Download presentation

Presentation is loading. Please wait.

Published byBlaze Corse Modified about 1 year ago

1
Dr. Michael R. Hyman, NMSU Chart Deception Main Source: How to Lie with Charts, by Gerald E. Jones

2
2 Information Information = Data + Labels =data (505) =information

3
3 Summaries Can Hide Needed Detail

4
4

5
5 Pie Charts Pie charts denote percentages Label slices with percentages, not raw values –True of two pie figures Most important slice –Right-hand slice –Exploded slice Bottom slice of 3-D pie seems overly important

6
6

7
7

8
8

9
9 Pie Charts Pie charts denote percentages Label slices with percentages, not raw values –True of two pie figures Most important slice –Right-hand slice –Exploded slice Bottom slice of 3-D pie seems overly important

10
10

11
11 Pie Charts Pie charts denote percentages Label slices with percentages, not raw values –True of two pie figures Most important slice –Right-hand slice –Exploded slice Bottom slice of 3-D pie seems overly important

12
12

13
13 Pie Charts Pie charts denote percentages Label slices with percentages, not raw values –True of two pie figures Most important slice –Right-hand slice –Exploded slice Bottom slice of 3-D pie seems overly important

14
14

15
15 Pie Charts Multi-sized pies on same figure (to denote trend) biased by dissimilar diameter-area rates of growth Large ‘Other’ slice can hide needed detail Use pie-column chart for more detail on a slice Donut charts confusing Omitting missing data increases relate size of all slices

16
16

17
17 Pie Charts Multi-sized pies on same figure (to denote trend) biased by dissimilar diameter-area rates of growth Large ‘Other’ slice can hide needed detail Use pie-column chart for more detail on a slice Donut charts confusing Omitting missing data increases relate size of all slices

18
18

19
19 Pie Charts Multi-sized pies on same figure (to denote trend) biased by dissimilar diameter-area rates of growth Large ‘Other’ slice can hide needed detail Use pie-column chart for more detail on a slice Donut charts confusing Omitting missing data increases relate size of all slices

20
20

21
21 Pie Charts Multi-sized pies on same figure (to denote trend) biased by dissimilar diameter-area rates of growth Large ‘Other’ slice can hide needed detail Use pie-column chart for more detail on a slice Donut charts confusing Omitting missing data increases relate size of all slices

22
22

23
23 Attend to Cultural Bias in Orienting Charts

24
24

25
25

26
26 Argo Sales Increase Sales, in units or dollars, listed along the y axis Avoid 3D effects

27
27

28
28 Bar Chart—Avoid 3D and Sideways

29
29

30
30

31
31

32
32

33
33 Bar and Area Charts Stacked vs. clustered bar charts –Mixing cumulative and clustered bars can confuse viewers When stacking, place least variable component on bottom Avoid paired-bar charts Viewers can misread 3-D bars

34
34

35
35

36
36

37
37 Bar and Area Charts Stacked vs. clustered bar charts –Mixing cumulative and clustered bars can confuse viewers When stacking, place least variable component on bottom Avoid paired-bar charts Viewers can misread 3-D bars

38
38

39
39 Bar and Area Charts Stacked vs. clustered bar charts –Mixing cumulative and clustered bars can confuse viewers When stacking, place least variable component on bottom Avoid paired-bar charts Viewers can misread 3-D bars

40
40

41
41 Bar and Area Charts Stacked vs. clustered bar charts –Mixing cumulative and clustered bars can confuse viewers When stacking, place least variable component on bottom Avoid paired-bar charts Viewers can misread 3-D bars

42
42

43
43 Bar Graph—Avoid 3D

44
44 Avoid 3-D Effects

45
45

46
46 Line Graph—Avoid 3D

47
47

48
48 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

49
49

50
50

51
51 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

52
52

53
53

54
54

55
55

56
56 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

57
57

58
58

59
59 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

60
60

61
61

62
62 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

63
63

64
64

65
65 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

66
66

67
67

68
68

69
69 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

70
70

71
71 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

72
72

73
73 XY (or Scatter) Charts Label data points Start axes at coordinates (0,0) Avoid distorting axes Avoid broken axes Cumulative charts seem optimistic because upward sloping Avoid dissimilar data on same chart Semi-log charts can minimize trends If x axis isn’t time axis, connecting dots can mistakenly suggest a trend Interpolation and extrapolation are subjective

74
74 Radar Charts Beware non-uniform scales on axes

75
75

76
76 Depicting Confidence Intervals

77
77

78
78

79
79

80
80 Headers and Footers Use footnotes to source table (avoid plagiarism charge) Font rules –Use proportional font –Don’t mix fonts in a chart –Use upper and lower case letters Spell out numbers 1 through 9 unless a measure

81
81

82
82 Examples of Bad Graphics

83
83

84
84

85
85

86
86

87
87 Examples of Bad and Better Graphics

88
88 Example #1

89
89

90
90 Example #2

91
91

92
92 Example #3

93
93

94
94 Example #4

95
95

96
96 Example #5

97
97

98
98 Example #6

99
99

100
100 Example #7

101
101

102
102 Example #8

103
103

104
104

Similar presentations

© 2016 SlidePlayer.com Inc.

All rights reserved.

Ads by Google