Data Journalism at The Guardian

This week we will look at what data journalism looks like at The Guardian. At The Guardian journalism used to just be about words and carefully crafting those words to create imagery for the reader. However, now data journalism has revolutionised the way journalists tell stories. Data journalism means that you are not just confined to text, but you can tell a story through images as well. Journalists are the least trusted by the public, however by being able to show their findings through data visualisations it gives their stories more credibility. The Guardian are pioneers for data journalism and define data journalism as just journalism as they realise it has changed the journalism world.

At The Guardian, journalism didn’t exist until 2009, however it is something they have always been trying to tackle from the beginning. The Guardian have always been displaying data however technology has changed how it is displayed.

During the London Olympics, The Guardian created a visualisation to show the medals won by each country. However, unlike other medal charts at the time The Guardian combined things like population of the country, team size etc to allow for greater comparison. This created discussion about things like “why one country was better at running than another?” and how economics affected this. Data journalism allows for endless content and when made interactive it allows people to explore the data themselves and create their own conclusions.

Most important aspect and why? 

I found the most important part of this lecture was understanding how data journalism and data designers work together to create stories and inform the public of information. I found it really interesting that data journalism has become the new journalism, that using imagery to tell stories is now the norm. This was important to learn as it provided a reference point to study in The Guardian – to learn how they do things and how they interpret data. However, it was also important as it gave greater weight to the field of design and that the industry of journalism now works more closely with designers to tell their stories.

References

The Guardian. (2013, April 8). What is data journalism at the Guardian? [Video File]. Retrieved from https://www.youtube.com/watch?v=IBOhZn28TsE

The Guardian. (2013, April 8). History of data journalism at the Guardian [Video File]. Retrieved from https://www.youtube.com/watch?v=IBOhZn28TsE

The Guardian. (2013, April 8). Data journalism in action: the London Olympics [Video File]. Retrieved from https://www.youtube.com/watch?v=WyjBJzigm0w

Historical & Contemporary Visualisation Methods: Part 2

In Part 1 of Historical and Contemporary Visualisation Methods we looked at historical examples where data visualisation has been influential in this field. We left with the question – “Why visualise?”

This post will show an example and discuss why visualisation design is more than designing something pretty but is used to provide the reader with insight and understanding.

Alberto Cairo, in his book functional art, discuses visualisation design and looks at an article he read on the worlds population. The article discussed a conflicting view about the poor/rich fertility rate. The author suggests that the fertility rates of the poor and rich will merge and the population will stabilise. Whilst, many views claim that the poor fertility rates and the rich will decrease, blaming the poor countries for overpopulation.

Ridley, M. (2010). Percentage increase in world population. The rational optimist: how prosperity evolves [Image]. Retrieved 22 August, 2018 from http://www.qsm.com/blog/2013/whats-story-your-data-0

The articles uses one graph (above) to support the claims. Whilst the graph is clear and simple, it is insufficient to support the claims made by the author.

Cairo, A. (2013). A comparison of Spain’s and Sweden’s fertility rates [Image]. Retrieved August 22, 2018, from https://vimeo.com/176255825

Cairo gathered his own data on total fertility rates from the United Nations and created his own graphs to show that their is more sufficient way to support the argument made by the author. Cairo created a comparative graph about Spain and Sweden’s fertility rates (above). This graph provides clear, comparative data and gives the numbers a shape that helps the reader understand the data easier. However, to fully understand where the original article got its information we need to see the global fertility rates.

Cairo, A. (2013). Fertility rate [Image]. Retrieved August 22, 2018, from https://vimeo.com/176255825

This graph on Fertility rates is an excellent example of someone who has taken something that worked for a small set of data and let the software producing the graph create something on a large scale. Rather than interpreting the data and creating a graph that gives insight and understanding, this is the “default setting.” This graph is hard to look at and there is no visual hierarchy.

Cairo, A. (2013). Fertility rate [Image]. Retrieved August 22, 2018, from https://vimeo.com/176255825

Cairo created the above graph to display global fertility rates over time. This graph highlights a few poor and a few rich countries to help support the authors view and the hypothesis presented.

From this example provided by Cairo, we can see the importance of providing enough data for a reader to follow and argument or to engage in an evaluation for themselves. Visualisations need to carry meaning and provide the reader with enough information for them to see how a conclusion has been made. The examples of the historical visualisations shown in the previous post were simple in the data provided. The final visualisation in this post of a modern example is quite data heavy, requiring more of the reader. This suggests that we have more graphically/ visually sophisticated viewers than in earlier times.

 

Most important aspect and why?

This lecture discussed a really important point when designing data visualisations. That being, that the main function of data visualisation is to create insight and understanding for the reader not just present a pretty image. This helped my understanding in what I need to aim to achieve when creating data visualisations myself. Whilst appearance is important, it needs to carry function. For example the use of colour, font and layout to create visual hierarchy needs to be with purpose and to add meaning to the data, not just to create a beautiful image to look at. Our newsfeeds are full of pretty infographics that carry little to no meaning. I want to aspire to create data visualisations that carry meaning and educate the reader on something they couldn’t have seen without it.

Cairo’s example of how to properly support an argument with visualisations was very helpful in understanding how to approach this myself. Whilst the graph the reader used was clear and simple, it didn’t provide the reader with the chance to compare the data themselves. I found this important and helpful to see how Cairo pulled apart the information and displayed the data in a meaningful way. What worked for one set of data (the comparison between Spain and Sweden) didn’t work for the global comparison. It is important to understand your data and not just use a graph that the computer generate as, seen in the messy Fertility Rates graph, it is not always the best method of approach.

Reference 

Cmielewski, L. (2016, July 25). Visualisation: Historical and contemporary visualisation methods- Part 2 [Video File]. Retrieved from https://vimeo.com/176255824

Data Types

Depending on the data being collected, we use different type of data measurement in order to make sense of it. There are four main types of data measurement:

1. Nominal
2. Ordinal
3. Interval
4. Ratio.

Nominal data refers to named categories. It is an unordered form of measurement as no category is greater than the other. An example of this is when you are in a grocery store there are different categories that products come under. For example foods that come under dairy, produce etc would be seen as a nominal way of measurement as no category has greater value than the other.

Ordinal data is used for ordered non-numerical data. An example can be seen below of ordinal data measurement.

Interval data is a numerical data measurement, however it doesn’t have a meaningful 0 value. An example of interval data is the temperature as the value of 0 doesn’t mean the absent of heat. Other examples are time of day, calendar and years.

Ratio data is numerical data that does have a meaningful 0 value. The 0 indicates an absence of whatever you’re measuring. Examples of things that use this type of data measurement are height, weight, age and money.

Qualitative and Quantitative Data

Qualitative is non numeric descriptive information. This is nominal and ordinal data,

Quantitative is numerical data as it is quantifiable. This is interval and ratio data.

Most important aspect and why?

The most important aspect of this lecture was understanding the different forms of data measurement in order to apply the correct measurement type when collecting data. The reason this is important is it helps to understand how to give meaning to the data by using the correct measurement in which the data is first collected. In order to help my understanding of each data type I have gathered images as a reference point for later on.

Nominal data: 

Unknown. (2017). Examples of nominal scales [Images]. Retrieved August 13th 2018 from https://www.mymarketresearchmethods.com/types-of-data-nominal-ordinal-interval-ratio/

 

Ordinal data:

Bertram, D. (n.d.) Likert scales [Image]. Retrieved August 13th 2018 from http://poincare.matf.bg.ac.rs/~kristina/topic-dane-likert.pdf

 

Interval data: Average Temperature Chart of Jos, Africa 

Unknown. (n.d.) Average temperatures jos [Image]. Retrieved August 13th 2018 from https://en.climate-data.org/location/46664/

Ratio Data: 

Soltas, E. (2013). Average salary in the securities industry in New York City versus in all other industries [Image]. Retrieved August 13th 2018 from http://evansoltas.com/2013/02/27/5-more-graphs-on-finance/

 

This lecture helped me understand the main types of data measurement used today. I learnt that overall data comes under two categories, categorical or numerical and from their we can work backwards to determine which data measurement should be used.

 

Reference

Waterson, S. (2016). Data types [Video File]. Retrieved from https://vimeo.com/176274669

What is Data?

We live in a time where we all generate a lot of data. Everything we do seems to be connected to the online world and thus creates data trails. Every app you use, every site you engage with, every payment you make – it all generates data. Data is essential to helping us understand social, environmental and political systems. As the world changes and data increases, new visualisation strategies are needed to make sense of this data.

So what actually is data? Data is values of qualitative and quantitive variables belonging to a set of items and/or the result of measurements. (Waterson, 2016) Data by itself actually has no meaning. For data to contain information it must be interpreted to take on meaning. This is where data visualisation comes into things. What is data visualisation? Exactly as it sounds, data visualisation is the visualisation of data. It is one of the steps in data analysis and its goal is to communicate data clearly through graphs.

You have probably heard of the term infographic before. However, did you know there is a difference between infographic and data visualisation? Infographics are not necessarily based on data wheres all data visualisations are all information visualisations. Infographics often look pretty but don’t really contain a lot of information wheres data visualisations add meaning to information. For example the image below of an infographic about a process is not a data visualisation. It is a list of process not based on data.

Mcguire, S. (2018). Happiness tips infographic [Image]. Retrieved 10th August 2018 from https://venngage.com/blog/9-types-of-infographic-template/

Effective visualisation makes complex data more accessible, understandable and usable. It helps users analyse and give meaning to information.

As Data Visualisation Designers processing, analysing and communicating data creates many challenges. It is important to use the right visualisation type for the different types of data we need to visualise. The bar graph is the most basic and common visualisation, however it is the best when comparing two variables (see example below). When trying to decide what type of visualisation to use you should ask yourself “Why should I not do a bar graph?”

Unknown. (2015). Total revenue by product [Image]. Retrieved 10th August 2018 from https://irina150.wordpress.com/2015/10/08/bar-charts-versus-pie-charts/

The line chart or timeline (see example below) are the best default choice for looking at data over time. People are familiar and comfortable with both these types of visualisations and can read them easily. This is important as it allows for effective communication which is the foundation for our visualisation. The main role of the Data Visualisation Designer is to use data to tell a story.

Excel Easy. (n.d.). Line Chart [Image]. Retrieved 10th August 2018 from https://www.excel-easy.com/examples/line-chart.html

 

The 4×4 Model for Winning Knowledge Content 

Shander, B. (2014). The 4×4 model for winning knowledge content [Image]. Retrieved 2 August, 2018 from https://vimeo.com/100429442

 

Most important aspect and why? 

The most important aspect I took away from this lecture was that we as designers are to use data to tell a story. It is not about producing a “pretty picture” but whilst it is important to make visually pleasing, data visualisation is about interpreting data to give meaning and understanding. My understanding of data visualisation has already increased as I didn’t know that their was a different between infographics and data visualisations. My perspective has already changed on how I am to approach this semester and look to interpret data to tell a story. The reason this is important is because as communication designers if we are designing pieces that are not communication anything we have essentially failed at our job. Therefore, it is important to analyse data, interpret it and communicate it clearly so that the audience understands something that they wouldn’t have previously.

 

Reference

Waterson, S. (2016, July 18). What is data vis? [Video File]. Retrieved from https://vimeo.com/175177926