Yiyang Shi

Good example:

Source: http://www.bain.com/publications/articles/winning-over-shoppers-in-chinas-new-normal.aspx

This is a graph on the trend of FMCG categories obtained from Bain company’s website. I think it is a fairly good example based on perceptual and cognition factors for two reasons. First, the use of symbol is reasonable. People can be good at slope sometime. But when slope changes and we try to figure out the large picture, we can become less accurate. The author solved this issue by providing a downward arrow to indicate the change. The arrow pops-up when looking at the graph, and we immediately get the idea that the growth is decreasing.

Second, the use of color is reasonable. Other than the two continuous data type on x-axis and y-axis, the graph also contains a categorical data type, the FMCG categories. Notice that “Food and beverage” and “Personal and home care” are two independent categories while “Total FMCG” is the sum of the two categories. The author used red color to contrast the two gray colors so that the total FMCG category can pop-up and differentiate itself with the two other categories. Some may argue that we should use two colors for the two independent categories. I think the author chose gray color for two reasons. One is that red and gray are the two colors of Bain Company so the author really doesn’t have many choices. Two is that using three different colors will mislead viewers to think that there are three independent categories while there are actually just two.

Bad Example 1:

Source: https://247wallst.files.wordpress.com/2016/02/tallest-building-height-by-state-map.jpg

This is a graph on tallest building height by state obtained from a website called 247 Wall Street. As a graph visualization on building height, it made several mistakes based on perceptual and cognition factors. The first mistake is the use of color. The height of building is continuous data. However, when presenting tallest building by state, the height data become categorical data where each state forms its own category. At most 12 distinct colors should be used to present categorical data. However, the author assigned some buildings to be the same color with no obvious reasons. The author also used some gradual changing colors such as light green and dark green, light brown and dark brown which creates more confusion. The second mistake is the use of bars. Using bar to represent magnitude or, in this case, height is only pre-attentive when bars are adjacent to each other. With the ways the heights are presented in the graph, viewers will have a hard time to distinguish the highest, the second highest, and the lowest building. The third mistake is the use of width. Width is a potential pre-attentive feature. In this case, the width of the building is obviously different and we don’t know why.

In general, this is a bad example of visualization. In fact, a very simple 2-dimensional bar chart would do a much better job on presenting tallest building height by state than this graph.

Bad Example 2:

Source: http://www.nytimes.com/2016/02/10/technology/fall-in-tech-stocks-is-faster-and-farther-than-broader-market.html?ref=technology Evaluation

This is a graph on F.A.N.G stock price obtained from N.Y. Times website. Based on perceptual and cognition factors, it is a bad example of visualization for two reasons. First, people are not accurate at slope when the slopes are not compared to each other. The viewers will like to compare the changes in stock prices of the four companies. However, since the four slopes are not presented together, we can hardly tell which one falls the most. Also, we should notice that the scale of the y-axis, i.e. the scale of the price, is different among the four stocks. This is very misleading. Second, for categorical data, different colors should be applied for each category. The four companies are categorical data with the same type of continuous data, share price, being evaluated here. However, the colors for the four companies are the same. If the colors are different, it will have a pop-up pre-attentive effect and viewers can automatically recognize that there are four different companies being discussed here.

Botanical Tree:

Based on perceptual and cognition factors, I think the design is good in 3-D, lighting aspect, and shape but bad in color. The two graphs are good in 3-D design because it has the depth cue, the linear perspective, and the shadows. The depth cues allow us to see the spheres and the vertebrals pre-attentively. The design also has linear perspective where the objects become smaller with more distance. By presenting the shadows on the branches, we can tell the location relationships between branches and objects. Other than 3D perception, the shape of the branches and the “fruits” are also pre-attentive. More than that, we can easily tell where the light comes from on the graph.

However, the color use is not very satisfactory. I am not sure about the color on the spheres, but the color on the branches should be continuous. It seems that the design is trying to approximate the gradual changing color on the branches with about 3 or 4 colors but it comes out as separate. It could be misleading at the first glance because viewers will mistake continuous data into ordinal or even categorical data. It is good that the designer controlled the total number of color under the acceptable amount but I still have no idea about the colors on the “fruits” and what they represent.

- I have neither given nor received aid while working on this assignment. I have completed the graded portion BEFORE looking at anyone else's work on this assignment. Yiyang Shi -