Once you've classified open-ended or written response data, either inside Displayr or in an outside program (such as Excel), you can reuse the response-to-category mapping (called the Code Frame) in the Text Categorization tool to classify similar or remaining responses. This allows you to use the old classifications as a better starting point within the Text Categorization module, which helps speed up classification of new responses using Displayr AI and similarity matching features. This is useful when you have:
- Similar surveys with similar open-ended or spontaneous awareness text variables, but different respondents.
- Different reports/documents using the same survey data.
- Classifications you created outside the Text Categorization tool (such as in Excel or using a List of Items output), and you want to use these as a starting point within the Text Categorization tool.
This article guides you through how to reuse the categorization by exporting and importing a QCodes file (which contains the Code Frame) into the Text Categorization module. If you want to reuse a categorization you created in Displayr within the same document on a separate variable, see How to Reuse a Categorization (Code Frame) on a New Variable.
This article is broken out into the following sections:
Requirements
- A categorized variable or variable set created using the Text Categorization module. These variables typically end with "- Categorized" in their name.
- A separate Displayr document with a text variable you want to apply the existing categorization to.
- Familiarity with Displayr's Text Categorization module, see How to Classify Text Data.
Method - Export a QCodes file
If you've used the Text Categorization tool to classify text data, you can export a QCodes file from the tool directly, otherwise, you will need to use an automation to generate a QCodes file.
Export a QCodes file from the Text Categorization tool
- Select the categorized variable set in the Data Sources tree (usually denoted by - Categorized).
- In Properties
, go to Data > Attributes > Text Categorization.
- Click the Export button at the bottom of the categorization screen.
- Displayr automatically downloads and saves a .QCodes file to your local downloads folder.
Generate a QCodes file from a variable set
If your classification was done in Excel or using some other method within Displayr, you can run an automation that uses the original text variable that was classified and the resulting variable set of classification themes/categories to generate a QCodes file to use in the Text Categorization tool.
- Go to a Page in your document or from the Report tree, click + > Page to create a new one.
- In the Data Sources tree, select the Text variable with the responses that were classified and the Variable Set (either Nominal or Binary-Multi) that includes the classifications - note these must be in the same Data Set.
- From the toolbar, click Anything
> Advanced Analysis > Text Analysis > Create QCodes File.
- A Select location to save box will appear. Select the location on your Displayr Cloud Drive where you want to save the QCodes file.
- In the File name field, enter an appropriate name for your QCodes file. Note: The filename must end in .QCodes
- Click Save. Displayr saves the QCodes file to your Displayr Cloud Drive.
- To download the QCodes file to use in the Text Categorization tool, click your initial in the top right-hand corner and select Displayr Cloud Drive.
-
In the Displayr Cloud Drive, find your QCodes file and click on the Name to download it.
Method - Import a QCodes file into the Text Categorization tool
- Go to your Displayr Document you plan to update using the existing categorization.
- Select the Text variable you would like to classify in the Data Sources tree.
- Hover and click + > Text Categorization.
- Select Multiple themes or Only one theme, depending on the structure of your responses.
- Click Start.
- In the Text Categorization window, click the Import button.
- Select the QCodes file you downloaded and select Open.
- The categories created from your original categorization will appear in the Themes pane on the left.
- [Optional] You can edit or create additional themes based on the new text variable, see How to Classify Text Data for details.
- Use the Classify button to classify the text responses into the themes using AI or you can manually classify responses into the themes using other matching and filtering features, see How to Classify Text Data for steps on how to classify responses to themes.
- Once done, click Save.
- Displayr adds a new variable to the dataset, with - Categorized in its name, representing the newly categorized variable.
Next
How to Reuse an Existing Categorization (Code Frame) on a Different Variable