D3 Projects.

This project visualizes data from a 1974 paper where 52 undergradautes were asked to judge the appropriateness of 15 behaviors in each of 15 situations. The data is represented as a matrix where rows represent situations and columns represent behaviors. The values may be represented by saturation and size and it is possible to switch from a single color gradient to a two color gradient with the buttons in the top left corner. Clicking on an entry will bring it to the top left corner and sort all other rows and columns based on similarity.

I was not aware while making this visualization but I later learned that this technique of reorganizing data in this way was originally developed by Jacques Bertin in the 1970s as a "reorderable matrix" where rows and columns were physically moved by hand.

This visualization uses data from a UK national survey that collects food consumption data from participants who maintain food diaries. The food is broken down into various categories and it spans many years. The dataset is very large so I decided to create very simplified representations where each category of food is represented by a different color and the quantities are represented by circular barcharts. This allows a viewer to see what is essentially a thumbnail for each year and quickly identify years with anomolous "shapes" and see how the shapes seem to change.

A large representation of the currently selected year is represented in the middle and mousing over different elements will show tooltips with more detailed information. Small simplified representations lie around the large representation and clicking on them selects that year to be shown in the center.

This project is a dashboard that combines datasets from several large databases. The main information sources are very detailed Body Mass Index (BMI) information from the CDC's Behavioral Risk Factor Surveillance System (BRFSS), additional BMI, poverty, and exercise data by state, cereal nutrition information obtained from consumer reports, and favorite cereals by state obtained from Google Trends. The datasets are very large and contain breakdowns for different demographics according to gender, age, race, education level, and income.

Almost every part of the dashboard is interactable. A user is able to click on a state to see data specific to that state on the left, they are able to select a year to see how the map looked at the time, the dial on the top left is used for selecting different demographics. The list of cereals on the right is reordered according to which cereal is currently selected and selecting a cereal will highlight it on the map, on the graph in the upper right, and on the parallel coordinate chart above showing nutrition information.