARE OF RADAR PLOTS
    Representation (areas) change with axis order !
    http://data.visualisation.free.fr/Blog/
    Why-you-should-never-use-radar-plots.nb.html

    View Slide

  55. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- WHY CODING ? -]
    If decoding is hard, why coding ?

    View Slide

  56. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GRAPHICS reveal DATA : ANSCOMBE (1973) QUARTET
    We use here 4 couples of random variables : (X1, Y1), (X2, Y2)
    (X3, Y3) and (X4, Y4). All four data sets have the same
    descriptive statistics.
    Xs Mean Std. Dev. Ys Mean Std. Dev. corr(Xi, Yi) N
    X1 9 3.32 Y1 7.5 2.03 0.8164 11
    X2 9 3.32 Y2 7.5 2.03 0.8162 11
    X3 9 3.32 Y3 7.5 2.03 0.8163 11
    X4 9 3.32 Y4 7.5 2.03 0.8165 11

    View Slide

  57. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ANSCOMBE (1973) QUARTET
    All four data sets are described by the same linear model
    (Yi = α + βXi + i), revealing apparently the same
    relationships :
    Dependent variable :
    Y1
    Y2
    Y3
    Y4
    Regressed on :
    Xi
    , i=1,...,4
    0.500 ∗∗∗ 0.500∗∗∗ 0.500∗∗∗ 0.500∗∗∗
    Constant 3.000∗∗ 3.001∗∗ 3.002∗∗ 3.002∗∗
    R2 0.667 0.666 0.666 0.667
    Resid Std. Error 1.237 1.237 1.236 1.236
    F Statistic 17.990∗∗∗ 17.966∗∗∗ 17.972∗∗∗ 18.003∗∗∗
    Note : Data from Anscombe (1973). ∗ p <0.1 ; ∗∗ p < 0.05 ; ∗∗∗ p < 0.01

    View Slide

  58. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ANSCOMBE (1973) QUARTET
    A simple scatter plot (regression overlaid) shows something
    very different.
    4
    8
    12
    5 10 15
    x1
    y1
    Regression of Y1 on X1 (with constant)
    4
    8
    12
    5 10 15
    x2
    y2
    Regression of Y2 on X2 (with constant)
    4
    8
    12
    5 10 15
    x3
    y3
    Regression of Y3 on X3 (with constant)
    4
    8
    12
    5 10 15
    x4
    y4
    Regression of Y4 on X4 (with constant)

    View Slide

  59. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ANSCOMBE (1973) QUARTET
    NP : Plots of the residuals shows also same differences
    −2
    −1
    0
    1
    2
    5 6 7 8 9 10
    Fitted values
    Residuals
    Residual vs Fitted Plot
    −2
    −1
    0
    1
    5 6 7 8 9 10
    Fitted values
    Residuals
    Residual vs Fitted Plot
    −1
    0
    1
    2
    3
    5 6 7 8 9 10
    Fitted values
    Residuals
    Residual vs Fitted Plot
    −1
    0
    1
    2
    7 8 9 10 11 12
    Fitted values
    Residuals
    Residual vs Fitted Plot

    View Slide

  60. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- RULES OF THUMB -]
    “There are no “good” nor “bad” graphics (...), there are graphics
    answering legitimate questions and graphics that do not answer
    question at all ” Bertin (1970)
    It is easy to criticize ... but are there some rules ?

    View Slide

  61. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    COMPARISON IS DIFFICULT
    Yes ! there are some basic rules :
    Compare what is comparable → statistics

    View Slide

  62. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    COMPARISON IS DIFFICULT
    Yes ! there are some basic rules :
    Compare what is comparable → statistics
    Use comparable elements → data visualisation

    View Slide

  63. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    COMPARISON IS DIFFICULT
    Yes ! there are some basic rules :
    Compare what is comparable → statistics
    Use comparable elements → data visualisation
    And many people have already worked on that !

    View Slide

  64. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    COMPARISON IS DIFFICULT
    Yes ! there are some basic rules :
    Compare what is comparable → statistics
    Use comparable elements → data visualisation
    And many people have already worked on that !
    Jacques Bertin

    View Slide

  65. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    COMPARISON IS DIFFICULT
    Yes ! there are some basic rules :
    Compare what is comparable → statistics
    Use comparable elements → data visualisation
    And many people have already worked on that !
    Jacques Bertin
    Edward Tufte

    View Slide

  66. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    COMPARISON IS DIFFICULT
    Yes ! there are some basic rules :
    Compare what is comparable → statistics
    Use comparable elements → data visualisation
    And many people have already worked on that !
    Jacques Bertin
    Edward Tufte
    Tuckey, Cleveland, .....

    View Slide

  67. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    FUNDAMENTAL QUESTIONS (CHEN ET AL. (2007))
    What to Whom, How and Why ?
    A graphic may be linked to three pieces of text : its caption, a
    headline and an article it accompanies. → should be consistent
    and complement each other.

    View Slide

  68. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    FUNDAMENTAL QUESTIONS (CHEN ET AL. (2007))
    What to Whom, How and Why ?
    A graphic may be linked to three pieces of text : its caption, a
    headline and an article it accompanies. → should be consistent
    and complement each other.
    Show or explore data ?
    Different purpose, different comparisons, different requirements !

    View Slide

  69. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    FUNDAMENTAL QUESTIONS (CHEN ET AL. (2007))
    What to Whom, How and Why ?
    A graphic may be linked to three pieces of text : its caption, a
    headline and an article it accompanies. → should be consistent
    and complement each other.
    Show or explore data ?
    Different purpose, different comparisons, different requirements !
    Choice of Graphical form ?
    Choice depends on the type of data to be displayed (e.g.
    univariate continuous data, bivariate categorical data, etc..) and
    on what is to be shown, what is tested (compared).

    View Slide

  70. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    FUNDAMENTAL QUESTIONS (CHEN ET AL. (2007))
    What to Whom, How and Why ?
    A graphic may be linked to three pieces of text : its caption, a
    headline and an article it accompanies. → should be consistent
    and complement each other.
    Show or explore data ?
    Different purpose, different comparisons, different requirements !
    Choice of Graphical form ?
    Choice depends on the type of data to be displayed (e.g.
    univariate continuous data, bivariate categorical data, etc..) and
    on what is to be shown, what is tested (compared).
    Unique solution ?
    There is not always a unique optimal choice → alternatives can
    be equally good or good in different ways, emphasizing different
    aspects of the same data.

    View Slide

  71. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] EDWARD R. TUFTE’S
    In his seminal book, Tufte (2001) propose some principles for
    displaying quantitative information.
    Data : Above all, show the data

    View Slide

  72. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] EDWARD R. TUFTE’S
    In his seminal book, Tufte (2001) propose some principles for
    displaying quantitative information.
    Data : Above all, show the data
    Question : Induce the viewer to think about the substance
    rather than about methodology, graphic design. Encourage the
    eye to compare different piece of data.

    View Slide

  73. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] EDWARD R. TUFTE’S
    In his seminal book, Tufte (2001) propose some principles for
    displaying quantitative information.
    Data : Above all, show the data
    Question : Induce the viewer to think about the substance
    rather than about methodology, graphic design. Encourage the
    eye to compare different piece of data.
    Data-ink ratio : Maximize the ink-data ratio. Erase all non
    data ink, Erase redundant information

    View Slide

  74. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] EDWARD R. TUFTE’S
    In his seminal book, Tufte (2001) propose some principles for
    displaying quantitative information.
    Data : Above all, show the data
    Question : Induce the viewer to think about the substance
    rather than about methodology, graphic design. Encourage the
    eye to compare different piece of data.
    Data-ink ratio : Maximize the ink-data ratio. Erase all non
    data ink, Erase redundant information
    Integrity : Avoid distorting what the data have to say

    View Slide

  75. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] EDWARD R. TUFTE’S
    In his seminal book, Tufte (2001) propose some principles for
    displaying quantitative information.
    Data : Above all, show the data
    Question : Induce the viewer to think about the substance
    rather than about methodology, graphic design. Encourage the
    eye to compare different piece of data.
    Data-ink ratio : Maximize the ink-data ratio. Erase all non
    data ink, Erase redundant information
    Integrity : Avoid distorting what the data have to say
    General to specific : Reveal the data at different levels of
    detail (from broad picture to fine structure)

    View Slide

  76. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] EDWARD R. TUFTE’S
    In his seminal book, Tufte (2001) propose some principles for
    displaying quantitative information.
    Data : Above all, show the data
    Question : Induce the viewer to think about the substance
    rather than about methodology, graphic design. Encourage the
    eye to compare different piece of data.
    Data-ink ratio : Maximize the ink-data ratio. Erase all non
    data ink, Erase redundant information
    Integrity : Avoid distorting what the data have to say
    General to specific : Reveal the data at different levels of
    detail (from broad picture to fine structure)
    Context : Graphical display should be closely integrated with
    the statistical and verbal descriptions of the data set.

    View Slide

  77. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    PRACTICAL EXAMPLE : DATA-INK RATIO
    Let’s start with a classical graph (R default - Boxplot )
    g1 g2 g3 g4 g5
    98 100 102 104 106 108 110 112
    Groupe
    Response
    FIGURE – Distribution of a continuous variable on 4 groups

    View Slide

  78. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ERASE ALL NON DATA INK
    Groupe
    Response
    1 2 3 4 5
    98 100 102 104 106 108 110 112
    FIGURE – Distribution of a continuous variable on 4 groups

    View Slide

  79. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ERASE ALL REDUNDANT !
    Groupe
    Response
    1 2 3 4 5
    98 100 102 104 106 108 110 112
    FIGURE – Distribution of a continuous variable on 4 groups

    View Slide

  80. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GOING FURTHER...
    Groupe
    Response
    1 2 3 4 5
    98 100 102 104 106 108 110 112
    FIGURE – Distribution of a continuous variable on 4 groups

    View Slide

  81. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AND SHOW THE DATA...
    Groupe
    Response
    101.0
    100.0
    101.0
    103.8
    109.1
    1 2 3 4 5
    FIGURE – Distribution of a continuous variable on 4 groups

    View Slide

  82. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    HAVE WE LOST SOMETHING ?
    g1 g2 g3 g4 g5
    98 100 102 104 106 108 110 112
    Groupe
    Response
    Groupe
    Response
    101.0
    100.0
    101.0
    103.8
    109.1
    1 2 3 4 5
    FIGURE – Distribution of a continuous variable on 4 groups
    Did you noticed that group 1 and group 3 had the same median
    (101.0) ? see the ggplot theme + theme_tufte()

    View Slide

  83. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    PRACTICAL EXAMPLE : INTEGRITY (THE LIE FACTOR)
    LieFactor =
    Size of effect shown in graphic
    Size of effect in data
    (1)
    A Lie Factor = 1 indicates a substantial distortion
    FIGURE – Fuel economy standards. (E. Tufte - from NY Times 1978)

    View Slide

  84. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    FIGURE – Fuel economy standards (revisited)
    The "18 mpg" line measures 1.5 cm (in 1978) ; the "27,5 mpg"
    measures 13 cm (in 1985)
    −→ Lie factor = 14.5% ! ! !

    View Slide

  85. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    PRACTICAL EXAMPLE : CONTEXT
    In the next 2 minutes, you will design “the best way” to
    compare two numbers
    75and
    37.
    From S. Ortiz

    View Slide

  86. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  87. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  88. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  89. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  90. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  91. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  92. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  93. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  94. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  95. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  96. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  97. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  98. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  99. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  100. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  101. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  102. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  103. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  104. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  105. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  106. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  107. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  108. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  109. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  110. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  111. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  112. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  113. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  114. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS !
    Solutions proposed by S. Ortiz

    View Slide

  115. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS (FOLLOW-UP)
    What are the problems you’ve encountered ?

    View Slide

  116. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS (FOLLOW-UP)
    What are the problems you’ve encountered ?
    Context ?

    View Slide

  117. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS (FOLLOW-UP)
    What are the problems you’ve encountered ?
    Context ?
    Audience ?

    View Slide

  118. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS (FOLLOW-UP)
    What are the problems you’ve encountered ?
    Context ?
    Audience ?
    What to compare ?

    View Slide

  119. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS (FOLLOW-UP)
    What are the problems you’ve encountered ?
    Context ?
    Audience ?
    What to compare ?
    Units ?

    View Slide

  120. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CONTEXT MATTERS (FOLLOW-UP)
    What are the problems you’ve encountered ?
    Context ?
    Audience ?
    What to compare ?
    Units ?
    Multiple solutions ?

    View Slide

  121. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] BERTIN’S APPROACH (A VISUAL
    LANGUAGE)
    If graphs are used to communicate, it is a form of language.
    Any language has a grammar, “words” and logic. Let us study
    the science that deals with signs or sign language : “The
    Semiology”.
    TABLE – Bertin’s definition of 8 visual variables
    Position (x, y)
    Size
    Value
    Texture
    Colour
    Orientation
    Shape

    View Slide

  122. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    SHAPE IS NOT SUITABLE FOR PROPORTIONALITY
    Price of land in the East of France Bertin (1970)

    View Slide

  123. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    SIZE IS SUITABLE FOR PROPORTIONALITY
    Price of land in the East of France Bertin (1970)

    View Slide

  124. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WHAT ELSE ?
    Is color suitable for
    proportionality ?

    View Slide

  125. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    A NOTE ON COLORS
    “Colors” are not suited for ordering nor for proportionality !
    Try putting the following hues in order from low to high.

    View Slide

  126. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    A NOTE ON COLORS
    These colors are easy to order from low to high.
    Few (2008) provides meaningful solutions for choosing palettes
    of colors, for example for heatmaps.
    See also the ggplot theme theme_few()

    View Slide

  127. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] CLEVELAND-MCGILL

    View Slide

  128. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE CLEVELAND-MCGILL EFFECT

    View Slide

  129. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE CLEVELAND-MCGILL EFFECT

    View Slide

  130. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE CLEVELAND-MCGILL EFFECT
    From Cleveland and McGill (1984)

    View Slide

  131. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WEBER’S LAW AND FRAMED BOXES

    View Slide

  132. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WEBER’S LAW AND FRAMED BOXES

    View Slide

  133. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WEBER’S LAW AND FRAMED BOXES
    From Cleveland and McGill (1984)

    View Slide

  134. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [RULES :] ACCESSIBILITY
    Normal →
    Color-blind →

    View Slide

  135. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- PERCEPTION -]
    //
    Source :
    http://flowingdata.com/2014/06/25/duck-vs-rabbit-plot/

    View Slide

  136. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    “PREATTENTIVE” VARIABLES
    How many "3" in that sequence ? (from Ware (2012))

    View Slide

  137. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    “PREATTENTIVE” VARIABLES
    How many "3" in that sequence ? (from Ware (2012))

    View Slide

  138. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    “PREATTENTIVE” VARIABLES
    How many "3" in that sequence ? (from Ware (2012))

    View Slide

  139. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AND NOW...
    Find the red dot !

    View Slide

  140. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  141. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  142. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  143. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  144. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  145. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  146. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  147. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  148. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  149. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  150. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TEST : FIND THE RED DOT !

    View Slide

  151. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    HARDER : IS THERE A "STRANGER" ?

    View Slide

  152. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    HARDER : IS THERE A "STRANGER" ?

    View Slide

  153. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    HARDER : IS THERE A "STRANGER" ?

    View Slide

  154. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    HARDER : IS THERE A "STRANGER" ?

    View Slide

  155. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    HARDER : IS THERE A "STRANGER" ?

    View Slide

  156. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    HARDER : IS THERE A "STRANGER" ?

    View Slide

  157. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    HARDER : IS THERE A "STRANGER" ?

    View Slide

  158. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THAT WASN’T EASY
    Preattentive concept, Treisman (1985)
    Some visual elements or patterns are detected immediately
    But there may be interferences (colour and form)
    Very useful (detection, explanatory and presentation)
    Helpful to highlight a message !

    View Slide

  159. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    TOO MUCH VARIATION DOESN’T HELP
    From Ware (2012)

    View Slide

  160. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    MOST PREATTENTIVE VISUAL VARIABLES
    From Ware (2012)

    View Slide

  161. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    [- BEFORE/AFTER -]
    Source : Emma Rathbone - The New Yorker

    View Slide

  162. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    BEFORE (SCHWABISH - JEP, 2014)
    FIGURE – An Unbalanced Chart - Original

    View Slide

  163. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AFTER (SCHWABISH - JEP, 2014)
    FIGURE – An Unbalanced Chart - Revised

    View Slide

  164. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    BEFORE (SCHWABISH - JEP, 2014)
    FIGURE – A Clutterplot Example - Original

    View Slide

  165. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AFTER (SCHWABISH - JEP, 2014)
    FIGURE – A Clutterplot Example - Revised

    View Slide

  166. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    BEFORE - NETWORK EXAMPLE
    Victor Hugo’s characters network (Les Miserables).

    View Slide

  167. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ALTERNATIVE : ADJACENT MATRIX PLOT
    An adjacency matrix, where each cell ij represents an edge from vertex i to
    vertex j.

    View Slide

  168. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ALTERNATIVE : SORTED ADJACENT MATRIX PLOT

    View Slide

  169. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ALTERNATIVE : ARC DIAGRAM PLOT
    FromGaston Sanchez

    View Slide

  170. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    ALTERNATIVE : VERTICAL ARC DIAGRAM PLOT

    View Slide

  171. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    BEFORE - STUDENTS GRADES IN 4 COURSES
    Source Munzner (2014)

    View Slide

  172. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    BEFORE - STUDENTS GRADES IN 4 COURSES
    Source Munzner (2014)

    View Slide

  173. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AFTER - STUDENTS GRADES IN 4 COURSES
    Source Munzner (2014)

    View Slide

  174. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AFTER - STUDENTS GRADES IN 4 COURSES
    Source Munzner (2014)

    View Slide

  175. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AFTER - STUDENTS GRADES IN 4 COURSES
    Source Munzner (2014)

    View Slide

  176. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    BEFORE : US AIRLINES MAP (C. HURTER)
    Source Christophe Hurter

    View Slide

  177. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AFTER : BUNDLING THE NETWORK (C. HURTER)
    To simplify the understanding, researcher propose bundling
    techniques. 1
    Source Christophe Hurter
    1. To bundle : to tie or gather things together

    View Slide

  178. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    AFTER : BUNDLING IN NETWORKS (C. HURTER)
    Edges get closer and density gets sharper :
    Source Christophe Hurter

    View Slide

  179. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    BUNDLING IN NETWORKS (C. HURTER)
    Edges get even closer and density gets even sharper :
    Source Christophe Hurter

    View Slide

  180. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    BUNDLING IN NETWORKS (C. HURTER)
    At the end of the day...one can see through darkness !
    Source Christophe Hurter

    View Slide

  181. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE STUDIES

    View Slide

  182. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Original data are percent of users in funnel data stages, or
    gates :
    Each stage represents some portion of the previous stage.
    Source : www.storytellingwithdata.com

    View Slide

  183. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Original data observed for 4 regions of the world

    View Slide

  184. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    ... and for 3 cohorts (Q1, Q2 & Q3)

    View Slide

  185. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Identify the interesting feature : What comparisons ?

    View Slide

  186. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Identify the interesting feature : What comparisons ?

    View Slide

  187. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    By cohort

    View Slide

  188. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Remove superfluous information (Tufte’s rule)

    View Slide

  189. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Remove superfluous information (Tufte’s rule)

    View Slide

  190. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Compare cohorts trends

    View Slide

  191. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Compare cohorts trends

    View Slide

  192. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Relabel the graphic

    View Slide

  193. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Relabel the graphic

    View Slide

  194. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Make comparison& visual choices (reference line)

    View Slide

  195. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Provide insights

    View Slide

  196. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Add color, context & analysis

    View Slide

  197. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : FUNNEL DATA ANALYSIS
    Add color, context & analysis

    View Slide

  198. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 1 : BEFORE/AFTER
    Source : www.storytellingwithdata.com

    View Slide

  199. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 2 : WHAT’S THE GOOD POINT SIZE ?
    FIGURE – Sample size = 500, point size = 3

    View Slide

  200. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 2 : WHAT’S THE GOOD POINT SIZE ?
    FIGURE – Sample size = 500, point size = 1

    View Slide

  201. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 2 : WHAT’S THE GOOD POINT SIZE ?
    FIGURE – Sample size = 500, point size = 6

    View Slide

  202. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 2 : WHAT’S THE GOOD POINT SIZE ?
    FIGURE – Sample size = 500, point size = 13

    View Slide

  203. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 2 : WHAT’S THE GOOD POINT SIZE ?
    FIGURE – With Transparency ! (Sample size = 500, point size = 13)

    View Slide

  204. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 2 : WHAT’S THE GOOD POINT SIZE ?
    FIGURE – With Transparency ! (Sample size = 500, point size = 13)

    View Slide

  205. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 2 : WHAT’S THE GOOD POINT SIZE ?
    FIGURE – Context matters ! (Sample size = 500, point size = 13)

    View Slide

  206. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    How learners behave during the course ?

    View Slide

  207. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    How learners behave during the course ?
    Learner’s "activity" over time /space

    View Slide

  208. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    How learners behave during the course ?
    Learner’s "activity" over time /space
    The MOOC’s stucture :
    FIGURE – The MOOC is made of 60 different steps (ressources)

    View Slide

  209. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    How learners behave during the course ?
    Learner’s "activity" over time /space
    The MOOC’s stucture :
    FIGURE – The MOOC is made of 60 different steps (ressources)
    Learners may pick whatever step, whenever they want.

    View Slide

  210. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Basic graphic # 1 : Barplot (stacked)
    4.11
    4.10
    4.9
    4.8
    4.7
    4.6
    4.5
    4.4
    4.3
    4.2
    4.1
    3.21
    3.20
    3.19
    3.18
    3.17
    3.16
    3.15
    3.14
    3.13
    3.12
    3.11
    3.10
    3.9
    3.8
    3.7
    3.6
    3.5
    3.4
    3.3
    3.2
    3.1
    2.12
    2.11
    2.10
    2.9
    2.8
    2.7
    2.6
    2.5
    2.4
    2.3
    2.2
    2.1
    1.16
    1.15
    1.14
    1.13
    1.12
    1.11
    1.10
    1.9
    1.8
    1.7
    1.6
    1.5
    1.4
    1.3
    1.2
    1.1
    0 1000 2000 3000
    Learners
    Step
    Completed Visited but not completed Comments
    Activity per step

    View Slide

  211. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Basic graphic # 2 : Heatmap
    FIGURE – Heatmap showing activity over time

    View Slide

  212. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Idea # 1 : Visualize steps visited over time (1 line = 1 learner)
    FIGURE – How to visualize “Trajectories” over time ?

    View Slide

  213. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Too many starting points and trajectories !
    FIGURE – How to visualize “Trajectories” over time ?

    View Slide

  214. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Idea # 2 : Align starting points ! Change time for clicks sequence
    FIGURE – “Trajectories” as a sequence of “clicks”

    View Slide

  215. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Steps vs clicks sequence should reveal = patterns
    FIGURE – “Trajectories” as a sequence of “clicks”

    View Slide

  216. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Steps vs clicks sequence should reveal = patterns
    FIGURE – “trajectories” as a sequence of “clicks”

    View Slide

  217. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Should work with many lines (many learners)
    FIGURE – “trajectories” as a sequence of “clicks”

    View Slide

  218. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    Back to R.....
    FIGURE – First try : Each line is a learner’s path

    View Slide

  219. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    FIGURE – Trick one : Apply transparency and grey lines

    View Slide

  220. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    FIGURE – Trick two : Tiny lines and "jitter"

    View Slide

  221. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    FIGURE – Trick three : Colored lines, "jitter" and simple labels

    View Slide

  222. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    FIGURE – Trick four : Add context (analysis)

    View Slide

  223. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    FIGURE – Trick four : Add context (analysis)

    View Slide

  224. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    CASE 3 : SEARCHING FOR LEARNERS IN A MOOC
    FIGURE – Trick four : Add context (analysis)

    View Slide

  225. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WHAT TO REMEMBER
    From the viewer“data visualisation” are implicitly or explicitly
    comparisons or even tests (in the statistical sense)
    Graphics should help questioning

    View Slide

  226. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WHAT TO REMEMBER
    From the viewer“data visualisation” are implicitly or explicitly
    comparisons or even tests (in the statistical sense)
    Graphics should help questioning
    They should provide elements, to answer (data at least)

    View Slide

  227. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WHAT TO REMEMBER
    From the viewer“data visualisation” are implicitly or explicitly
    comparisons or even tests (in the statistical sense)
    Graphics should help questioning
    They should provide elements, to answer (data at least)
    If the question implies comparison, they should truthfully
    show the comparison

    View Slide

  228. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WHAT TO REMEMBER : THERE ARE RULES
    Data visualisation is a visual language, so there are :
    Elements of language

    View Slide

  229. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WHAT TO REMEMBER : THERE ARE RULES
    Data visualisation is a visual language, so there are :
    Elements of language
    Rules of use (spelling)

    View Slide

  230. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    WHAT TO REMEMBER : THERE ARE RULES
    Data visualisation is a visual language, so there are :
    Elements of language
    Rules of use (spelling)
    Grammar

    View Slide

  231. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE TRUTH DOES NOT LIVE IN 2D
    The “truth”....

    View Slide

  232. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    THE TRUTH DOES NOT LIVE IN 2D
    The “truth”....

    View Slide

  233. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GRAPHS I LIKE
    Visualizing thousands of runners (A. Ottenheimer)

    View Slide

  234. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GRAPHS I LIKE
    Atlas of Places (Muriz Djurdjevic & Thomas Paturet)

    View Slide

  235. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    GRAPHS I LIKE
    Experiments in Time-Distance map (Kohei Suguira)

    View Slide

  236. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    REFERENCES I
    Anscombe, F. J. (1973). Graphs in statistical analysis. The American
    Statistician, 27(1) :17–21.
    Bertin, J. (1970). La graphique. Communications, 15(1) :169–185.

    View Slide

  237. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    REFERENCES II
    Bertin, J. (1983). Semiology of graphics, translation from sémilogie graphique
    (1967).
    Buja, A., Cook, D., Hofmann, H., Lawrence, M., Lee, E.-K., Swayne, D. F., and
    Wickham, H. (2009). Statistical inference for exploratory data analysis and
    model diagnostics. Philosophical Transactions of the Royal Society of London
    A : Mathematical, Physical and Engineering Sciences, 367(1906) :4361–4383.
    Chen, C.-h., Härdle, W. K., and Unwin, A. (2007). Handbook of data
    visualization. Springer Science & Business Media.
    Cleveland, W. S. (1994). The Elements of Graphing Data. Hobart Press,
    Summit : NJ, 2 edition.
    Cleveland, W. S. and McGill, R. (1984). Graphical perception : Theory,
    experimentation, and application to the development of graphical
    methods. Journal of the American Statistical Association, 79(387) :531–554.
    Dix, A. and Ellis, G. (1998). Starting simple - adding value to static
    visualisation through simple interaction. In Eds. T. Catarci, M.
    F. Costabile, G. S. and Tarantino, L., editors, Proceedings of Advanced Visual
    Interfaces, pages 124–134. L’Aquila, Italy, ACM Press.

    View Slide

  238. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    REFERENCES III
    Few, S. (2008). Practical rules for using color in charts. Visual Business
    Intelligence Newsletter, (11).
    Gelman, A. (2004). Exploratory data analysis for complex models. Journal of
    Computational and Graphical Statistics, 13(4).
    Munzner, T. (2014). Visualization Analysis and Design. AK Peters Visualization
    Series. A K Peters/CRC Press, 1 edition.
    Treisman, A. (1985). Preattentive processing in vision. Computer Vision,
    Graphics, and Image Processing, 31(2) :156–177.
    Tufte, E. R. (2001). The Visual Display of Quantitative Information. Graphics
    Press, 2 edition.
    Tukey, J. W. (1977). Exploratory data analysis. Reading, Mass.
    Ware, C. (2012). Information visualization : perception for design. Elsevier.

    View Slide

  239. Definitions [De]coding Howto ? Rules Perception Before/After Cases So what ? Références
    STAY IN TOUCH !
    My website Data.visualisation.free.fr
    @Xtophe_Bontemps

    View Slide