You are hereHsia-Ching Chang and Eric Upchurch

Hsia-Ching Chang and Eric Upchurch

By admin2019 - Posted on 18 December 2014

Hsia-Ching Chang is an assistant professor in the College of Information, Department of Library and Information Sciences at the University of North Texas. She was the founder/business analytics consultant at CrystalWave Analytics, worked as a project manager for the Hyweb Technology Co. Ltd and a product manager for in Taiwan. Her doctorate dissertation focused on applying Twitter data analytics to explore the dynamics of information architectures. She holds Ph.D. and M.S. degrees in information science from University at Albany, State University of New York and a M.S. in public policy from National Taipei University in Taiwan. Her research interests concentrate on business analytics, social media, knowledge mapping, cloud security, enterprise information architecture and emerging IS/IT adoption and diffusion.

Eric Upchurch is an Interdisciplinary Information Science PhD student at the University of North Texas. His research focus in 2014 has been oriented towards Big Data analytics, Text mining, and Knowledge Management. Mr. Upchurch is also a Senior IT Manager for the City of Fort Worth IT department whose primary responsibilities include management of the Data Center facilities, the Disaster Recovery Program, the Operations team, and the Production Control team. Mr. Upchurch is always endeavoring to exhibit a transdiciplinary approach that bridges the best of the academy with business in an effort to create unique value.

Big Data Visualization as Storytelling: A Means-Ends Chain Approach

The big data taxonomy classifies big data into six knowledge domains: data, compute infrastructure, storage infrastructure, analytics, visualization, security and privacy (Cloud Security Alliance, 2014). The Uniqueness of the big data phenomenon stems from its volume, variety, velocity and veracity, which has changed the way work and life revolve around data. Much like the initiative, open government policy enables open data availability and accessibility where citizens can benefit from the usefulness and value added by mashups of open data. A growing number of vendors including Socrata, Junar, and CKAN provide open data platforms that make open data publishing and use more intuitive by supporting a variety of users, such as citizens who are interested in gaining insights into open data, decision makers, and data experts. With those platforms, consumers can utilize data visualization tools to analyze, visualize, and interpret open data on the Web. Different consumers may have different means and ends (goals) with respect to the creation of the visual representation of data, which may tell different stories.

This study aims to explore big data in the context of open data, understand the relationship between big data and open data, the role of visualization in open data, the status quo of open data platform products, how these products can be used to realize open data visualizations and how information and knowledge professionals use them. To understand the decision making process of information and knowledge professionals in an open data visualization context, this study takes a means-end chain theory approach, which assumes consumers consider products as means to achieve important goals. Proposed by Rokeach (1968) and applied to consumer behavior by Gutman (1981), the meansā€end chain represents users’ cognitive structure by associating product (concrete/abstract) attributes with both the functional and psychological consequences of product use and the personal values behind the consequences. Analyzing means-end chains assists in identifying users’ demands, means to achieve their goals and underlying values. Pilot laddering interviews on decision-making of adopting open data visualization tools will be conducted with several information and knowledge professionals. Beginning with extracting key dimensions of (visualization) product attributes based on the product descriptions and advertisements from multiple open data platform vendor websites, the laddering interviews will ask the respondents to choose the identified product attributes that matter most to them and several words from the big data visualization taxonomy regarding their stories/experience on open data visualization. For instance, big data visualization techniques can be broadly divided into three categories: spatial layout visualization (including charts & plots and trees & graphs), abstract/summary visualization (compact/reduced dimension representation of data) and interactive/real-time visualization. The interview results which describe means-end chains and selected concepts/words from the taxonomy can generate meaningful associations, which reveals consumers’ means, the major processes of open data visualization that affect the consequences, and perceived values. As a result, a hierarchical value map can be constructed to address the gaps between the product attributes and consumers’ demands.