Interactive Visualization of Diversity in Multivariate Data Sets Unified across Fields of Study

The study of the diversity of multivariate objects shares common characteristics
across disciplines, including ecology and organizational management. Nevertheless, experts
in these two disciplines have adopted somewhat separate diversity concepts and
analysis techniques, limiting the ability of potentially sharing and cross comparing these
concerns. Moreover, while complex diversity data may benefit from exploratory data
analysis, most of the existing techniques emphasize confirmatory analysis based on statistical
metrics and models. To bridge these gaps, interactive visualization is especially
appealing because of its potential to allow users to explore diversity data in a direct and
holistic way, prior to further statistical analysis.
This dissertation addresses the problem of designing multivariate visualizations that
support exploration and communication of diversity patterns and processes in multivariate
data. To this aim, the dissertation presents design considerations as well as
implementation and evaluation of interactive visualizations targeting diversity analysis.
The contributing visualization techniques and tools include (1) Diversity Map - a
novel multivariate space-filling representation emphasizing diversity patterns in separate
attributes; (2) Ecological Distributions and Trends Explorer (EcoDATE) - a web-based
visual-analysis tool that is built upon the Diversity Map and facilitates the exploratory
analysis of long-term ecological data with an emphasis on distribution patterns and temporal
trends; and (3) HIST - a visual representation for communicating team diversity
faultlines across multiple attributes that is based on multiple linked, stacked histograms.
Further, drawing upon lessons from these designs, this dissertation cross compares analyses
of species diversity (ecology), microbial diversity (microbiology), and workgroup
diversity (organizational management) and introduces a unified taxonomy of analytical
tasks to guide the creation and evaluation of future diversity visualizations. The design
considerations, visualization techniques, tools, and task taxonomy are evaluated
and refined in empirical user studies involving human participants and subject-matter
experts.