Graph Redesign
Introduction
This post discusses Graph and how to redesign it using cognitive design principle. 2 Graphs are taken which shows the Economic and Human impact of Natural Disasters from 2005 to 2014(Explainer: Are Natural Disasters on the Rise?, n.d.) and they are briefly discussed in this paper, their strengths and weakness and best ways to redesign them as per cognitive design principle.
Graphs to Redesign
Graphs are taken from “https://theconversation.com/explainer-are-natural-disasters-on-the-rise39232”(Explainer: Are Natural Disasters on the Rise?, n.d.). They discuss the Economic and Human Impact of Disasters from year 2005 to 2014. The data for the graph is downloaded from “https://public.emdat.be/data”(EM-DAT Public Data, n.d.).
The Economic and Human Impact of Disasters from 2005 to 2014
First Graph (Explainer: Are Natural Disasters on the Rise?, n.d.)
This graph shows damages in $ Billion, number of people affected in millions and number of people killed from year 2005 to 2014.
Strengths:
1. Graph is easily understandable.
2. Numbers are written which helps to understand exact value of damage for a year.
3. Total is given to summarize the damage of 10 years
Weakness:
1. Area is used to show how big is the value which is not a good way of representing value as per cognitive design principle.
2. Pie Chart is used which is not a good to showing result.
3. There is no x-axis.
4. There are too many small texts which is difficult to read.
5. Value shown is not matching with the database, possible reason can be database was updated after the graph was created.
Top 10 countries with most disasters, 2005–2014
Second Graph (Explainer: Are Natural Disasters on the Rise?, n.d.)
This Graph discusses top 10 countries with highest number of disasters and damages done in $ Billion in those countries.
Strengths:
1. Numbers are written which helps to understand exact value.
2. Bar graph is good way of representing value as per cognitive principle.
Weakness:
1. It uses area to represent how big the value is which is not a good way of representing information as per cognitive principle.
2. Value shown is not matching with the database, possible reason can be database was updated after the graph was created.
Redesign of graph
Objectives
The objective of Redesigning is to overcome the weakness of Graph using cognitive design principle. The data for the graph is downloaded from “https://public.emdat.be/data”.
All Graphs are redesigned in Tableau.
Yearly Damages
First Try at Redesigning Figure 1
This Graph represents yearly damage, the year which have highest damage will have the highest area and also it will have the darkest color and lowest damage will have lowest area and lightest area. This graph uses area and color which is not so good way of representing information, therefore this is not a good redesign of graph.
Impact of Natural Disasters from 2005 to 2014
Final Redesign of Figure 1
It is a side-by-side horizontal grouped bar graph. This Graph represents yearly damages done in $Billions, number of people affected in million and number of people died in thousand. Years are in ascending order from 2005 to 2014 and for each year damage, people affected and people died is
plotted in same order for all year and legends is also mentioned in the right side and in the same order. This graph uses length to represent value which is a good way of representing value as per cognitive design principle.
Country wise Damages
First try at Redesigning Figure 2
This Graph represent country wise distribution of damage, darker color has higher damage and lighter color have lesser damage. It uses color to represent damage which is not a good way of representing information as cognitive design principle. Therefore, it is not a good redesign of graph.
Top 10 countries with most Natural Disasters
Second try at Redesigning Figure 2
This is row labelled plot diagram. It shows top 10 country with highest damages in $billion and its corresponding values. It uses position along a scale which is the best way of representing information. For better representation row banding is done and data is arranged in descending order. This is a good graph, it can be further improved.
Top Countries with most Damages by Disaster Type from 2005 to 2014
Final Redesign of Figure 2
This is row labelled plot diagram. The graph is divided by disaster type and for each disaster top countries are shown with highest damages. Damages are plotted using point and information is arranged in descending order per disaster type and row banding is also used. Information is
represented using position along a scale which is the best way of representing information.
Top Countries with most Damages by Disaster Type from 2005 to 2014
Improvement of Figure 8 by Richard Sigman
This graph is improvement of Figure 7 done by Richard Sigman (Linkedin) really appreciate it thank you for your guidance.
Software Solution
In Tableau, there is a functionality of creating parameter and creating a calculated column using that parameter. This functionality is used to create a dynamic solution for Redesign Graph 1. A calculated column was created with values of damages, people affected and people died. Using this dynamic selection of a field can be done and 1 parameter can be seen at a time.
Cognitive Testing
Questions are for Redesign graph 1.
Responses are given by Swetha Annapoorna (Linkedin). Thank you for your input.
Question 1. What Information you are able to gather, after looking at above image once?
Response 1. The graph above depicts a bar graph representation of the effects of natural disasters from 2005 to 2014. The year 2010 appears to have seen the most damage, effects, and deaths.
Question 2. Do you think showing all 3 at once (Damages, People Affected, Death) gives better overall idea or it should be shown 1 at a time?
Response 2. The graph depicted is an excellent method of forecasting the effects of a disaster. However, I believe that having a separate graphical representation for each effect will aid in comparisons.
Note: This is discussed in software solution section.
Question 3. Is it confusing to read graph because of different power of 10 (Thousand, Million, Billion) for death, people affected and damages
Response 3. The graph is simple to read for various effects in powers of ten. Labelling each bar with the impact value would also make it easier to read and avoid confusion.
Conclusions
Graphs are used to represent information in efficient way, but not all graphs are easy to understand. Graphs that show Economic and Human impact of Natural Disasters are studied in this paper and improved those Graphs using Cognitive Design Principle. Cognitive-based Data Visualization Design Principles can be used to improve quality of Graph and make it more efficient and easier to understand.
References
EM-DAT Public Data. (n.d.). Retrieved October 5, 2022, from https://public.emdat.be/data
Explainer: are natural disasters on the rise? (n.d.). Retrieved October 5, 2022, from
https://theconversation.com/explainer-are-natural-disasters-on-the-rise-39232