Hierarchy charts visualize a hierarchy, helping you to see patterns in your coding or view the attribute values of cases and sources. You can display a hierarchy chart as a tree map or sunburst. You need NVivo for Mac (Version 11.2 or later) to work with hierarchy charts.
This topic gives you an idea of what you can do with hierarchy charts and provides links to other useful topics.
In this topic
- What is a hierarchy chart?
- What can I show in a hierarchy chart?
- Which type of hierarchy chart should I use?
- How display options affect the chart
- Understanding aggregation in a hierarchy chart
Hierarchy charts are a way of visualizing hierarchies, helping you to see coding patterns or view the attribute values of cases and sources.
When you create a hierarchy chart, it displays in Detail View.
1 Select or refine the content that you want to display in the hierarchy chart. For example, you can change the data in a hierarchy chart of nodes or sources or change the order of attributes displayed on a hierarchy chart.
2 Click the panes across the top to visualize the hierarchy chart as a tree map or sunburst or to see a tabular display.
4 You can interact with your hierarchy chart to explore a specific data point in more detail or view the data underlying the chart.
5 If you want more space to view the hierarchy chart, click the disclosure triangle to collapse the panel at the top. You can also display Detail View on the right.
A tree map is a diagram that shows hierarchical data as a set of nested rectangles of varying sizes. For example, use size to represent the amount of coding at each node.
Size indicates amount, for example, the number of nodes coded or amount of coding references. The tree map is scaled to best fit the available space so the sizes of the rectangles should be considered in relation to each other, rather than as an absolute number.
Larger areas display at the top left of the chart, smaller rectangles display toward the bottom right.
A sunburst is a radial tree map. Hierarchy levels are presented as rings—the innermost ring is the top level of the hierarchy. The rings are divided into segments which represent children in the hierarchy. Use this diagram to see the largest contributing segments (children) in a hierarchy of multiple levels.
The summary pane displays the underlying data in a table format.
Sources With a hierarchy chart of sources, you can:
Compare the amount of coding of your sources—are some sources more heavily coded than others?
Identify sources with most coding references at specific nodes—for example, which sources contain the most coding references to Economy?
Nodes With a hierarchy chart of nodes, you can:
Compare the amount of coding at your nodes—do some nodes contain more coding references than others?
Visualize prominent themes in your project
Identify areas that need further investigation or research
Check that you have consulted a variety of sources—for example, have I relied too heavily on journal articles that are more than ten years old?
View the demographic spread of your survey respondents
Hierarchy charts are most useful when you want an overview of data that shows multiple levels of your hierarchy at once.
Use a tree map to analyze your hierarchies in one view and compare relative sizes of different aspects of your data. It's easier to compare rectangles than curved segments, so use this chart when you want to compare lots of data and make the most use of your space. It's also easier to read the labels on this chart. Tree maps expand to take up the size of the Detail View.
Use the sunburst to see every level of your hierarchy and the contributing segments in each group. Use this chart when you have a deep hierarchy to visualize. It's hard to fit a circular chart into a rectangular window, so the sunburst is best used when space is not an issue. You can draw attention to a single segment or highlight a particular node.
Hierarchy charts use size to convey meaning. You can use color to show additional information. This means you can size items in the chart to compare them by one metric, and then color the items to compare them by a second metric—in the same chart.
By default, hierarchy charts are sized by items coded, and colored by hierarchy. Use the Size By and Color By options on the ribbon to change the way data is displayed on the chart. Refer to Change the appearance or content of a hierarchy chart for more information.
By default, hierarchy charts are colored by Hierarchy and sized by Coding References. This means that each different parent item in the hierarchy is assigned a different color, and the child and grandchild items are lighter shades of the parent color. The items are sized by the number of items coded. This combination makes it really easy to differentiate the branches of your hierarchy.
This example is both sized and colored by the number of Coding References. Economy has the most coding references so it is the largest area and has the darkest color. Development is smaller and has a lighter color because it has less coding references. This combination uses color and size to emphasize what you're trying to show.
This example is colored by Items Coded and sized by Coding References. Economy has the most coding references so it is the largest area and Development has the darkest color because it codes the most sources. This combination results in an information-rich visualization, using the same chart to convey two different aspects of information about your project data.
Hierarchy charts represent your data as aggregated. Even if you have not turned on node aggregation, any node with child nodes will be represented on a hierarchy chart as a parent node including its children.
When you hover over an area of the chart, a tooltip displays information for that item and specifies the number of items coded and number of coding references that relate directly to the parent, as well as an aggregated figure of parent plus children.
Hierarchy charts size a node based on whether or not it has children—therefore any node containing child nodes may appear larger than its actual coding. For example, if your hierarchy chart displays hierarchical nodes, and you have chosen to size by number of items coded, you may find that parent nodes are sized larger than expected. Use the tooltip information to understand the data underlying an area.