1 00:00:09,235 --> 00:00:12,673 In his book, the Functional Art, Alberto Cairo provides a tool for 2 00:00:12,673 --> 00:00:16,537 thinking about design tradeoffs when building information graphics, and 3 00:00:16,537 --> 00:00:18,849 he calls this tool the Visualization Wheel. 4 00:00:20,010 --> 00:00:23,110 In this conceptualization there are two poles of a circle. 5 00:00:23,110 --> 00:00:27,560 The top one represents highly complex data which informs at a deep level. 6 00:00:27,560 --> 00:00:32,790 While the bottom provides easier access to data but only informs in a shallow manner. 7 00:00:32,790 --> 00:00:37,960 Inside the circle are dimensions which describe tradeoffs between two approaches. 8 00:00:37,960 --> 00:00:40,630 While Cairo provides a number of interesting dimensions. 9 00:00:40,630 --> 00:00:42,767 It's important to note that this is a tool for 10 00:00:42,767 --> 00:00:45,400 designers to think about their visualizations. 11 00:00:45,400 --> 00:00:48,070 And not so much an analytics tool itself. 12 00:00:48,070 --> 00:00:50,870 Your needs from any particular problem might change which of 13 00:00:50,870 --> 00:00:52,790 these dimensions are important or 14 00:00:52,790 --> 00:00:56,575 might introduce new dimensions which you should be thinking about. 15 00:00:56,575 --> 00:01:00,760 Cairo actually suggests a role in an organization or a professional background 16 00:01:00,760 --> 00:01:04,410 might also influence the kinds of graphics we want to make. 17 00:01:04,410 --> 00:01:07,799 So I'd like to ask you as someone who's studying data science, 18 00:01:07,799 --> 00:01:10,893 who are you trying to reach through your visualizations? 19 00:01:14,139 --> 00:01:18,800 Let's dig in and look at a few of the trade offs that Cairo considers. 20 00:01:18,800 --> 00:01:21,710 The first is abstraction and figuration. 21 00:01:21,710 --> 00:01:25,320 A highly figurative visual describes the phenomenon using physical 22 00:01:25,320 --> 00:01:29,830 representations of the phenomena, such as photographs or drawings. 23 00:01:29,830 --> 00:01:32,230 As the representations become less real and 24 00:01:32,230 --> 00:01:36,570 more conceptual, the emphasis shifts from figuration to abstraction. 25 00:01:37,830 --> 00:01:42,660 The second dimension he discusses is, is functionality and decoration. 26 00:01:42,660 --> 00:01:45,710 A completely functional graphic has no embellishments and 27 00:01:45,710 --> 00:01:48,980 is closer to a direct representation of the data. 28 00:01:48,980 --> 00:01:52,600 While a heavily decorated graphic has more artistic embellishments. 29 00:01:53,670 --> 00:01:57,550 As it is with all of these dimensions, there isn't a clear better or worse. 30 00:01:57,550 --> 00:02:00,150 Embellishments may increase the amount of time a viewer 31 00:02:00,150 --> 00:02:02,240 spends considering the visual. 32 00:02:02,240 --> 00:02:03,900 Exploring it's nuances and 33 00:02:03,900 --> 00:02:08,680 forming mental associations which may increase familiarity and memorability. 34 00:02:10,010 --> 00:02:11,990 The third dimension is density and 35 00:02:11,990 --> 00:02:16,050 lightness as they relate to the amount of information being shown. 36 00:02:16,050 --> 00:02:19,910 There are lots of great examples of this in scientific visuals, 37 00:02:19,910 --> 00:02:22,910 where some figures are intended to be studied in depth, 38 00:02:22,910 --> 00:02:26,580 while others are meant to quickly augment a narrative. 39 00:02:26,580 --> 00:02:30,930 You can see this dimension at play when comparing magazine infographics, 40 00:02:30,930 --> 00:02:34,960 where the reader is likely to be more heavily engaged in the content. 41 00:02:34,960 --> 00:02:38,610 Two advertisements in the same magazine where readers are likely to only 42 00:02:38,610 --> 00:02:39,680 quickly consider an ad. 43 00:02:40,890 --> 00:02:44,120 Here's an example from my field of study which was published by Harvard and 44 00:02:44,120 --> 00:02:47,230 MIT describing the patterns of access of users 45 00:02:47,230 --> 00:02:50,060 in the edX massive open online course platform. 46 00:02:52,120 --> 00:02:54,320 Take a moment to study this image. 47 00:02:54,320 --> 00:02:56,590 What sense can you make of this image? 48 00:03:04,246 --> 00:03:06,977 If we consider Cairo's third dimension. 49 00:03:06,977 --> 00:03:11,780 It'd be tough not to say that this is a fairly dense graphic. 50 00:03:11,780 --> 00:03:14,460 The main portion of the graphic, a scatterplot, 51 00:03:14,460 --> 00:03:19,000 has two axes which are labeled, as do each of the two subplots to the left and 52 00:03:19,000 --> 00:03:20,200 the bottom, which are histograms. 53 00:03:21,230 --> 00:03:21,930 The axis for 54 00:03:21,930 --> 00:03:26,940 the histograms can be a bit confusing as they bleed into the scatterplot axis, but 55 00:03:26,940 --> 00:03:31,218 change the unit of measure and the direction of the measurement. 56 00:03:31,218 --> 00:03:34,601 For instance, in the X axis at the top of the scatter plot, 57 00:03:34,601 --> 00:03:38,607 percentage of chapters access is being used as the unit of measure and 58 00:03:38,607 --> 00:03:42,354 increases as you move towards the right hand side of the figure. 59 00:03:42,354 --> 00:03:46,425 You have the same X axis for the histogram to the left of the graph is in 60 00:03:46,425 --> 00:03:52,060 thousands of persons and increases as you move to the left side of the figure. 61 00:03:52,060 --> 00:03:54,560 There's some red lines overlayed on the graphic. 62 00:03:54,560 --> 00:03:59,110 Breaking the scatter plot roughly into quadrants, but with minimal labeling it's 63 00:03:59,110 --> 00:04:02,670 a little unclear what the quadrants end up representing. 64 00:04:02,670 --> 00:04:04,690 Regardless of whether this is a good figure or 65 00:04:04,690 --> 00:04:08,270 not, it would certainly be considered an example of a dense figure. 66 00:04:09,770 --> 00:04:13,730 The fourth aspect Cairo maps is the dimensionality of the graphic. 67 00:04:14,930 --> 00:04:18,410 A multidimensional graphic describes a phenomena as a whole and 68 00:04:18,410 --> 00:04:22,880 invites the viewer to explore many different aspects of the phenomena. 69 00:04:22,880 --> 00:04:27,182 A unidimensional graphic instead focuses on a single or 70 00:04:27,182 --> 00:04:30,940 a few items and explores them in one or more ways. 71 00:04:30,940 --> 00:04:34,601 The fifth dimension is originality and familiarity. 72 00:04:34,601 --> 00:04:38,792 In the modern world, we're used to seeing a plethora of different kinds of 73 00:04:38,792 --> 00:04:42,463 information graphics, things like bar charts and line charts. 74 00:04:42,463 --> 00:04:45,796 And thinking in terms of these representations is taught at a very 75 00:04:45,796 --> 00:04:47,010 young age. 76 00:04:47,010 --> 00:04:49,140 For instance, my daughter, who's almost five, 77 00:04:49,140 --> 00:04:52,720 came home from preschool the other day with a page that showed a bar chart. 78 00:04:52,720 --> 00:04:54,900 That she and others had made in class. 79 00:04:54,900 --> 00:04:58,550 To celebrate learning about the American election, the whole class got together and 80 00:04:58,550 --> 00:05:00,060 had a vote on what they would have for 81 00:05:00,060 --> 00:05:04,690 snack, chocolate ice cream or mint chocolate chip ice cream. 82 00:05:04,690 --> 00:05:08,410 Of course, mint chocolate chip ice cream won, as one might expect. 83 00:05:08,410 --> 00:05:12,343 But, my point here, is that, these are basic ways of thinking about data in 84 00:05:12,343 --> 00:05:15,489 a graphical form and they are now being taught very early on. 85 00:05:15,489 --> 00:05:18,154 And this makes them familiar to a broad population. 86 00:05:19,395 --> 00:05:21,085 This wasn't always so though. 87 00:05:21,085 --> 00:05:25,515 It's safe to say that the bar chart is fairly familiar to most people but 88 00:05:25,515 --> 00:05:27,555 a graphic with more originalities, 89 00:05:27,555 --> 00:05:31,585 elements that need to be explained or studied by the user. 90 00:05:31,585 --> 00:05:35,183 Here's a very famous graphic by Charles Minard. 91 00:05:35,183 --> 00:05:39,770 It describes Napoleon's march into Russia in 1812. 92 00:05:39,770 --> 00:05:42,130 Take a moment to study the graphic. 93 00:05:42,130 --> 00:05:46,010 Broadly speaking, there are five different kinds of information being visualize in 94 00:05:46,010 --> 00:05:47,100 this graphic. 95 00:05:47,100 --> 00:05:49,190 How many of those five pieces do you see? 96 00:05:53,290 --> 00:05:54,692 Here's what I see. 97 00:05:54,692 --> 00:05:57,340 First there are elements of geography showing us the various 98 00:05:57,340 --> 00:05:59,640 rivers in towns along the way. 99 00:05:59,640 --> 00:06:04,241 The width of the tan upper bar represents the size of Napoleon's army, and 100 00:06:04,241 --> 00:06:09,135 you can see it shrinks from the beginning of the campaign from 422,000 to 101 00:06:09,135 --> 00:06:12,365 only 100,000 once the French reach Moscow. 102 00:06:12,365 --> 00:06:16,421 The lower black bar shows Napoleon's retreat from Russia, and there are various 103 00:06:16,421 --> 00:06:20,850 points along it which are mapped to dates and temperatures in Celsius. 104 00:06:20,850 --> 00:06:24,730 We see the bar thins dramatically as the army shrinks in size. 105 00:06:26,420 --> 00:06:31,405 So the five kinds of data in Minard's visualization include location, 106 00:06:31,405 --> 00:06:34,848 direction, temperature, army size and dates. 107 00:06:36,163 --> 00:06:40,911 The last dimension which Cairo shares is the novelty and redundancy dimension. 108 00:06:40,911 --> 00:06:45,115 Redundancy is the tendency of a graphic to tell the same story in many 109 00:06:45,115 --> 00:06:46,232 different ways. 110 00:06:46,232 --> 00:06:50,870 For instance, you might use the height of bars on a bar chart with an axis as well 111 00:06:50,870 --> 00:06:53,203 as color to emphasize the largest bar. 112 00:06:53,203 --> 00:06:56,407 This one's a bit tricky, you don't want to bore your readers or 113 00:06:56,407 --> 00:06:58,289 make your graphics overly complex. 114 00:06:58,289 --> 00:07:02,478 But you do want to encode information in a way which supports their understanding of 115 00:07:02,478 --> 00:07:04,960 the phenomena you've described. 116 00:07:04,960 --> 00:07:06,007 On this dimension, 117 00:07:06,007 --> 00:07:10,084 novelty is the act of describing each phenomena in the graphic in only one way. 118 00:07:11,818 --> 00:07:14,780 There are no rights and wrongs in the visualization wheel. 119 00:07:14,780 --> 00:07:17,170 The purpose of the wheel is to help you understand and 120 00:07:17,170 --> 00:07:20,610 compare the visual approaches you might take. 121 00:07:20,610 --> 00:07:25,580 As a reflective activity, Cairo suggested you can plot your thinking, a long each of 122 00:07:25,580 --> 00:07:29,720 these dimensions and then join those points together, to create a radar plot. 123 00:07:30,900 --> 00:07:34,980 Here Cairo provides two examples of the visualization wheel in action. 124 00:07:34,980 --> 00:07:39,330 On the left hand side you see there is more emphasis towards complex visuals. 125 00:07:39,330 --> 00:07:43,750 Those which are dense, multidimensional and have high functionality. 126 00:07:43,750 --> 00:07:45,730 On the right, you see the visualization wheel. 127 00:07:45,730 --> 00:07:50,900 There are more elements of decoration, lightness, and figuration. 128 00:07:50,900 --> 00:07:54,000 Cairo suggests that the left wheel is more indicative of work 129 00:07:54,000 --> 00:07:57,860 being done by scientists and engineers, while the right wheel is more 130 00:07:57,860 --> 00:08:02,170 indicative of work being done by artists, graphic designers, and journalists. 131 00:08:03,580 --> 00:08:07,350 This tension between how scientists and engineers might view the visualization of 132 00:08:07,350 --> 00:08:11,330 data, and how artists and journalists might, is an interesting one. 133 00:08:11,330 --> 00:08:14,210 And we'll revisit it at the end of next week. 134 00:08:14,210 --> 00:08:17,550 In the next lecture, we'll go a bit deeper, and consider some heuristics from 135 00:08:17,550 --> 00:08:20,249 Edward Tufte, as to what makes a good visualization. 136 00:08:21,380 --> 00:08:25,820 The visualization wheel is one tool which we have in order to better compare two 137 00:08:25,820 --> 00:08:28,920 different ways of visualizing information. 138 00:08:28,920 --> 00:08:30,810 In the next lecture, we'll dig in a bit and 139 00:08:30,810 --> 00:08:34,690 see how you might use the visualization wheel to evaluate your own graphics.