In NVivo, cases represent the 'units of analysis' in your research study—for example, the people you interviewed or the communities you studied. Your project might have cases for people, places, organizations, events or other entities that you want to analyze and compare.
Cases are like containers that hold all the information related to a person, place or other unit of analysis. You can open a case to see all the information that has been coded there—for example, you could review everything that Mary said or everything about Harkers Island.
When you begin a project, it is a good idea to decide on your units of analysis and plan for the cases you will need to represent them.
Cases can have attributes (variables) such as Age, Gender or Location. For example—you might make a case for a participant called Ken, assign attributes like Age Group, Gender and Education level, and then code all of Ken's content at the case:
When collecting data, record the descriptive information to support the comparisons you want to make. For example, asking whether young people see issues differently from older people requires that you record the age of your participants. You can store these attributes directly in NVivo or in a separate file
Do I need cases?
Not all projects require cases. Whether you need to set up cases depends on your research question and methodology.
If you want to make comparisons based on the attributes of people, places, organizations or other 'units of analysis' then cases provide a useful way to do this.
But if your project does not require these types of comparisons, then there is no need to create cases. For example, you may not need cases if you are doing a literature review or conducting in-depth interviews with only a few stakeholders.
Creating cases for different types of data
In general, setting up cases involves:
- Adding one or more case classifications (and associated attributes) to your project
- Creating cases
- Assigning cases to a classification and defining attribute values
- Coding relevant content at the case
Depending on the type of data you're analyzing, NVivo provides ways to automate this process. For example—if you're working with interviews, you can automatically create cases from your interview files. Create cases automatically
Explore cases using queries
After creating cases, assigning attribute values and coding content at themes—you can use queries to make comparisons or gather the data in revealing ways.
|Review what participants over the age of forty said about water quality.|
Check for the dominant themes across different age groups and genders.
Check for coding co-occurrence - see where content is coded to environment and positive attitude or negative attitude
See all participants by community or age.
Run a Coding query, and then a Word Frequency query
Gather all the material by female participants and create a new code based on the results. Then run a Word Frequency query on this new code to check for commonly used terms. Do the same for the male participants and compare your findings.
Explore cases using charts and other visualizations
After you set up cases and code your content, you can use visualizations to look for commonalities, patterns and trends.
Compare the coding for two cases using a Comparison Diagram.
This is a useful way to check for similarities and differences between cases.
In this example, the cases Maria and Paul both have coding to the codes Jobs and cost of living and Economy.
Check the demographic spread of cases in a Hierarchy chart.
For example, you may want to check the representation of male and females in different age groups.
Use a chart to explore the coding for a case.
This gives you a feel for the dominant themes reported by a case.
Use a chart to explore cases that have coding to a selected theme.
This example shows the cases that have coding to the code Real Estate Development. It also shows the percentage of each case that is coded to the code.
It can be a useful way to see which cases mention a theme most often.
Use an Explore diagram to see the items connected to a case and then step through the related project data
See the most frequently used words for a case in a Word Cloud.
Creating case hierarchies
You can organize cases in hierarchies (parents and children) in ways that make sense for your project.
For example, you might want to organize places into broader groupings:
- Down East Communities
- Cedar Island
- Cape Lookout
- Cape Point
- Snug Harbor
If there is no logical connection between your cases (or perhaps these connections are not yet apparent) you can just add cases at the top level of the hierarchy.
You can right-click on a parent and choose to aggregate coding from all the children. When you open the parent case, you will see content coded to all the child cases. This is useful when you want to review or query the content for a broad grouping—for example, run a Word Frequency query on Down East Communities.
Summarize cases in a Framework Matrix
Framework matrices provide a way to summarize or condense your files in a grid that has rows for cases (for example, people you interviewed) and columns for codes.
Each cell in the grid represents the intersection of a case and theme—when you enter text into the cell you can create a summary of the content relevant to that case and theme.
Using the grid, you can easily scan down the columns to compare cases or across to look more closely at an individual case. Framework matrices