Posts by Louis Keating
The Novel Coronavirus - a Pandemic Disease Visualised
 

The Novel Coronavirus (2019-nCoV) has caused global alarm as the originally defined pneumonia cluster found in China has spread beyond its original source into upwards of 25 countries.

To keep you informed with the latest statistics on where it’s being diagnosed and how it’s being contained, the team at White Box Analytics have extracted the most up-to-date World Health Organisation data on the virus, providing you with an updated source on the numbers behind the virus.

 

The visualisation paints a picture about where the coronavirus is concentrated, as well as the countries it has spread to since its identification in early January. The map of China sheds light on the high number of occurrences in Hubei, where the Wuhan seafood and live animal market thought to be the source of the virus, is located. Latest updates of WHO data shows this number is now in excess of 27,000 diagnoses in Hubei alone.

The trends and incremental change component of the visualisation further shows the exponential-like growth in the spread of the virus, acting as an alarming warning sign for the world, particularly those who have or are likely to travel to majorly affected locations. The WHO is however implementing a number of strategies to combat the virus that is yet to have a vaccine or proven treatment (see below).

Update 19th February 2020

This last week has been very fascinating in terms of how the Coronavirus is both being diagnosed, and how it is impacting the societies and economies of China and those related to it.

An issue with the virus has been that it has taken approximately two weeks to diagnose (using medical diagnostic tests) each person affected. The problem with using the diagnostic test as the sole measuring tool is that, due to the huge number of people picking up the virus on a daily basis, hospitals have not been able to test all people likely to have picked it up. By changing the diagnosis criteria to include those people who have been clinically diagnosed, many reports have shown a huge spike in the number of new diagnosed cases. Take, for example, this John Hopkins University chart showing new diagnosed cases, using the new criteria.

John Hopkins University Data

The jump reflects the first day the new criteria came into place. Despite this rise, however, the World Health Organisation have not changed the way they are recording the number of diagnoses. Looking at our visualisation now, we have updated it so you have the option to filter between clinically diagnosed and laboratory confirmed cases, as well as total cases which includes both. This paints an important picture about the way data is presented in the news, and in business projects; we need to be consistent with the way we use data, and when changes are made, stakeholders need to be informed so that there aren’t misconceptions about its implications.

In terms of its impact on the Chinese economy, Chinese e-Commerce giant Alibaba has rolled out a number of measures through its retail platform Tmall, to minimise the impact of Coronavirus on businesses using its marketplace. These measures include providing a free set-up tool (to launch on the platform), cutting annual service fees for the first six months of the year, and lowering automatic settlement, interest and trading related fees (source).

Despite these efforts however, the impact is having a huge effect on both China and the rest of the world. A New York Times update today (19/02/2020) has reported that more than 10% of China’s population are confined to their homes. This confinement has taken its toll, leading Hong Kong bank, HSBC, to plan to cut up to 35,000 jobs with reported losses of $4.5 billion (source).

Information about the virus

Origin -

It has been reported that the likely source of the virus was the Huanan Wholesale Seafood Market in Wuhan (source).

Symptoms -

The coronavirus exists in a number of forms and is associated with respiratory and gastro-intestinal issues often associated with a common cold or the flu. These symptoms vary from mild to moderate.

Current strategies for prevention -

Education through online training - informative articles and video tutorials about the source, spread and warning signs of coronavirus. By identifying the signs and symptoms identified above, people can help prevent the further spread of the virus.

Accelerating research and innovation - The WHO is playing a coordinating role by bringing the scientific community together, identifying key research priorities. It will produce a global research agenda and a framework to decide which projects are undertaken first. The aim will be to fast-track development and evaluation of effective diagnostic tests, vaccines, and medicines, and to bring affordable access to vulnerable populations, leveraging community engagement.

Expenditure for global preparedness - a US$675 million preparedness and response plan has been developed to cover strategies to combat the virus between February and April 2020.


For the latest updates, visit this page or follow White Box on LinkedIn.

 

For more fascinating visualisations and data stories, click here.

 
VisualiseLouis Keating
Brexit and the next James Bond; how do they mix? - Visualisation

How do Brexit and the next James Bond character relate? Ije Ireumi’s Monday Makeover chart from a YouGov survey reveal Brits who voted to leave the EU do not wish to accept a non-british nationality, gay or female person playing the next James Bond character. Did you spot any other interesting insights in this data visualisation?

 

For more fascinating visualisations and data stories, click here.

To keep up with all things data and White Box, follow us on our LinkedIn page.

 
VisualiseLouis Keating
The man who got rich on data - years before Google - Article Analysis
how data analysis developed into the booming industry it is today
 

The man who got rich on data - years before Google is a time-lined walk-through of how a man called Herman Hollerith came across a flawed process, that being the government’s ability to capture, sort and derive value from increasing amounts of census data, and came up with a solution that has changed the landscape of the modern economy forever.

In 1880 Hollerith developed the first tabulating machines which were first used in the 1890 census, saving millions of dollars and enabling the government to discover far more powerful insights from the information it was capturing.

The backbone of the story is based around the common and now frequently used saying in data related content: data is the new oil. Harford both agrees and disagrees with this statement. Data is like oil in that when it is crude and unrefined, it isn’t of much use to anyone. However, once it is refined, unlike oil, data can be used more than once to power whatever it is being used for.

He attributes this fact to the reason why tech companies like Alphabet, Alibaba, Amazon, Facebook and Tencent have grown to become 5 of the 10 biggest companies in the world, which was once dominated by oil companies. The way these companies have achieved success is dissimilar to that of Hollerith’s Tabulating Machine Company, which eventually went on the become IBM. This is because we now produce data in everything we do, meaning that compared to the 1890s and early 1900s where data was produced in much stricter and refinable formats, the most successful businesses are the ones who know and understand what data is valuable, and why.

Now we know this is much easier said than done, but is the reason we believe all companies who want to remain competitive over the next year and next decade need to carefully consider whether their data refining systems are producing outputs that drive effective decision making.

In essence, what’s the point of having a data strategy if the data strategy isn’t giving you outputs that create value for your business and customers?

Check out the original story here.

For more data analysis and visualisations, click here.

Or, get in touch for a discussion about your data strategy.

 
Data is the new oil. But is it really?
Data is much more complex and is constantly being used for different purposes
CommentaryLouis Keating
The best Christmas movie

One of the joys of Christmas is movie watching and without doubt, I’ll have the same conversation I do every year when I dutifully select Die Hard for everyone’s viewing pleasure and someone will say “but this isn’t a Christmas movie?”.

So, I was excited to see an article by the blogger Stephen Follows (a film researcher with a love for data) that goes into extensive detail to test what a Christmas movie is by using data. So what better way to spend some pre Christmas party drinking time than to brush up your defence for putting on what I’ve titled above, is the best Christmas movie.

Die Hard analysis_1.JPG

He breaks down his analysis into understanding the creative elements, the commercial strategy and the cultural impact. There is a lot here, so I’ve pulled out some highlights for your ammunition.

What I like about this analysis is the diversity of the data Stephen has found, from looking at the above video analysis, the prevalence of Christmas songs within Christmas movies to the spread of Wikipedia pages views by month (December being the key indicator, see below).

Die Hard analysis_2.JPG
Die Hard analysis_2_3.JPG

He finishes with Google Trends and takes the time to annotate the peaks, which again indicate the Christmasness of Die Hard.

One of my favourite quotes to take from this from one of the two credited writers of the film, Steven de Souza has publicly declared “If ‘Die Hard’ is not a Christmas movie, then ‘White Christmas’ is not a Christmas movie”:

Unlike White Christmas, Die Hard took place entirely at Christmas, featured a Christmas party and the “Christ-like sacrifice” of John McClane walking on broken glass.

Hope you enjoy watching Die Hard at Christmas and please take time to read the original article here.

For more data analysis and visualisations, click here.

Or, get in touch for a discussion about your data strategy.

CommentaryLouis Keating
Sydney lock-out laws interactive map
 

Methodology

We looked at four key crime offences (domestic and non domestic violence related assault & offensive language and conduct), from 59 months pre lock-out (Mar-09 to Jan-14) compared to 59 months post lock-out (Feb-14 to Dec-18).

Data from Bureau of Crime Statistics and Research.

Highlight zoning includes the lock-out zone, proximal displacement and distal displacement areas.

We have also contextualised by showing the NSW overall trend.

 

Interactive map

Hover over any suburb to see the volume of crimes before and after the lock-out laws were introduced.

Use the filters at the top to highlight zoning and see specific crimes.

You can also pan and zoom in/out using the options in the map (top left).

If viewing on a mobile device, you can also use this link to view the desktop version.

 

Analysis

Using this tool and data, we have summarised the findings in this analysis.

If you’d like to discuss how we can help you use data for insights, get in touch or look for more inspiration here.