After more than 15 years of development drawing on research in cognitive psychology, statistical graphics, computer science, and cartography, micromap designs are becoming part of mainstream statistical visualizations. Bringing together the research of two leaders in this field, Visualizing Data Patterns with Micromaps presents the many design variations and applications of micromaps, which link statistical information to an organized set of small maps. This full-color book helps readers simultaneously explore the statistical and geographic patterns in their data.
After illustrating the three main types of micromaps, the authors summarize the research behind the design of visualization tools that support exploration and communication of spatial data patterns. They then explain how these research findings can be applied to micromap designs in general and detail the specifics involved with linked, conditioned, and comparative micromap designs. To compare and contrast their purposes, limitations, and strengths, the final chapter applies all three of these techniques to the same demographic data for Louisiana before and after Hurricanes Katrina and Rita.
Offering numerous ancillary features, the book's website at http://mason.gmu.edu/~dcarr/Micromaps/ provides many boundary files and real data sets that address topics, such species biodiversity and alcoholism. One complete folder of data examples presents cancer statistics, risk factors, and demographic data. The site includes CCmaps, the dynamic implementation of conditioned micromaps written in Java, as well as a link to a generalized micromaps program. It also contains R functions and scripts for linked and comparative micromaps, enabling re-creation of all the corresponding examples in the book.
Daniel B. Carr is Professor of Statistics at George Mason University in Fairfax, Virginia. Building on his early experience in developing static and dynamic graphics at Pacific Northwest National Laboratory, Dr. Carr continues to use new data as the motivation to create new graphics. He has taught statistical graphics to hundreds of graduate students in the computational data sciences. For more than 30 years, he has developed graphical and computational methods for exploratory visualization and has collaborated with researchers at several federal agencies, including working with Dr. Pickle at the National Cancer Institute (NCI). Dr. Carr is a Fellow of the American Statistical Association. Linda Williams Pickle is principal and chief statistician at StatNet Consulting, LLC in Gaithersburg, Maryland, and Adjunct Professor of Geography and Public Health Services at Pennsylvania State University. She has devoted more than 30 years to the cancer research community, working for the National Cancer Institute (NCI), the Vincent T. Lombardi Comprehensive Cancer Center at Georgetown University, and the National Center for Health Statistics (NCHS). Dr. Pickle has published extensively about the spatial patterns of disease, including the award-winning Atlas of United States Mortality, which was the first of its kind to use statistical modeling in the background and to produce age-specific mortality rates based on modeled data. She is a Fellow of the American Statistical Association.
An Introduction to Micromaps Introduction Row-labeled plots Linked micromaps Conditioned micromaps Comparative micromaps Summary and preview of book chapters Research Influencing Micromap Design Influence of statistical graphics research on micromap designs Contributions from other research areas Human perceptual and cognitive strengths and limitations impacting data visualization Summary Data Visualization Design Principles Introduction Enabling accurate comparisons Strive for simple appearance Engage the analyst Summary Linked Micromaps Introduction Page layout Data encodings Micromap highlighting Multivariate data Multivariate sorting Pushing the envelope Software Summary Conditioned Micromaps Introduction One-way conditioned layouts Two-way layouts Higher order layouts Describing and comparing groups of regions Weighted descriptions and comparisons Alternative views CCmaps software options Summary Comparative Micromaps Introduction Representing change Types of comparisons Two-way comparisons Rates of change Alternative views Summary and future directions Putting It All Together Summary Exploration of Louisiana population changes after the 2005 hurricanes Concluding remarks Appendix 1: Data sources and notes Appendix 2: Symmetric perceptual groupings References Index