Data Deconstruction Lab

Update History:

  1. 2017-01-20 /First draft/

01 UI Haus mobile graph kit - Internal/Ratio


  • Description: This is a UI design of Web Monitor App for iOS like Google Analytics, from the screen the user can get the info of website like comments, likes, clicks summary and the history.
  • Variables: time as a interval variables, users/clicks/likes are ratio variables.
  • Visual encoding: The line chart for the score demonstration, the circle size and color for the key element monitor (green: >50%; red: <50%), the table for the specific check time, .etc
  • Pros: It uses several ways to show the real time data for user to monitor the website.
  • Cons: the table view can also change the transition color from green to red.

02 Inside Fortune’s 2016 Fastest-Growing Companies List - Ordinal/Ratio


  • Description: The companies on the list are ranked by the individual components used to determine the overall placement. This is non-aggregated dataset, the visual use the d3 to show the different ranks together with line. there is a combine infographic, the breakdown of industries and ranks.
  • Variables: different ranks is the ordinal variable, the number of different industries is the ratio variable.
  • Visual encoding: The vis use different colors to show 7 industries and the curve to connect 4 kinds of ranks. The red color also showed the big category of company list.
  • Pros: It use colors to show the different category of dataset.
  • Cons: It would be better, if they can add the quantity attributes to the vis, like the size of the line to show the quantity or the ratio.

03 Increasing diversity - Ratio


  • Description: To quantify how America is changing, they used the diversity index, which measures the chance that two people chosen at random will not be the same race and ethnicity. to show the county-level change in diversity since 2000. It is a transformed map with transition colors.
  • Variables: diversity index is a ratio variables.
  • Visual encoding: The color scale, shown in the top right, represents two things: level of diversity and change in diversity.
  • Pros: With just the diversity index dataset can show the absolutely index and the increase change is cool.
  • Cons: I think the four dimensions are too much for this graphic, too much different colors in this map may make the audiences confused. It would be better if the vis just use contrast colors and less than 8 kinds.

04 Gender wage gap, how much less women make than men - Nominal/Ratio


  • Description: This vis shows the pay gap between women and men in different industries. It is a non-aggregate dataset with the statistics median number for the vis.
  • Variables: Gender is the nominal variable, Industry is the nominal variable, Median Salary is ratio variable.
  • Visual encoding: The color mapped to sample belongs to which industry.the assistance line shows the three degree line.
  • Pros: In this vis, there is a assistant line chart to show the overall change based on the different year.
  • Cons: The salary comparison line can be used with different color like red, yellow and green. If the dots in different degree(10%less/20%less/30%less)they can shown with different shapes like rectangle and triangle, it can show the different category more clearly.

05 Diversity in Tech - Nominal


  • Description: In the vis, employee breakdown of key technology companies, from this vis, we can have a view of gender and ethnicity diversity ratio. This is an aggregated dataset, it uses the summary of variables to make a ratio.
  • Variables: Gender and Ethnicity, they are both nominal variables.
  • Visual encoding: the vis’s x axis split into two parts 1. Gender, 2. Ethnicity. The primary visual encoding is number and color .
  • Pros: This vis is simple but clearly to understand with color and the compare with US Population.
  • Cons: It can add a sort button and it would be better if there is a line chart for the comparison between 2014/2015.