Data visualization is the graphical representation of information and data that is designed to facilitate understanding. Information about your data - or your raw data - is described and displayed through position, color, size, and shape Good data visualization techniques can help make access to your data more equitable. Good data visualization is not simply a collection of facts. Rather, good visualization tells a story. There are two types of data visualization: exploratory and explanatory. Exploratory data visualization is to recognize patterns in your data. Explanatory data visualization is to share an insight about your data.
Interested in learning more? Penn State users have access to a variety of tutorials and training materials on data visualization through LinkedIn Learning.
Choosing the best visualization to represent your data can be tricky. When selecting a chart, consider what you want to do with your data. You can use your data: to inform, to compare, to show change, to organize, or to reveal relationships.
INFORM: If you want to emphasize a single, important data point, consider using a single big number, a donut chart, or a pictograph.
COMPARE: If you want to compare categories or show comparison, consider using: bar charts (for many categories), bubble charts (for a few categories), pie charts (for composition), stacked bar charts (to show composition over time or across categories), or a word cloud (to show word frequency).
CHANGE: If you want to show change over time or by location, consider utilizing: line charts (many series over time), area charts (few series over time), timelines (to show distinct events through time), or maps (showing location).
ORGANIZE: If you want to show groupings, rankings, or processes, consider choosing: lists (to show a simple process), flow charts (to show a complex process), Venn diagrams (for groupings), mind maps (for groupings and connections), pyramid diagrams (for simple hierarchies), or ordered bar charts (for numerical rankings).
REVEAL: If you want to showcase relationships such as correlations or distributions, consider using: scatter plots (to show relationships between two variables), histograms (to show distribution of one variable), or a multi-series chart (to show relationship between multiple series over time).
Interested in learning more about data visualization and data analysis? The University Libraries Research Informatics and Publishing department offers individual research and statistical consultations on a variety of topics:
Research Informatics and Publishing additionally supports the Advanced Analytics and Visualization Digital Lab. This computer lab space is designed for those seeking access to and support for data analysis and visualization software.