Nobel Peace Prize Winners visualisation using Tableau

It’s possible to interact directly with Tableau workbooks using Tableau Public, which is probably best described as a ‘Youtube’ for data.  This means that viewers are no longer passive consumers of data visualisations. Instead, data consumers are turned into active, interactive data consumers who can visually navigate their way, slicing intuitively through their data sets in order to make strategic decisions based on data and fact.
I’ve made use of Tableau Public here in order to try and ‘practise what I preach’ with respect to data. In other words, that data visualisations should be fluid enough for navigation, and robust enough to withstand scrutiny. Click on the image in order to go to the relevant page at Tableau Public: 

This particular image shows the sex of the Nobel Peace Prize winner as male, female or ‘N’, which represents an institution. If you go to the Tableau Public site, you can play with the worksheet and see where the male/female/institutions have earned the Nobel Peace Prize. 
As in the case of a previous blog, most of the countries had up-to-date names apart from the following:

  • Burma – is now Myanmar
  • Tibet – contentious
  • Russia – should be the ‘Russian Federation’
  • Palestine – contentious

If Tableau does not recognise the country, then it assigns it as a Null value and it appears in the ocean. This occurs with Tibet. There is a debate over whether Tibet is independent of China, and I am afraid I cannot answer that here. However, since the Dalai Lama is a peaceful man, in my opinion, I could not find it in my heart to represent him as anywhere else other than Tibet; so I’m afraid I’ve left it as a Null value in order to highlight it.
If this was Microsoft SQL Server 2008 R2 Reporting Services, we could make the most of the geographic and geometrical points stored in SQL Server 2008 R2; this would remove the obstacle around the political associations with naming countries since we would simply have an absolute point in a map. More on this on later blogs!

Tableau Public is free, so what are you waiting for? Get vizzing!

Visualisation of Nobel Peace Prize Winners

Every year since 1901, the Nobel Prize is an international award dedicated to outstanding achievements to humanity. There are prizes in physics, chemistry, physiology or medicine, literature. It is administered by the Nobel Foundation in Stockholm, Sweden.

The most commonly recognised award is the award for peace, and this year, the Nobel Peace Prize winner has been announced as Chinese dissident Liu Xiaobo. In 1968, Sveriges Riksbank established The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, founder of the Nobel Prize. Personally, I love this award since it allows us, as humans, to stand and take stock of what we do to one another, and gives us a chance to put peace in the frame as a discussion point for the world.

The ever-brilliant Guardian Data Store has released data for every peace prize winner. I have taken this data, and visualised it to show the number of prize winners per country.  Here is the image, or click here for a larger image:

Tableau Nobel Prize Winners

To do this, I used Tableau. This tool is excellent because it recognises country names. However, when visualising data, it is important to be careful if you are just relying on country names alone. For the Nobel Peace Prize winners list, this proved problematic because country names can change over time, and Tableau seemed to recognise the most recent country list. For example, Baroness Bertha Sophie Felicita Von Suttner won the award in 1905, and the country was then known as Austria-Hungary; now, it is known as the Czech Republic.

Also, countries can be known by more than one name, which can be difficult for software to negotiate.  For example, Burma is officially known as the Union of Myanmar, and East Timor is officially known as Timor-Leste.In this case, Tableau didn’t recognise ‘Great Britain’ but did recognise ‘United Kingdom’. So, you have to be very careful of your data; the software can only do so much for you, and you need to double check. Fortunately, Tableau has a ‘View Data’ option which allows users to double-check. A simple tip in Reporting Services is to add in a table which shows your base data, and then remove it before you complete the report. This will allow you to double-check that the report is really showing that which you would like it to show.

Additionally, country names can be deeply contentious; just look at the Middle East for some examples there. No software in the world is going to be able to resolve this; you have to be clear about whether you are using one country name or another. For this visualisation, I replaced ‘old’ names with the most up-to-date country names, and replaced the informal names given in the original data  (e.g. Burma) with the official name (Union of Myanmar).

To summarise, if you are using geographically-based display if data, be careful f you are relying on country names.  Be sure to check whether the country name is picked up properly. Tableau and Reporting Services can both use Longitude and Latitude, and Reporting Services 2008 R2 can use SQL Server 2008 Geography and Geometry data types for absolute certainty. More on this at a later point!

Add to Technorati Favorites

Using WMS Servers to load non-US maps

Residents of the United States are lucky; then get free US maps, with lots of US-specific information, which are loaded easily into reporting software. Tableau provides the US maps by default, and Microsoft have released this functionality in SQL Server 2008 R2 so you can display your data efficiently using the new geometry and geography data types.

However, what about the case where you don’t want to see United States data?

This question was posed to Microsoft team members at last weekend’s SQLBits. Namely, when can we get free maps for the rest of the world? Unfortunately, Microsoft have no plans to provide us with non-US specific information at this point in time. The reason is that US borders and boundaries are very well-defined and there are no conflicts regarding territory. However, for the rest of the world, it isn’t that straightforward. In some countries, it is possible that you may even go to prison for drawing incorrect borders, which are perhaps being fought over. So, drawing maps is much more contentious than you might think. Microsoft quite like their users, and decided it was less contentious to provide maps, preferring instead to allow you to make and create your own.

In the meantime, you can load your own ESRI shapefiles to both Tableau and SQL Server 2008 R2 Reporting Services. I saw the latter at SQLBits last week at a presentation by Andrew Fryer, and I was very impressed with the mapping capabilities that Microsoft now offer.

If you would like to use WMS Servers in Tableau, this is also easy. Here are some publically accessible WMS Server URLs listed here. Please note that you should check the servers’ usage policy before using it; just because it is publically accessible does not mean that we’re automatically allowed to use it!

This URL belongs to NASA, but be aware that your connection may be refused if it is too busy!
It is also possible to use the World Mineral Deposits service:
In Tableau, it is easy to load in the background maps. Here are some images to help you:

1. Find the WMS Servers dialog box

2. Type in the URL

3. Choose your map!

I hope that this helps someone, somewhere. If I get time to look into this more, then I’ll post.

Quick Geographical Mapping using Tableau

I’ve become interested in geospatial mapping features of SQL Server 2008, and how these can be used to benefit users’ understanding of their data. As a starting point for teaching myself, I used Tableau to display some ‘Complaints’ data, using Longitude and Latitude in order to put points on a map. In order to understand more about the SQL Server geometry and geography datatypes, my next step will be to expose some data using the new data types.
While I’m learning, I thought I’d apply Tableau to UK data so that there is a non-US-centred example. So, the attached video shows how easy it is to put data on a map. The video shows the number of complaints made, broken down by specific postcode area. The size of the circle represents the number of complaints in a specific area, and the colour refers to different postcodes within the area of Scotland that we’re looking at.
It’s the first time I’ve tried something like this, so I’d be interested to know what you think!


Add to Technorati Favorites