SparkTable: Generating graphical tables forwebsites and documents with R

Abstract

Visual analysis of data is important to understand the main characteristics, main trends and relationships in data sets and it can be used to assess the data quality. Using the R package sparkTable, statistical tables holding quantitative information can be enhanced by including spark-type graphs such as sparklines and sparkbars. These kind of graphics are well-known in literature and are considered as simple, intense and illustrative graphs that are small enough to fit in a single line. Thus, they can easily enrich tables and texts with additional information in a comprehensive visual way. The R package sparkTable uses a clean S4 class design and provides methods to create different types of sparkgraphs that can be used in websites, presentations and documents. We also implemented an easy way for non-experts to create highly complex tables. In this case, graphical parameters can be interactively changed, variables can be sorted, graphs can be added and removed in an interactive manner. Thereby it is possible to produce custom-tailored graphical tables - standard tables that are enriched with graphs - that can be displayed in a browser and exported to various formats.