Designing data exploration for non-expert audiences

Abstract

As digital technologies permeate everyday life more than ever before, data is increasingly salient to the everyday lives of laypeople, or non-experts, as this dissertation defines them. Open source databases, such as government or nonprofit databases, have proliferated as organizations seek to provide data to the general public. However, many databases that are intended to be open source come with built-in barriers for non-experts, as they require individuals to have data-handling knowledge, technical skills, and/or field expertise in order to interact with, interpret, or utilize said data. Such data may be relevant to, or even taken directly from, the lives of non-experts, yet often remains inaccessible to prospective non-expert audiences.

Interactive data exploration systems intended specifically for non-expert audiences can address such barriers, serving as a bridge between raw data sources and non-experts. Moreover such systems can be useful tools for science communication and knowledge dissemination and activities, such as citizen science and community organizing.

As such, this dissertation presents a series of design principles, aimed at designers, for designing data exploration systems for non-expert audiences.

This Human Computer Interaction (HCI) dissertation, which is rooted in the Design Science Research (DSR) methodology, identifies trends for designing data exploration systems for non-expert audiences via a systematic mapping study, proposes design principles via a second literature review that builds upon that systematic mapping study, and finally evaluates and refines these design principles via two user studies that involve the creation and iterative design of a data exploration system based on said principles. Culminating in four final design principles that are rooted in data exploration and HCI research, and built upon the DSR methodology.