Visualizations

Charts help you present or explore the coding in your project and ask questions—for example, What are the dominant themes in this interview transcript? or How is coding spread across the different authors in my project? NVivo for PC and NVivo for MAC have different ranges of chart types and the availability of the chart depends on the data you want to view. The full range includes create bar charts, column charts, pie charts, bubble charts, heat maps and radar charts.

Hierarchy charts visualize a hierarchy, and can help you see patterns in your coding, cases, attributes and files. Typically when you are using a hierarchy chart to analyze your files, you would be comparing the amount of coding between files. Similarly, when analyzing your codes, you would use hierarchy charts to question if some codes are used more than others or there are prominent themes in your project. All in all, hierarchy charts are useful when you want an overview of your data on multiple levels. There are two types of hierarchy charts - the Tree Map and the Sunburst. A Tree Map uses nested rectangles to represent the amount of, for example coding to a code. A Sunburst represents hierarchy levels as rings, and can give a detailed view of every level of your hierarchy, eg. parent and child codes.

Make a visual representation of your brainstorm when starting your project and use the assumptions, concepts and ideas to develop codes. The structure of the mind map can be used to set a boundary for the project and create consensus for your project team. The mind map can also be used to create the code hierarchy with a single click.

Similar to the mind map, the concept allows you to map out your ideas and explore connections in your data. There are many different shapes and icons you can use to conceptualize your project and data and arrows to show the direction of relationships in your data. You can also add items from your project such as articles or interviews to the map as well as the attributes that describe these objects.

A project map is only available on NVivo for PC.  Project maps are a graphic representation of the different items in your project and how they are connected. You can see how project items are related through codes, how they are linked to other project items and the centrality of project items. You can also use project maps to identify any gaps or isolated items that may require you to seek out additional data.

Comparison Diagrams show you what two items in your project have in common and where they differ. Thus you can compare two files to each other, two cases or two codes. This means you can quickly assess if, for example, two cases are addressing the same or different topics, what they have in common and what is unique about them.

Explore diagrams enable the visualization of connections between a single project item and the other project items. The diagrams are dynamic, meaning that the history of interacting with the diagram is remembered, allowing you to step forward and back through your project data to explore the connections between items. For example, you have a parent code called "Europe" and you want to explore all files coded to Europe. By making an explore diagram, you can explore each child code to Europe and see what different countries your files have been coded to.

Cluster Analysis Diagrams are only available on NVivo for PC. A cluster analysis group files or nodes that share words, attribute values, or codes. Project items that appear closer together are more similar than items that are far apart. However, as written on the NVivo help pages, cluster analysis is not a way to produce research results! Any associations must be further investigated using other methods. In NVivo such other methods are coding queries and matrix coding. The cluster analysis groups items using word similarity, coding similarity or attribute similarity and different similarity metrics that you need to be able to interpret to argue for validity of the resulting associations. Further, the analysis uses "stop words" in the similarity calculation, so make sure the stop word list for the language you are working in is sufficient.