A short video, showing the decline of manufacturing jobs and the rise of tech and science jobs. This explorable is created by EVA (eva.cmucreatelab.org).
EVA is online: eva.cmucreatelab.org
Explorable Visual Analytics, EVA, is a web-app designed for visualizing and exploring large and complex datasets. It was started as my PhD thesis project around 2 years ago at CMU’s CREATE Lab. Since last year, Amir Yahyavi, a PostDoc at our lab, has also joined the project. Now, EVA is capable of scaling to any data size.
Over the past few months, we have collaborated with the U.S. Census Bureau to import their Longitudinal Employer-Household Dynamics dataset into our tool. Now, using EVA, you can explore this fascinating dataset on jobs and workforce, build your stories based on the data, and easily share the entire exploration process.
Give it a go and check out our vimeo channel for how-to videos and stories: https://vimeo.com/channels/eva
My friend Amir and I will be presenting a workshop on EVA using LEHD dataset. It will be held in US Census on June 25. This is the flyer info:
U.S. Census Bureau and Carnegie Mellon University CREATE Lab
Storytelling with Data
A Workshop on Visualizing and Exploring Very Large Datasets
U.S. Census Bureau Headquarters – June 25, 2015
Screenshot: A 3D spatial and temporal explorable view of jobs and income levels near Philadelphia
The U.S. Census Bureau’s LEHD Program and Center for Economic Studies invite you to participate in a day of cutting edge data exploration and engagement on usability and visualization analytics using large scale economic data. This workshop is being held in coordination with the CREATE Lab at Carnegie Mellon University and will take place at the Census Bureau’s headquarters in Suitland, MD.
During the day, participants will learn to use Explorable Visual Analytics (EVA) – a free, open-source, web browser-based tool for data exploration and visualization that was built by CREATE Lab. Data for exploration will include the Residence Area Characteristics (RAC) from the public-use LODES dataset, which is produced by the LEHD Program and includes demographic and employment information on 120 million workers in the United States as well as their residential distribution. In total, the dataset comprises 2.7 millions entries and 46 dimensions (worker age, sex, race, ethnicity, and educational attainment; firm ownership and industry; and job earnings).
Invitees will take part in guided investigation of data for the Washington DC metropolitan area (DC, Delaware, Maryland, Pennsylvania, and Virginia). The interactive and easy to use nature of EVA facilitates intuitive data exploration and leads to new forms of hypothesis generation and inquiry. Invitees will then be able to generate and explore their own questions from their diverse experiences, skill level, and professional expertise.
EVA will allow invitees to explore this data in space and time in variable resolutions, fluidly create new views by selecting desired data dimensions, and construct a narrative supported by a sequence of interactive visualizations. Because EVA will be publicly available, invitees will be able to continue using the tool to share big data tours on standard web browsers after the workshop is complete. It will also be possible to host data using EVA’s visualization engine on invitees’ own websites and an afternoon session will be held to discuss the technical considerations and requirements for doing so.
The Longitudinal Employer-Household Dynamics (LEHD) program is part of the Center for Economic Studies at the U.S. Census Bureau. The LEHD program produces new, cost effective, public-use information combining federal, state and Census Bureau data on employers and employees under the Local Employment Dynamics (LED) Partnership. State and local authorities increasingly need detailed local information about their economies to make informed decisions. The LED Partnership works to fill critical data gaps and provide indicators needed by state and local authorities.
 The Center for Economic Studies (CES) partners with stakeholders within and outside the Census Bureau to improve measures of the economy and people of the United States through research and the development of innovative data products.
I presented my current research in our ISP Seminar series. Gaurav has written a very nice blog post on it (link), summarizing the talk.