Codes represent the concepts to be tagged in the research as well as the organization and relation between these concepts.
Nodes are the primary concepts or codes. Nodes are created under the Create tab of the toolbar at the top.
Nodes can be hierarchically arranged.
Note: great thought should be paid to thinking about the code schema before coding files with these codes commences. Nodes can always be added, but review of coded files will have to be done again when new codes are added.
Double click a node to see every fragment coded with that node in each file.
Each document will be linked, followed by each fragment with the specific node.
By default, text representations are included in the default view.
On the right side, there are tabs that represent each type of document tagged by this node.
For example, the default view will display the transcript of an interview, but clicking on Video. will display the transcript and the video snippet.
The proportion of each document that is captured by each coded section is displayed above the coded section.
Create Case Classifications
Classifications are the attributes associated with objects in the research. Both files and cases can be classified.
Case classifications store attributes of the research subjects in the project. In the example above, the characteristics for Barbara are displayed. Think of these as hierarchically nested nodes under each interviewee.
Case classifications can also be imported from a spreadsheet, where each case is a row, and each column is an attribute of the case, including that case's name. The import wizard will let you define the case names and other attribute names.
Coding Files
Coding is the process of associating concepts (or codes) to your files. The codes leave the files unaltered, and selecting a code will bring up all associated references to files.
The sample project contains files that have already been coded. To see the existing codes and associated text, enable Coding Stripes from the toolbar at the top (the Document tab, PDF tab, or tab for whatever type of file is displayed.
When there are a lot of codes, specific nodes can be selected instead. Most Coding under Coding Stripes is also a good option.
To code a piece of text, first select the text to code (highlighted above). Then, there are several ways to code the text.
Below the document is a toolbar. Type the name of a node following "Code At." You can also code cases and relationships from the left hand side of that toolbar (expanded dropdown box above).
Right click and click "Code..." Then select the code or codes to associate.
Click Code from the toolbar at the top.
Tip: if a complex coding structure is present, it might easier to code files in two or more waves, each wave focused on a few related codes that are easy to remember.
Tip: Codes can overlap, and coding narrowly increases the value of your work. For example, there may be a 4 sentence paragraph concerning economic development, but if the third sentence is focused on tourism, for example, and the second focused on agriculture, coding those two sentences as both economic and subcategories of economic development helps to identify specific examples.
Tip: Codes can be removed by selecting text and using the uncode command.
The coding tools that are available will depend on the type of file being coded. In this example, a transcript in spreadsheet form is coded (in pink). In addition, the timeline at the top is also coded so that the video can also be searched based on code.
Autocoding
The autocoding wizard in the toolbar at the top allows for rapid coding of well structured data.
In the interview transcript above, data were structured in Microsoft Word to identify the names of the speakers. Paragraph styles can also be used to identify different types information, such as questions and replies.
In this example, the case "Barbara" is attached to each of her responses automatically, saving coding time. Question numbers could also be attached, so each response to question 1 could be looked at, for example.
Query Types
Queries allow for searching for specific words, codes, or combinations of codes to extract text from the research data.
Queries can be accessed from the Explore tab in the toolbar at the top.
In this example, results from a text search are displayed. Again, note the sidebar on the right showing the variety of file types that are represented by the results. The Reference tab will display any text for all search results, the rest will display file type-specific results.
The coding query allows for queries to recall text that has been tagged by one or more codes.
In the example above, results are displayed for all text that were tagged with the Economy code or any if its sub-nodes, and the attitudes code and any of its sub-nodes.
Again, note the sidebar on the right, which delineates the results by the type of file that is contained in the project.
Use the query wizard for these and more complex queries, and see this NVivo documentation for more information on the types of queries that are available.
