Graphics and How We Display Our Statistical Information
In the day and age of technology we want to see bright colors and interactive displays for graphical information. But, what is the best way to display our data? Gelman (2013), discussed how graphical communication of data and models is important in statistical practice. Infovis which is information visualization is intended to seize the reader’s attention and tell a story where interactive graphics have become a go to for the exploratory data analysis area. Like the Baby Name Wizard where people can see the popularity of first names over time. Per Gelman & Unwin (2013) this graph type of display followed the recommendations of Tufte, Cleveland and others. Looking up my own name “Jennifer” showed that my name was popular during the 80’s with over 60 million births being named Jennifer. This was very engaging and eye-catching to myself, the reader.
Graphs can look decent and practical but not be competent with what is being displayed to the reader. Graphics show us many things about the data being represented (ex: legend, title and x — y representation) this way the reader knows what the outcome of the data that is being shown. We have come a long way from taking small data sets to “big data” now being presented to readers. Some of the goals discussed in the article by Gelman & Unwin (2013) were to make graphics eye-catching and how to make readers more engaged in what they are looking at. This may help readers become more committed to understanding and thinking about the data.
Graphics assist the reader in understanding statistical data visualization and infographics can tell a story to the reader about the data being presented. The graph, shown in the Gelman (2013) article by Florence Nightingale, of the Crimean War, seemed impressive enough to myself for its time, however, it was a bit uneasy on the eyes for looking at it from a data standpoint. Although, it was informative and as a reader drew me into that information being represented.
Most of the graphs we use as researchers seem to be similar by using histograms, scatter or line plots. Adding some color or making a graph look more appealing are ok, however, when we go too far with graphics they may take over the data that is being presented. I am interested in learning more about making graphs more appealing in R. In the HSE598 class we have been working with ggplot and have been practicing with making different types of plots and colored graphs of data sets. This has been both fun and informative because all plots and graphs do not show the information or data in the same way.
So as statisticians in graduate school, presentations are the norm for classes and thesis work. It is important to make sure graphics during presentations are an informative type of graph for the reader and that they are displayed in a pleasingly graphical manner that still allows the reader to be able to understand the data being presented.
References:
Gelman, A. and Unwin, A. (2013). Infovis and statistical graphics: Different goals, different looks. Journal of Computational and Graphical Statistics. Vol. 22, №1, 2–28.
