Imagine a data scientist said to you, “collecting and cleaning data is laborious and tedious, but once I had the data, an infinite variety of views opened up.” While sounding like a familiar lament from any contemporary data conference, this quote is 250 years old. It was written by Joseph Priestley, the polymath who created some of the first visualisations of visualise historical data.
The first data analysts faced same challenges as today
Priestley’s quote reminds us that we should not forget that we’ve been wrestling with data analytics and visualisation for a long time. The challenges the pioneers faced are the same ones we face today. Aspiring data-driven organisations can learn as much from the pioneers as any contemporary text book.
William Playfair, another pioneer of visualisation in the early 19th Century, said this about data analysis: “As knowledge increases among mankind, and transactions multiply, it becomes more and more desirable to abbreviate and facilitate the modes of conveying information from one person to another, and from one individual to the many.” He’s talking about the modern challenge – summarise information to enable more people to understand a system.
Data must be visualised to make a difference
Playfair’s words emphasise how much the first data analysts realised the absolute necessity of visualising data to get their message across. Raw data is invaluable but must be transformed into something else to make a difference.
The battle between letting the data speak for itself and creating a vivid visualisation that grabs the audience began two hundred years ago. Playfair said, “nothing is more dry and tedious as statistics,” and data must “set the mind and imagination to work”.
But a contemporary of Playfair, the medical statistician William Farr, said, “statistics should be the driest reading of all”. Reconciling these two viewpoints continues to challenge us today!
A long history of bad and good practices
The pioneers experimented, creating the wonderful line and bar charts that are so effective today. Yet, they also gave us pie charts, leading to many crimes against data that persist nowadays.
Finding the “best” charts has been a driver for pioneers in data visualisation. American engineer Willard Cope Brinton published his “Graphical methods for presenting facts” in 1914 with a clear belief that “facts do not speak for themselves. When they do speak for themselves, the wrong conclusions are often drawn from them. Unless the facts are presented in a clear and interesting manner, they are about as effective as a phonograph record with the phonograph missing.”
His book offers today’s data analysts an extremely rich and wise set of instructions and ideas about the dos and don’ts of using graphics to explain data insights. Notably, Brinton slams the pie chart as “inaccurate” and applauds bar graphs as a better way to convey data comparisons.
Challenge the status quo (but don’t forget the roots)
Brinton is incredibly modern – he foresaw the data-driven enterprise when he said the executive of the future will depend on his powers of accurate data analysis. Brinton also realised that merely having the data and visualising it is not enough. To be a data-driven organisation, people at all levels need to be better educated in data literacy.
Even as data analysis and visualisation are integral to digital businesses and are supported by advances in modern business intelligence, including AI and machine learning, some fundamentals have usefully been set out and debated for years, if not centuries.
Andy Cotgreave, Technical evangelist and senior director, Tableau
Andy has a strong passion for visual analysis and data technologies. He is a visual analysis expert at Tableau and has a strong perspective on the issue of delivering basic digital skills. He has been with Tableau in various roles since 2011 ranging from product consultant to product marketing to his current role as Senior Technical Evangelist. Prior to joining Tableau Andy was a data analyst at the University of Oxford.
