by joseph12

My first impression of Sankey was that it looked like a very detailed and easy to understand visualization tool that accurately portrayed the amount of used/unused energy in the United States. Perhaps to no one’s surprise, Oil was the most abundant and used energy in the U.S with the majority of it being used for transportation and industrial purposes. Natural gas was second and was followed by coal in terms of energy usage. Solar and wind were far lower than I was expecting, as there are major companies that generate billions of dollars yearly using wind/solar energy.

For areas such as Maryland and Ohio, the most used form of energy was coal, and it was mainly used for electrical and even some industrial purposes. Although the top 3 sources of energy (i.e. Coal, Oil, and Natural Gas) remained the same as the rest of the United States, unlike most of the other other states, Maryland and a few other states most used source of energy seemed to be coal, followed by oil and Natural Gas.

The first thing I noticed about California’s energy expenditure compared to other states was that it was far greater than the rest. As a matter of fact, when comparing the results, it appears that California alone uses more energy than a combination of multiple states from the previous screenshot. Furthermore, like most other states, oil was its main source of energy, and natural gas was close second.

All in all I do believe that Sankey was a very good and efficient tool to portray energy used and energy lost. It is incredible detailed and all the while manages to somehow retain a simplistic aspect to it. The visualizations are very easy to understand. So much so that one would be able to understand what is being portrayed even without the numerical values being shown. The only criticism that I would personally give would be the fact that the data used may have been a little bit outdated as there were estimated energy usages going back as far as 2011.

By Joseph

Pie charts are one of the many types of visualizations that one can use to portray both complex or simple data. A pie chart can be especially good when the data being portrayed are all parts of a whole. As shown by the pie chart above, the values for the Recommended diet intake are all individual parts that add up to a 100%. Furthermore there are both small and large data sets that needed to be represented such as, the dairy value and the fruits value. Both data sets are vastly different in terms of percentage and hence the amount of space that they take up visually. To convey this, a pie chart is the perfect visualization to use in this instance. This is especially true when one considers that the appearance of a pie chart also falls in line with the topic of the graph itself which is diet and food intake. the values not only add up to a 100%, but also the data appear to reflect a plate, with each data set representing a specific food intake. Upon making this pie chart, I did come across some suggestions as to how I could improve my data such as changing the color of the data set or make one value stand out by removing color from every value except for one. One could also disconnect a certain piece from the whole to visually represent something to the audience. These design methods are all very valid and useful in data visualization, but I decided to make the pie chart I created very simple to understand. The more elements that one adds to something, the more important it is that the information is clearly conveyed. Not only was there not a need to apply extra designs for the data that was conveyed, but there would also be a risk that the audience may misinterpret certain designs choices if they poorly represent the data. All in all, a pie chart is very efficient visualization to represent data that is part of whole.

Data visualizations are a major aspect in simplifying complex data to a broad population of people with different cultures, and languages. As shown by the New York Times bar graph of the speaking times of political candidates and the amount of attacks that each political candidate received in comparison to their speaking times. Taking a look at the bar graph, it is very clear what it is trying to convey. The graph not only labels the time in increments with labels, but also states the total amount of time that each candidate ended up speaking for. Though the graph for comparison is separate from the bar graph of the total time, the clarity of the information being presented leaves very little room in regards to which graph is what, and what it is they are trying to convey. What was interesting about this comparison is that the person that was the most attacked was surprisingly not the person that spoke the most, but rather someone that they didn’t even speak at all. Furthermore, there are also a plethora of other data visualizations that compare numbers that could have also been used to portray the data besides the a bar graph. Two such examples would be doughnut chart or a pie chart. Both would not only accurately represent the data, but also offer a visual representation of the percentages that each candidate spoke in comparison to the total time given. With that being said, Data visualizations for comparison of numbers can be applicable for a number of different topics, aside from political topics, and portrayed in many different kinds of graphs.