Data visualization is nothing but
“How to tell stories with Data” or “Helping audience to understand, which sometimes otherwise difficult to see” or “Capturing an issue and delivering visual representation so that audience can connect immediately ”
Today all content lives in the Mr. X’s model of Universe starting from You-tube, Face book, twitter, Instagram, LinkedIn to mail boxes (Gmail, Yahoo, Outlook, Rediff etc.) getting the data is not important, getting the insight is essential, getting the insight is not enough, visualizing it with accuracy, clarity and integrity is the ultimatum.
As per Eric Schmidt “We now create more content every two days, that must created from the dawn of Civilization to 2003” & by the way at this moment of 2015 it’s every single or half of the day now. Now in this age of data overload where people are empowered by choice and access to information online or offline, one question arises which is the accuracy of collected information.
So few things along with accuracy i.e. Visualization, Storytelling, Interactivity & share ability are vital while channelizing the audience.
Coming to audience “Culture”, “level of expertise”, “Consumption of context or channel”, “Accessibility”, “True believers or skeptics” and “action” are the six basics to note down while channelizing.
Make sure audience need to know the underlying story which you talking about, otherwise provide them enough background to connect to the topic. Digital story telling is all about visualization and audiolization.
10% of people are color blind (Most are red and green color blind). They find difficulty in differentiating the stock market’s UP price (Which indicated by green) & DOWN price (Which indicated by red). Before going to finalize the color pattern I will suggest to upload and check your chart how it is visible to normal eye, Red- Blind, Blue- Blind, Green-weak etc. on www.color-blindness.com
Contrast and font size is must to be checked (For 40 year aged audience font size must be 11+, if college kids don’t worry about much. Still suggest to take care of it).
Gestalt’s Principles on Psychology on how we perceive visual information & these are http://graphicdesign.spokanefalls.edu/tutorials/process/gestaltprinciples/gestaltprinc.htm
- Figure/ ground
- The eye differentiates an object form its surrounding area. A form, silhouette, or shape is naturally perceived as figure (object), while the surrounding area is perceived as ground (background).
- Balancing figure and ground can make the perceived image clearer. Using unusual figure/ground relationships can add interest and subtlety to an image.
- The most often used logo FedEx: A forward arrow mark subliminally motioned.
- In data visualization back ground is used as least aspect while the foreground the max.
- When we have items next to each other, they are perceived as grouped/ similar/ together
- Similarity occurs when objects look similar to one another. People often perceive them as a group or pattern.
- Similarity in terms of size, color & shape.
- Common fate
- When all dots are random positioned and some are moving towards one direction then moving dots are grouped different from the static ones
- Closure occurs when an object is incomplete or a space is not completely enclosed. If enough of the shape is indicated, people perceive the whole by filling in the missing information.
- Continuation occurs when the eye is compelled tomove through one object and continue to another object.
Understanding (Your) Data
Data should be represented with less mistake, more accuracy & compelling content with 3 basic math/ stats literacy
- Mean vs Median
- To use median when outliers are present
- Actual vs rank index
- Comparing GDP and Entrance exam results
- Comparing GDP and Entrance exam results, makes a rough idea about total no of contestants.
All visualizations at the end are Explatory (Either Explanatory or Exploratory / both). Consider the following marathon data chart which explains all things at the same time able to explore minute details of each runner by searching by name or moving the cursor on the dots. Which tells the gender, age distribution, finish time etc. amazing right.
At the end of every chart I would recommend to give something for the audience to explore something which will make our story theirs.
The 6Ws (What, when, why, where, how & who) which I feel are the great ways to organize and think about any story. In data visualization you don’t need to use all 6 at a time, but some are most important ones for the particular study. For an example suppose consider. The case where can I go for a specific treatment at a decent price and good quality. Applying 6Ws (Logical questions). I have “who, what & where” where who is the granular answer, where is for which city, which hospital & what is for the conditions. (What -> Where -> Who).
Sometimes you need to look what’s missing in the data or what’s wanting may be a column, may be a question missing from the survey. Just take care of that.
Explore your data
You need to know your data first before going for visualization. Tools like Ms. Excel, Tableau, QlikView, Gephi (Good form of network visualization i.e. nodes and links), for geographical data go for Google Maps, Map box, Carto-DB are the best visualization tools where programming knowledge is less needed.
Convert your data (The inevitable step you should always anticipate)
Common data adjustments before story making are calculating indexes and ratios, Calculating percentiles, Aggregating, Re-grouping, Converting from Excel/ csv to JSON/XML/My SQL as per requirement.
Sketching & wire framing (Analog part of data visualization)
There are reasons why to go analog before going to digital.
- Speed (When I use pen & paper, white board I can go very fast with my ideas)
- Flexibility (I can modify quickly with as many experiments as I can/ I can push my envelope)
- Scale (Which is not possible in small screen like a computer)
- Good body-mind connection (Very comfortable)
All are about brain storming. Point to go for analog is that its natural and it reduces the layers of things between me and my idea. I can move quickly, can generate as much ideas as possible. I can comfortable on it and can’t go wrong easily.
4 basic type of types, which helps audience what type of information audience are looking at
- Axes and legends (Should be in short text, not too much of stories)
- Call outs (Indicative bench mark in charts)
A best saying “The new graphic designer no longer creates visualizations by choosing a rigid collection of shapes, positions and colors but rather by choosing the rules needed by data to breathe form into geometric abstractions”
Position, Color, Contrast, Shape & Size are the 5 ways to show the differentiation between different data points or observations.
It’s always better to provide notes on the data and notes on the technology.
In the end
While entering data visualization most important which chart to use at which situation. Before you can walk you need to learn how you can crawl, like that you need to know the basic charts before going for alternative or hierarchical charts.
(Basic charts includes Bar charts, Line charts, Area charts, Time lines, Scatter plots, Bubble charts, Pie charts. Alternative charts includes Box plots, Heat maps, Radar or spider chart, parallel co-ordinates, Scatter plot matrix etc. Hierarchical charts include tree diagram, Node link diagram, Fish eye distortion, Jason-C matrix, Tree map, Core diagram, Venn diagram with opacity etc.)
Provide your comments below
If you loved this story, do join our Telegram Community.
Also, you can write for us and be one of the 500+ experts who have contributed stories at AIM. Share your nominations here.
Analytics learner, believer and follower having one year analytical experience as a Business Analyst. An Electronics & Communication Engineer holding a master degree in "Business Analytics & Retail" from Praxis Business School. Who believes Today all content lives in the Mr. X’s model of Universe starting from You-tube, Face book, twitter, Instagram, LinkedIn to mail boxes (Gmail, Yahoo, Outlook, Rediff etc.) getting the data is not important, getting the insight is essential, getting the insight is not enough, visualizing it with accuracy, clarity and integrity is the ultimatum.