Lecture 4 – Data Presentation Styles: Why we use graphics

This lecture discussed presentation styles and why we use them.

So, why do we use graphics?

Essentially, it is to make comparisons easier to understand. However, often graphic designers choose the wrong way to present information due to aesthetics what is most fashionable at the point of time. An example is bubble charts, as seen below.

Screenshot 2016-08-15 12.58.06

(Cairo, 2013).

This was a chart that was recreated by Alberto Cairo, The Functional Art. It’s displaying the change in market capitalisation in various banks between 2007 – 2009 and from analysing the image, it is clear that there was a decrease in market capitalisation.

Screenshot 2016-08-15 13.08.23

(Cairo, 2013).

The lecture then focused on one of the banks. On the left hand side, the bubble chart makes the the inner circle and the outer circle appear roughly the same size however, on the right hand side, the bar chart is telling the viewer a different story. The market capitalisation is 2009 looks roughly 1/3 of the size as it does in 2007. The lecture explained that while some visual designers create graphs that wants their readers to compare areas, readers will automatically compare heights and widths. Further, using circles make accurate comparisons harder and lead viewers to underestimate size difference.

Screenshot 2016-08-15 13.21.05
(McCandless, 2010)

In this image we can instantly see that squares are much easier to compare more accurately. Through height and width we can see that the Americans (yellow square) spent a huge amount of money – $7800 billion – to get themselves out of the financial crisis.

Screenshot 2016-08-15 13.45.29
(Cairo, 2013).

This is a ranking of different graphic approaches to comparing data, which is based on visual perception. The more accurate the judgement is for readers to make, is the more likely they will take away the correct perception. A map for example (which is situated next to ‘more generic judgements’) uses shading and colour saturation to show height, though there is other more important things on the map, the shading and colour saturation portray a realistic view of the image and make it easier for readers to make relevant comparisons. Whereas, when comparing things such as, dollar values, we need a more accurate judgement, therefore is is likely designers would use length to create clear comparisons for readers (as seen at the top next to ‘more accurate judgements’).

Screenshot 2016-08-15 13.49.50
(Cairo, 2013).

Above are the three most common charts used.

  1. Time series chart – often seen in stock market movements
  2. Bar chart – makes comparisons between things and is usually one dimensional
  3. Scatter plot – which as a variable on each axis and makes for an interesting way to compare a set of results.

Greater Insight can Resolve Big Issues

If a graph is designed well, they can lead to greater insight and a famous case of a disastrous use of poorly designed charts is a major launch accident.

The lecture explained the story:
Leading up to launch day, it had been unusually cold weather at the Kennedy Space Centre, Florida. The night before the launch their was a discussion regarding the safety of the o-rings (sealed sections of the rocket) and the possibility of them becoming damaged due to the cold weather. The booster rocket engineers made a number of launch recommendations and faxed thirteen graphics to support this. Here is one of them.

Screenshot 2016-08-15 14.11.59(Tufte, 2002)

In this graphic, there is nothing regarding the temperature and degree of damage, instead they ordered their information by time (that is the order of launch). Overall, any pattern is difficult to see.
After this skeptical image the rocket manager said the engineer managers evidence was inconclusive and they recommended the launch. The following morning the rocket launches and blew up a minute later.

Screenshot 2016-08-15 14.18.54
(Tufte, 2002)

Edward Tufte took that same data and re-created the rocket diagram as a simple scatter-plot graphic showing the relationship between the two variations that were of interest – temperature and o-ring damage. It reveals a clear pattern of damage, severity as temperature drops and shows there is always damage below 65 F.
This is a clear case of the mode of data presentation effecting the way we think about an issue.

The lecture then ended with an explanation of a line and pie chart, which hadn’t been discussed throughout the lecture as yet.

Screenshot 2016-08-15 14.28.04
(Google Ngram Viewer, 2016)

  1. Line chart – this image is taken from Google Books. Google books have been scanning books from the 1800s to the present and under ‘books.google.com/ngrams’, you can type in words and make comparisons of thousands of books that have been scanned for the occurrence of these words. The above image is an example of the rise and plato of commercial art and graphic design.

Screenshot 2016-08-15 14.37.23.png
(Cmielewski, 2016)

2. Pie chart – these are commonly used to show the relative proportions or perceptions of information such as, a percentage of a budget that is spent on different departments. It is important to limit the number of wedges on the pie to six or seven – more than this becomes too difficult for a reader to analyse and should be created into a bar chart instead. The above image is a good example of a pie chart – it is clear that the comparison between the electronics (in red) and the other subjects portray a large difference in price value.

Reflection
Overall this lecture clearly explained why we use graphs, how to best use graphs and what graphs suit certain information more than others. I learnt five things:

  1. That using circles make making accurate comparisons harder and lead viewers to underestimate size difference.
  2. That readers automatically compare heights and width more so then an area of a shape, therefore using a square is a more successful approach then using a circle to convey comparisons.
  3. I learnt about the ranking of different graphic approaches to comparing data based on visual perception by understanding what ‘allows more accurate judgements’ and what ‘allows more generic judgements’.
  4. That if a graph is designed well it can lead to greater insight and resolve issues large or small.
  5. That a time series chart, bar chart, scatter plot, line chart and pie chart are the most successful and commonly used graphs today.

References

Cairo, A. (2013). The functional art: An introduction to information graphics and visualisation. (1st ed.). Berkeley, CA: New Riders

Cmielewski, L. (Speaker). 2016. Lecture Pod 04: Data Presentation Styles: Why we use graphics [Vimeo video]. Western Sydney University

Google Ngram Viewer. (2016). Books.google.com. Retrieved 15 August 2016, from https://books.google.com/ngrams/graph?content=commercial+art%2Cgraphic+design&year_start=1800&year_end=2000&corpus=15&smoothing=2&share=&direct_url=t1%3B%2Ccommercial%20art%3B%2Cc0%3B.t1%3B%2Cgraphic%20design%3B%2Cc0

McCandless, D. (2016). Information is Beautiful. Information is Beautiful. Retrieved 15 August 2016, from http://www.informationisbeautiful.net/

Tufte, E. (2002). O-ring Damage Index: Scatter Plot. Retrieved from http://motherboard.vice.com/read/how-mistakes-were-made

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s