Types of queries

NVivo provides several different types of Queries that you can use to explore your data. A focused query will display results that can help you to identify patterns, themes, gaps, and inconsistencies. Some queries can be run early in the project to help you code, find potential relationships or check that your coding is meeting your needs. Other queries are best employed when the coding process has been finished (for example, to test a hypothesis).

There are three types of Queries:

  • Text-mining queries: search text material in the database for specific words or phrases. This group includes Word Frequency and Text Search queries.
  • Theory-building queries: help you explore relationships between items. This group includes Coding and Matrix Coding queries.
  • Clerical queries: provide output to help you manage components of the analytic process. This group includes Compound and Group queries

The PDF document below outlines each type of query, together with its purpose and prerequisites (if any).

Managing queries

To run a query: select the type you want from the Query tab at the top of the screen, choose your parameters, then select the Run Query button to get your results.

By default, results of queries are not automatically stored. Select the Save Results button to store the results in your project (in the Queries/Results folder). You can also save your query parameters in case you want to run it again once more sources have been added and coded. To save your query parameters select the Add to Project... button (it will be stored in the Queries/Queries folder).

To access saved queries and results select the Queries section from the left hand navigation pane.

The PDF below has more details on managing queries.

Word frequency queries

Word frequency queries tell you how frequently words appear. You can run a Word frequency query on all words, or limit the number of words. The results wil be sorted with the most frequently used words first.

Word frequency queries can be used as a starting point for a text search, as a word count, or to analyse, for example, the most common responses in open-ended questions.

Note: the video tutorial below is provided by QSR for NVivo 10. The interface in NVivo 11 Pro looks slightly different; pop-up boxes now open in the Display view section of the main window. Core functions and the steps to be followed remain the same.

Text search queries

Text search queries look for keywords, or expressions, using boolean operators (and, or, not) or proximity operators (number of words, or words in the same paragraph).

Text search queries can be used to find ideas, investigate linguistic expressions, or analyse the content in which participants use specific words or expressions. Text searches can also be converted into nodes; this process is known as brush coding using keywords.

Note: the video tutorial below is provided by QSR for NVivo 10. The interface in NVivo 11 Pro looks slightly different; pop-up boxes now open in the Display view section of the main window. Core functions and the steps to be followed remain the same.

Coding queries

Coding queries find text or other data in response to a single, sometimes complex question, involving multiple nodes and/or attribute values. Simple coding queries are used to look up a node when it is used in a particular context. Advanced queries respond to complex questions that require a combination of nodes or of attribute values and nodes. Boolean terms (and, or, not) as well as proximity operators can be used within this query.

Coding queries can be used to explore associations between topics and investigate exceptions.

Matrix coding queries

Matrix coding queries allow you to compare results across different groups and test hypothesis. For example, you may want to check whether there are differences in the emotional responses of females and males. Double-clicking on a cell of the matrix allows you to access the qualitative content- both detailed content and numbers are useful products from this query.

Matrix coding queries can be used to compare your participants using demographic data or other variables that you have stored as attributes. You can also use matrix coding queries to run within-case and cross-case analyses.

Note: the video tutorial below is provided by QSR for NVivo 10. The interface in NVivo 11 Pro looks slightly different; pop-up boxes now open in the Display view section of the main window. Core functions and the steps to be followed remain the same.