Processed

How can we make data visualizations for scientists better?

• scientists struggle w/data management & designing visualizations
• clearer, more consistent figures in science articles
• maintaining good datasets for easy reuse
• adding context and metadata to autogenerate data reports

Design

• combine input that encourages best practices for data management as well as better outputs than MS Excel default options
• preserve data and metadata info when generating images

Tools used

• d3.js, Raw, KnitR

Steps

1. Easy upload *.csv file
2. Automatic parsing
3. Prompt for names of variables as well as types

• continuous or categorical
• independent or dependent
• unit of measurement

4. Add additional dataset information (author, date, etc)
5. Select type of graph: line, bar, scatter
6. Drag variables to x, y, or color
7. Output to image file and/or print report