The Truthful Art (Ch11 & Ch12) : - MariaAguilarV/JMM-622-Infographics-and-Data-Visualization GitHub Wiki

In previous chapters, we learned about how to collect and interpret data, but in this process, it is also important to know how to deliver this information to the audience, analyzing and interpreting this data as accurate as possible. In Chapter 11, Cairo introduces terms as standard error and confidence interval in order to understand the uncertainty and significance that arise when collecting and interpreting data from a sample of individuals. It is impossible to reach a standard error of 0, there will be always uncertainty in the data, but as the author states “what really matters to reduce the amount of uncertainty is the size of the samples and if they have been drawn following the strict rules of random sampling”. How can we envision and report this uncertainty? Using a confidence interval. To calculate a confidence interval, we need to decide on a confidence level, it could be any number, but the most common ones are 95 and 99 percent level of confidence. The greater the confidence level we choose, the greater the margin of error becomes. Then, the author also shows a variety of graphics that we can use to visualize these concepts such as error bars, and more detailed and creative versions as the gradient plot, the violin plot and box plots. Finally, in chapter 12, the author finishes the book by showing inspirational visualizations that are contributing to expanding the vocabulary and grammar of graphics. Although there are some well-known sources of great data visualization as The New York Times, Cairo illustrates the work of people that represented creative ways to analyze business data, to show unexplored fields of literature and movies, and to change the whole experience of data visualizations by making interactive projects and place the reader at the center of the stage. Furthermore, there is a valuable message given by the author in this last chapter, the way these authors came up with these amazing ideas comes after a process of trial and error, practicing and understanding what data visualization is about, I really loved these phrase: “You cannot think “out of the box” if you don’t know really well what the inside of the box looks like”.

After reading these chapters, I thought about a common cover page of a renowned newspaper in Peru, “Diario Gestion”, which is published from time to time and talks about the acceptance of the Peruvian president (Martin Vizcarra) by the population. Let’s see their last publication, it says “Martin Vizcarra’s disapproval increases by 46% and approval drops by 42% in April”. After reading this chapter I immediately thought that the Peruvian population is really disappointed by the president; however, I read the whole publication and I found, at the end of the page, the data sheet that says that the sample size was 1252 people distributed in 17 departments, and the margin of error was +/- 2.8%. So, is it correct to assume that the entire population is disappointed by the president? Maybe not. Taking into account that the population of Peru is 32.1 million and that it is constituted by 24 departments. The whole report can be found here