Recommended
Recommended
More Related Content
What's hot
What's hot (20)
Viewers also liked
Viewers also liked (7)
Similar to Vital visualizations 2016 10 05
Similar to Vital visualizations 2016 10 05 (20)
Recently uploaded
Recently uploaded (20)
Vital visualizations 2016 10 05
- 1. Vital Visualizations Drew Marco Daugherty Business Solutions andrew.marco@daugherty.com
- 2. Agenda Overview History of Visualizations Visual Design Influences The Goldfish Effect Design Considerations Listener Segmentation Correlation != Causation Best Practices Q&A References
- 3. Overview
- 4. Overview How do you ensure your visualizations are vital to your business How the brain processes visual images Why the processing of visual images impacts your design Important design considerations
- 5. Characteristics Non-Visual Data Produce an Image Readable & Recognizable
- 7. Ptolemy’s World Map (2 A.D.) History of Visualizations
- 8. Bar and Time Series Plot of Economic and Agricultural Data History of Visualizations
- 10. History of Visualizations Cholera Map of London John Snow did know something after all!!!
- 14. Author vs. Reader Driven • Does not allow interactivity for the reader • Author determines information gleaned from the visualization • Also known as Explanation or Guided Analytics Author Driven • Allows interactivity for the reader • Users play with data • Users gather their own interpretation and understanding of information • Also known as Exploration or Self-service analytics Reader Driven Source: Narrative Visualizations: Telling Stories with Data
- 18. Telling a Story Narrative Descriptive Expository Persuasive Writing: • Narrative • Telling a story (fiction, non-fiction, essay, etc.) • Descriptive • Creating a vivid picture of a noun (person, place or idea) • Expository • To the point; factual writing (instructions, directions, etc.) • Persuasive • Using writing to express an opinion or take a stance through your writing Visualizations: • Narrative • Combination of the story and numbers • Descriptive • Utilizing visual presentation to describe the data set • Expository • Minimalist visualization conveying meaning to audience • Persuasive • Presenting data in a way which allows the viewer to easily understand the point of view
- 19. The Writer’s Block Many writers have provided advice on their writing process Some of this information can be applied to data visualizations Rudyard Kipling: “If history were taught in the form of stories, it would never be forgotten.” Roald Dahl “I don’t care if a reader hates one of my stories, just as long as he finishes the book.” William Zinsser “To defend what you’ve written is a sign that you are alive.” Tom Clancy “I do not over-intellectualize the production process. I try to keep it simple. Tell the damned story.” Carl Sandburg “Beware of advice – even this.”
- 20. There are lies, damn lies and statistics Mark Twain Writing Advice “Get your facts first and then you can distort them as much as you please.” “You need not expect to get your book right the first time.” “Use plain, simple language, short words and brief sentences. That is the way to write English – it is the modern way and the best way. Stick to it; don’t let fluff and flowers and verbosity creep in.” “Facts are stubborn, but statistics are more pliable.” “A successful book is not made of what is in it, but what is left out of it.” “One should never use exclamation points in writing. It is like laughing at your own jokes.” “Write what you know.”
- 21. “Don’t use passive voice; Timid writers like passive verbs for the same reason that timid lovers like passive partners. The passive voice is safe.” “Don’t obsess over perfect grammar. The object of fiction isn’t grammatical correctness but to make the reader welcome and then tell a story.” “You have three months. The first draft of a book – even a long one – should take no more than three months, the length of a season” “Write one word as a time” “Eliminate distraction. There should be no telephone in your writing room, certainly no TV or video games for you to fool around with”
- 22. Overall Visualization Process Research Data Vet Data Sources Filter Findings Determine Visualization Type Craft Story Review & Revise • Business Question • Focus on Nouns • Identify Data Sources • Trustworthy Data Only • Data Consumer Confidence • Start with full data • Pare down • Important Elements Only • Refer to flow chart • Keep Interesting • Keep Simple • Creative & Analytical Resources • Be original • New POV • Unbiased 3rd Party • Feedback • Enhance
- 24. Photography Influence Photograph Visualization Message Bountiful snow, how small people are compared to mountain of snow What is the message you are trying to tell to your audience? Intent Everything (picture frame, shutter speed, depth of field, etc.) derived from intent Everything (Colors, visualization type, scale, scope, etc.) derived from intent Considerations Snow in 2/3 of photo, skiers small, focused on primary subject (snow), filtered sunlight accordingly Vivid color scheme, appropriate scale, scope of visualization, filtered data accordingly
- 26. Visual Design Influences Principles of Design Balance Gradation Repetition Contrast Dominance Harmony Unity
- 27. Visual Design Influences Balance Visually consistent arrangement of elements
- 28. Visual Design Influences Gradation Showing linear perspective through changes in size or color
- 29. Visual Design Influences Repetition Presentation of the same image in a pattern; with variation is interesting; without variation can be monotonous
- 30. Visual Design Influences Contrast The juxtaposition of opposing elements – colors, shapes, sizes, etc.
- 31. Visual Design Influences Dominance Using object size (typically larger) to draw attention
- 32. Visual Design Influences Harmony Combining similar, related elements for visually appealing result
- 33. Visual Design Influences Unity Visual linking of various elements in the work
- 34. Visual Design Influences Cognitive Psychology
- 35. Cognitive Psychology Deals with mental processes such as attention, perception, creativity, thinking, etc. Of these processes, perception has the largest impact on visualizations Understanding perception allows you to “hack” your users’ brain Create more efficient visualizations
- 36. Perception Most important for visualizations Process by which human brain takes in external stimuli from the surrounding environment (i.e. the way the human brain “sees” information in dashboards and visualizations) Pre-attentive processing Electronic process within the brain Attentive processing Physical process requires changes in synapses between neurons in the brain Involves the subconscious processing of visuals Quickly processes: Position, Color (Hue and Intensity), Alignment, Size and Shape Does not quickly process: Text and Numbers – requires focus and attention to process
- 37. Number Crunching On average it probably took 5-10+ seconds to spot TURTLE
- 38. Number Crunching Spot the different number Sales by Date Date Monday Tuesday Wednesday Thursday Friday 12/10/2015 12345 12345 12345 12345 12345 12/11/2015 12345 12345 12345 12345 12346 12/12/2015 12345 12345 12345 12345 12345 12/13/2015 12345 12345 12345 12345 12345 12/14/2015 12345 12345 12345 12345 12345 12/15/2015 12345 12345 12345 12345 12345 12/16/2015 12345 12345 12345 12346 12345 12/17/2015 12345 12345 12345 12345 12345 The average person takes 3-4 seconds to spot the different number
- 39. I Spy with My Little Eye Spot the different shape Pre-attentive visual features allow processing within 40-50 milliseconds
- 40. Spitting image How many black dots can you see at one time?
- 41. Principles of Perception Proximity Objects placed close together are perceived as a group Similarity Objects sharing similar attributes, such as color or shape, are perceived as a group Enclosure and Continuity Objects with a boundary or continuation around them are perceived as a group Closure Open structures can be easily perceived as closed, complete
- 43. The Goldfish Effect 2013 Average American Attention Span Goldfish Attention Span 12 seconds 9 seconds 8 seconds Average American Attention Span 2000
- 44. Attention Span More than ever, research is highlighting a trend in reduced attention and concentration spans … the younger generations appear to be the worst afflicted. Sociologist David Moxon … individuals who multi-task emails, phone calls and social-networking sites have more trouble paying attention and focusing on important information. The New York Times
- 45. “Calgon, Take Me Away” When you are stressed … this stress takes a lot of resources from your brain and interferes with your capacity to encode any new information. Center for Studies on Human Stress
- 47. Design Considerations Audience Considerations Avoid Charting Tricks Common Mistakes
- 48. Audience Considerations What information is needed to be successful? What is the correct level of detail? What actions can be taken and what values need action? What learned or cultural assumptions may affect design choices? What do colors mean? Which icons are familiar? Don’t forget color blindness Color blind friendly color palettes http://colorlab.wickline.org/colorblind/colorlab/
- 51. Design Considerations Gridlines are best kept subtle Avoid displaying excessive detail Don’t highlight unimportant information Dashboards should fit entirely on one screen Avoid adding unnecessary/useless decoration Avoid an unappealing visual design Avoid text heavy Infographics
- 52. Common Mistakes Inappropriate display choices distorting reality Variety just because… Too much information Poorly designed (noisy fill patterns, line styles, or saturated/bright colors) Inaccurate encoding of quantitative data Inconsistent ordering and placement Inconsistent or reversed scales Proportional axis scaling Using counts vs. percentages when comparing periods with different totals
- 54. Listener Segmentation Novice • New to the subject, but does not want oversimplification Generalist • Aware of the topic, looks for an overview and the story’s major themes Management • Seeks in-depth actionable understanding with access to detail Expert • Wants more exploration and discovery with less story-telling Executive • Needs to know the significance and conclusions of weighted probabilities Source: Jim Stikeleather, Dell Executive Strategist in Harvard Business Review Article
- 56. Correlation doesn’t imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing ‘look over there’.
- 60. Facts are stubborn, but statistics are more pliable Running Back “A” Most rushing TDs since entering the NFL Sixth in fantasy points among RBs since joining NFL He has never missed a game Only three RBs have more 1st Downs in the last two years Eighth in total yards after contact since 2014 (among active RBs) Averaged 7 yards per reception No WR/RB/TE touched the ball more on his team last season His team’s offense was one of the 10 highest scoring offenses in the league Running Back “B” Averaged 3.56 yards per carry (42nd among RBs with at least 100 carries) 49th in running routes among RBs Finished with fewer than 6 fantasy points in 8 different games Had at least 5 different games with at least 9 touches and 2 or fewer points Averaged 2.2 yards per carry on red zone carries No RB in the NFL has lost more fumbles over the last two seasons Career lows in total yards, yards per carry, receptions, receiving yards and fantasy pointsBengals RB Jeremy Hill
- 61. Best Practices
- 62. Best Practices Empower Context Digestible Engagement • Present KPIs (metrics) compared to goals • Use color to differentiate different metrics • Regularly scheduled email delivery • Metric driven report notifications • Interactive Visualizations • Information inspires action • Actionable and meaningful content • Cannot be everything to everyone • Center of Excellence (COE) • Supports consistency • Tips and Tricks
- 63. Visualization Categories • Compare two variables • Show trends over time • Also known as Time Series Line Charts • Compare quantities of different quantities • Grouped Data ElementsBar Charts • Similar to Line Chart • Show how much one variable is affected by another (Correlation)Scatter Plots • More complex Scatter Plot • Shows how much one variable is affected by two others • The size of the bubble denotes value of third variable Bubble Charts • Compare parts to the whole (proportion of slice arc to whole) • Can be difficult to compare different slices of pie or across different pies • Total of values should equal 100(%) Pie Charts
- 64. Choose the Right Tool
- 65. Additional Visualization References Stephen Few http://perceptualedge.com Edward Tufte http://edwardtufte.com/tufte Jen Underwood Blog http://www.jenunderwood.com/gallery.htm Word Clouds http://www.wordle.net Chandoo http://chandoo.org/wp/excel-dashboards/ Clearly and Simply http://www.clearlyandsimply.com
- 66. Q & A Verily, this vichyssoise of verbiage veers most verbose
- 67. References
- 68. References https://en.wikipedia.org/wiki/Charles_Joseph_Minard https://public.tableau.com/en-us/s/blog/2016/08/lessons-photography-how- better-visualize-data http://www.express.co.uk/life-style/science-technology/709791/12-Black- Dots-Optical-Illusion-Ninio-s-Extinction https://xkcd.com/552/ http://vis.stanford.edu/papers/narrative http://www.tylervigen.com/spurious-correlations http://www.bl.uk/learning/timeline/large126360.html
Editor's Notes
- Non-Visual Data – A visualization’s purpose is to be the communicator of data Produce and Image – The visual must be the primary means of communication Readable & Recognizable – Must provide a way to learn something about the data
- Pre-History: Early visualizations focused on movements of stars, geography, maps, crop planting, and city development. They took the form of sand, rock and cave drawings, and later drawings on clay and papyrus. Ptolemy revolutionized map making with use of latitude and longitude to develop global coordinate system
- William Playfair – Scottish economist and politician; time series graphic showing trends in data over time This compares reigning monarchies (at top) to price of wheat (middle) to average wages for mechanics, smiths, masons and carpenters
- Charles Joseph Minard – French civil engineer noted for his contribution to Information Graphics in civil engineering and statistics Napolean’s Disastrous Russian Campaign of 1812 Represents two dimensions across six data types – Troops, Distance, Temperature, Latitude/Longitude, direction and location relative to dates Translation of text on graphic: Figurative Map of the successive losses in men of the French Army in the Russian campaign 1812-1813. Drawn by Mr. Minard, Inspector General of Bridges and Roads in retirement. Paris, 20 November 1869. The numbers of men present are represented by the widths of the colored zones in a rate of one millimeter for ten thousand men; these are also written beside the zones. Red designates men moving into Russia, black those on retreat. — The informations used for drawing the map were taken from the works of Messrs. Chiers, de Ségur, de Fezensac, de Chambray and the unpublished diary of Jacob, pharmacist of the Army since 28 October. In order to facilitate the judgement of the eye regarding the diminution of the army, I supposed that the troops under Prince Jèrôme and under Marshal Davoust, who were sent to Minsk and Mobilow and who rejoined near Orscha and Witebsk, had always marched with the army. Look at some of the things that we use today – Multiple dimensions on the graphic, combination of text and numbers, general flow of information from left to right, legend, etc.
- Prevailing wisdom was that cholera was spread by foul smelling mist “miasmas” John Snow believed that contaminated drinking water was the cause; if disease was carried on the air, instances of it would be widespread; however, if disease was caused by water, then instances would be localized. He created a map plotting London streets and incidences of cholera and public water pumps Based on this, he tracked the origin of the outbreak to the pump at Broad Street and he quickly advised decommissioning the pump. Subsequently, the cases of cholera quickly disappeared.
- Nathalie Miebach – an artist who generates the “visual articulation of scientific observations” Translates scientific data related to astronomy, ecology and meteorology into woven structures Basically creates three dimensional representation of the data on the grid of a woven basket This not what you typically think of when you hear the term visualization, but it is a type of visualization What kind of story do you think it tells? How effective is it?
- Focused on what information is needed to tell the story – Launch Angle; Batting Average and Average Exit Velocity Visually appealing Uses color (red vs. blue) to denote different meaning for some of the data points Shows relative launch angle by placement on the graphic – quickly understandable for the viewer
- Network distribution visualizations
- Stanford Researchers: Edward Segel, Jeffrey Heel
- Get your facts first and then you can distort them as much as you please Make sure that your underlying information is accurate; you can make the numbers lie afterwards You need not expect to get your book right the first time Develop, review, revise – Agile! Use plain, simple language, short words and brief sentences. That is the way to write English – it is the modern way and the best way. Stick to it; don’t let fluff and flowers and verbosity creep in Keep the data simple; do not add in too many layers of data or too many axes to the matrix Facts are stubborn, but statistics are more pliable Be careful to make sure that you do not skew the information presented in your visualization, as it is easy to make the numbers show what you want. (Example of statistical comparison from Matthew Berry)
- Don’t use passive voice; Timid writers like passive verbs for the same reason that timid lovers like passive partners. The passive voice is safe. Don’t stick with the traditional data visualization methods just because they are simple and easy; they may not be the best method Don’t obsess over perfect grammer. The object of fiction isn’t grammatical correctness but to make the reader welcome and then tell a story. Don’t obsess over perfect data quality when telling you story. Get the data quality sufficient for the type of story that you are telling You have three months. The first draft of a book – even a long one – should take no more than three months, the length of a season Time box your work (cleansing, transformation, analysis, reporting and formatting) to avoid heading into a rabbit hole from which you cannot return Write one word as a time Keep the scope of your work manageable; Do not try to present data to show the most profitable customers, products, stores, salesmen, month, region, etc. There will be too much information and it will be too difficult to accomplish all of this at one time Eliminate distraction. There should be no telephone in your writing room, certainly no TV or video games for you to fool around with Data Analysis and presentation teams should be able to focus on their work without being distracted by other time demands (work)
- A picture is worth 1000 words!
- Balance – The arrangement of elements in a way which is visually consistent Art shows a large shape close to the middle balanced by a small shape close to the edge Visualization shows the percentage of the total US population by age and Gender for a particular year (the values for each gender are shown against each other, balanced on either side of the middle of the visualization Visualization is from The Next America by Paul Taylor, April 10, 2014
- Gradation – Changing size and direction to produce linear perspective The gradation of color in the art draws the attention of the viewer’s eye down to the main subject – the person standing on the rock/mountain Visualization uses the gradation of color to show the frequency of pitch location for Trevor Bauer curve balls in 2016. The more intense the color on the heat map, the more frequently a pitch hit that area of the strike zone (from the catcher’s perspective) Visualization is from MLB Blogs (Bastian) in an article discussing the effectiveness of Trevor Bauer’s curveball in 2016 vs. 2015.
- Repetition – Presentation of the same image in a pattern; with variation is interesting; without variation can be monotonous Andy Warhol like print of Marilyn Monroe on the left (example from art) The visualization shows an arc diagram, which is a method of mapping the musical sequences within a particular song. This particular one is an arc diagram for Mozart’s Moonlight Sonata
- Contrast – The juxtaposition of opposing elements – colors, shapes, sizes, etc. Art shows how the form of the Frisbee throwing dude is visible due to the contrast of black and white and the alternating of black white in the same line (double constrast of sorts) Visualization shows Human breast lymphatic vessel (arrows) identified using CPS after intradermal injection of the microbubble contrast agent into the periareolar margin of the breast Visualization is from the Journal of Ultrasound in Medicine.
- Dominance – Using object size (typically larger size) to draw attention Art uses dominance to draw the viewers attention to the primary subject of the piece – the motorcycle rider, which is prominently placed in the “front” of the artwork and therefore draws the viewers attention Star Trek Visualization maps the relationships of every Star Trek character to other Star Trek characters based on the selected TV shows and/or movies Visualization is from FlowingData.com, a visualization blog with examples of visualizations
- Harmony – Combining similar, related elements for visually appealing result Art shows the harmony of colors, gradations, balance, dominance, etc. All of these design elements are tastefully used to make a pleasing visual work of art The Visualization shows the first 140 Twitter employees and how they relate to each other through mutual follows. Each arc shows a connection between two employees. The visualization contains balance, color gradiation, dominance and contrast put together to make the visualization more visually appealing AND easier to understand. Visualization is from FlowingData.com
- Unity – Visual linking of various elements in the work Art shows use of contrast (light/dark colors, small/big shapes), dominance (large dark spots draw attention), gradation (color shifting from lighter to darker) and repetition (repeating the linear looking shapes throughout the piece) Visualization also shows balance (visualization is blanced left to right of image), gradation (intensity of colors), repetition (vertical lines representing music notes), constrast (colors to black), etc. Visualization is from theappguruz.com and an article discussing how to build a music visualization in Unity
- Metaphor * Pre-attentive is like talking to a friend over Facetime * Attentive is like building the LTE network over which Facetime runs and then using Facetime to talk to a friend
- Your brain works with text differently than numbers. It also works with both differently than images. This is due to the effort your brain has to put into processing the text and numbers, versus what it can gain from perception.
- In order to view the information in this graph your brain must process it, since it is simply a set of numbers. The effort required to process this information takes time and therefore, it can be considered more difficult to understand.
- Your brain fills in the gaps in your peripheral vision with what it thinks should be there, assuming more grey lines The dots appear and disappear from your view of the image as your eyes move around it, based on what is seen in the peripheral vision
- Your brain fills in the gaps in your peripheral vision with what it thinks should be there, assuming more grey lines The dots appear and disappear from your view of the image as your eyes move around it, based on what is seen in the peripheral vision
- “When nothing is occupying my attention, the first thing I do is reach for my phone.” 77% of people ages 18-24 responded “Yes” 10% of people over age 65 responded, “Yes.” * Study from Microsoft Corp.
- Vertical Scaling Using large scaling on Y axis to exaggerate slope of lines Absolute Values Comparing actual numbers to percentages Narrower Context Skewing the estrapolation by filtering data out (one graph filters out past recessions, which skews the results presented in the graph)
- The first graph does not do as good of a job of showing the relationship between the actuals and budget, since the budget value is not normalized. The second graph keeps the budget value constant which more accurately compares the actual value to it.
- Jim Stikeleather, Dell Executive Strategist in Harvard Business Review Article
- Spurious Correlations
- Spurious Correlations
- Spurious Correlations