Month: March 2017

Living in silicon valley means you will be surrounded by numerous start-ups. But how many of these startups actually succeed? Here success is identified as crossing the ‘Billion Dollar’ mark. Most of the companies are venture-backed private companies. The claim of this visualization is to show how the Billion dollar start-up boom happened after 2014.

The visualization tracks all the start-ups who touched the billion dollar mark and how they went beyond that in a span of 2 years. The visualization defines a semi-circle of 3 radii. The first one marks the billion dollar mark, the second marks $10 billion and the 3rd one marks $40 billion. The start-ups across 3 major geographic locations are marked, viz., United States, Asia, Europe, and others. The visualization being interactive has a time slider below which shows dates from the year 2014 to 2017. Also, you can differentiate the companies on the basis of the industry. The drop down list shows different types of industries through which you can drill down to a specific industry or you can look at all the industries at once.

The visualization strongly supports the claim to show the boom of startups after the year 2014. The idioms used could have been a bit different so as to show the name of the companies extensively but the main claim of the visualization is to show the ‘Boom’ which appropriately displayed.

 

Reference: http://graphics.wsj.com/billion-dollar-club/

In 2009, great recession overwhelmed the United States’ economy. This data visualization clearly shows the performance (jobs number and salary number) of each industry in this ten years.

Visualization Link

The first visualization gives a big picture of the overall domestic performance and gives the overview of the performance of every industry in the United States. It uses color to categorize how industry performs and also gives the details information of ten years performance of each industry.

1. The second visualization highlights the data from the low-paying job such as fast food and high-paying job such as consulting. It finds that both sectors are growing.
2. The third visualization highlights medical industry and especial those in middle-wage industries.
3. The fourth visualization displays Housing industry. It is seriously suffering until now.
4. The fifth visualization shows the manufacturing industry. It compares each sector with another.
5. The sixth visualization shows oil industry performance. Almost all of sectors in this industry grow dramatically.
6. In this ten years, we also face digital revolution. The seventh visualization compares traditional media industry with IT industry.
7. Last but not least, it shows the hottest and fast growing industries’ performance in the United States.

This Visualization did a great job at grouping the data, using color to classify the performance, and arranging, for example, it put poor performance industry at the lower part. Moreover, it tries to look at the same data from different angles and makes story interesting.

Reference: https://www.nytimes.com/interactive/2014/06/05/upshot/how-the-recession-reshaped-the-economy-in-255-charts.html?&_r=2&abt=0002&abg=0

 

The main task of dashboards is to provide the insights. Key Performance Indicators (KPIs), as the name suggests, help us to know how a business is doing or how is a specific process doing. This usually involves comparing current values or historical trends against a target value.

I have two methods to point you towards the blogs, for building KPI dashboards:

1. Tableau provides dynamic shapes to show good news/challenges. The upward pointing green triangle is shown for any good news/success whereas the downward pointing red arrow reflects failure or a lower result from the last one.

It’s very simple to generate these in Tableau and the steps are explained on the website:

http://www.thedataschool.co.uk/emily-chen/building-kpi-dashboards-with-shapes/

2. This method involves making two views – one showing the current values, the other showing the historic trend (in this case, revenue). This post explains how to align these two views together or how to deal with them by interlinking the two separate sheets on a dashboard.

https://www.interworks.com/sites/default/files/RobertKPI1.png

This is explained on the following website:

A New Way to Visualize KPIs in Tableau

Personally, I liked both. It totally depends on which one you choose to present your research and insights.

References:

A New Way to Visualize KPIs in Tableau

http://www.thedataschool.co.uk/emily-chen/building-kpi-dashboards-with-shapes/

Original Visualization

The visualization shows the data for surplus singles in the US. Gender is shown by the color and the size of the circle determines the number of surplus singles. This chart just shows the number of surplus singles, I believe the audience would be confused of how big or large the number is if they do not have an idea about the population of the city. There is no different chart for different age groups to have targeted claims.

Redesigned Visualization

This visualization was created to overcome the above disadvantages. The visualization shows the number of surplus singles for every 1000 instead of the actual number, this gives the audience more background of the “how many extra” singles. This gives a nice and clear information to the audience. This visualization also has Slider using which the audience can select the age group they are interested in, this would make the claim to be more targeted.
Usage of map, color and size of the circle is a very easy way for the audience to grasp what the visualization represents. By redesigning the visualization the author has solved the problem the audience had with the original visualization i.e. understand how many more unmatched for 1000 rather than just the entire unmatched singles.

We’ve learned the nine principles of how to make effective visualizations and applied it into our exercises.

Today I’d like to share what I have learned from aesthetics session and how important that is to understand human perception when it comes to creating a good visualizations.

Bellow questions can answer the aesthetic lecture in different angle.

1. Does it meet physical attributes of vision? Our vision sensitively recognizes length, position, size, shape and color of views. Those determines how we think and how we decide. Sometimes pie charts, donut charts can’t fully represents the attributes of vision. Understanding attributes of our visions can help us to have better design when it comes to visualization.
2. Does eyes can discern and brains can understand? First our eyes see and then our brain understand. If the visualization is not easy to understand, not clearly indicates the relationship of the data then our brain can hardly understand what we see.
3. Does our seeing In-Balance with our thinking? In class professor taught eye beats memory principles. Our brain has limitation of working memory. If what we see is not balance with what we thinking then the visualization can’t convey the accurate message that it suppose to deliver.

Reference

Data Visualization for Human Perception

https://www.interaction-design.org/literature/book/the-encyclopedia-of-human-computer-interaction-2nd-ed/data-visualization-for-human-perception

 

 

Visual exploration is dealing with open-ended data-driven visualizations that needs experts like Data Scientists, and Business Intelligence analysts. Although new tools have begun to engage general managers in visual exploration. It’s exciting to try, because it often produces insights that can’t be gleaned any other way.

During this exploration we don’t know what we are looking for, these visuals tend to plot data more inclusively. In extreme cases, this kind of project may combine multiple data sets or load dynamic, real-time data into a system that updates automatically. Statistical modeling benefits from visual exploration.

Exploration also lends itself to interactivity: Managers can adjust parameters, inject new data sources, and continually revisualizes. Complex data sometimes also suits specialized and unusual visualization, such as force-directed diagrams that show how networks cluster, or topographical plots.

Skills like analytical, programming, data management, and business intelligence are more crucial than the ability to create more presentable charts .These skills are crucial for managers to help setup systems to wrangle data and create visualizations that fit their analytic goals, and therefore it mostly needs experts to create visualizations.

Source: https://hbr.org/2016/06/visualizations-that-really-work

This visualization reveals the secrets of animal trading around the world in 2013. It was created by National Geographic and Fathom Information Design, which believe the visualization is useful for researchers and could helping policy makers view animal trade in a different light.

The visualization uses packed circles, which is similar to the classical idiom of Bubble Chart, but the difference is it has interactivity. The data is shown in a hierarchical order, with large animal groups, such as bird and mammals shown initially. When you click on circles, it will drill down to specific species.

The graph only uses one mark and three channels. Each mark presents a specific specie, and surrounded by larger mark which represents a specific group of species. The channel of size encodes the volume of trade and the color is to differentiating the animal species. The channel of hue is applied to represent the purpose of trade. However, one important information, the change rate of trade was not encoded here. I believe that element will show up in its next version, which could provide their audiences with more insights.

Reference:

http://news.nationalgeographic.com/2015/06/150615-data-points-infographic-animal-trade/

 

Donut charts and pie charts are very similar and some would even say they are the same type of chart. I am in the camp that says that they are the same and therefore I have approached donut charts the same way I have approached pie charts. I do not use it.

The things about donut charts and pie charts is that pie charts are in actuality better than donut charts. Donut charts are basically pie charts with a hole in the middle. While this might not seem that big of a deal, it however reduces the amount of information that you are seeing and makes it even more confusing. For example in the figure above, by removing the center of each circle, the chart is presenting the information with only a tiny portion of information. It becomes harder to see the size of each section and how they compare with each other. Of course while the information about each sections size is represented as a number, this could have easily been done in a bar chart instead. So for future references, stay away from all charts that have a name related to baked goods. Its far more satisfying to eat them then to use them as a visualization tool.

Reference:

http://www.vizwiz.com/2012/06/donut-charts-are-worse-than-pie-charts.html

 

The data visualization landscape can be divided into two broad categories:

Hedonistic Visualization: It only shows how cool something is or represents an “Interesting to know”  information. Here is an example – 

https://fivethirtyeight.com/features/when-donald-trump-attacks-gop/

Narrative Visualization: It supports a narrative (often journalistic, sometime scientific). Data Journalism is all about telling a simple story in the most attractive way to entertain the audience. But the problem here is there is a high chance of over exaggeration to bring out an interesting story. While data journalism is a tough art, the other extreme of narrating a scientific story is harder. In order to demonstrate the hypothesis scientists have generated, they face the challenge of where to stop overloading the audience with too much information. Do they have an option? Will the scientists be able to convince the audience if they condense their visualizations?

How about Problem Solving Visualizations?

This is a less commonly known category that is gaining importance. Singapore’s MRT Circle Line was hit by a spate of mysterious disruptions in recent months, causing much confusion and distress to thousands of commuters. Data scientists at GovTech’s Data Science Division in Singapore used visualization to discover the origin of this recurring problem. The below article discusses about the interesting case:

https://blog.data.gov.sg/how-we-caught-the-circle-line-rogue-train-with-data-79405c86ab6a#.sebeshx7o

Important point to note here is for most of the problems in automated systems, solutions really need to come from human reasoning(in this case through a series of visualizations). Hence visualizations are very powerful problem solving tools.

References:

http://fellinlovewithdata.com/

https://blog.data.gov.sg/how-we-caught-the-circle-line-rogue-train-with-data-79405c86ab6a#.sebeshx7o

When Trump Attacks!

 

 

CrunchBase is one of the most widely-used databases of technology companies, people, and investors.  Financial companies and entrepreneurs use this dashboard to analyze investments, acquisitions, and start-ups all in real-time, and make the smartest investments possible.

Good points about the dashboard.

1. The dashboard conveys all the important KPIs for the audience (Investors) in a very simplified manner.
2. It solves the purpose of being interactive. Investors can filter by country, categories, type of investment, investment rounds, etc for global crunchbase data in a comprehensive manner.
3. It has a summary on top of the dashboard which means the user gets an overall view of what is happening in the data. For e.g. Here, the important factors like total money raised, the number of investments, Median raised and No. of startups and their percentage growth which gives us an overview of the yearly change in the data.
4. One color legend for the entire dashboard.

Improvements:

1. Instead of representing the Funding by round graph in a pie chart, the user can use a different idiom like a bar chart which gives a comparative idea for each funding by round.
2. Instead of mentioning the color legends at the bottom, the user could have defined color legends at the right-hand side so that the user of the dashboard is not required to scroll every time he wants to see what color is represented by what category.

 

Reference:  https://www.sisense.com/dashboard-examples/investor/