Adding a dimension to anscombe's quartet: Open source, 3-d data visualization

Abstract

At the same time that Information Scientists have been asked to visually analyze larger and larger datasets, the options for three-dimensional expression of those datasets has grown tremendously. This demonstration will share a new customization of open source software that links data and 3-D modeled space. Once the data is expressed in 3-D to the analyst's liking, methods will be detailed to take a 3- D representation of data to stereoscopic images, to interactive virtual space, and to a 3-D printed object. The theoretical underpinning of this presentation is based on a hypothetical expansion of Anscombe's Quartet from a two dimensional construct to a three dimensional one. Anscombe's Quartet was developed in 1973 by the statistician Francis Anscombe to demonstrate how datasets can be extremely statistically similar, but visually map in totally different ways. The development and research goal of this work is to develop an accessible 3-D data tool that allows for a high level of control by the user, and facilitate future studies on the effectiveness and best use practices associated with 3-D visualization. © 2013 Erik Choi.