How to Compare Sentiment by Entities in Text Data

Sometimes, instead of looking for keywords or grouping responses into themes, you want to look for entities in your text responses. This helps you identify what types of items were mentioned and adds them into a generic group of what they are, such as Country, Person, Number, etc. This article describes how to create an entity list from open-ended text data:

and sentiment scores calculated from an open-ended text variable:

And crosstab them to see sentiment scores by saved categories from an entity list, for example:

Requirements

You will need at least two Text variables to perform automatic categorization. Text variables are represented by a capital A next to the variable and can be found in the Data Sources tree:

One text variable should be suitable for creating an entity list, and another text variable should be used to calculate sentiment.

Method

Step 1 - Create an entity list

1. Hover in the Report tree, and click + > Advanced Analysis > Text Analysis > Automatic Categorization > Entity Extraction.
2. In Properties , under Data > Text variable, select the text variable you want to automatically code into an entity list.
3. From Data > Save Variable(s), click Categories, and a new variable will be inserted into your Data Sources tree.

NOTE: The variables created from this using Save Variable(s) > Categories may become invalid and need to be deleted and recreated if the output has changed, either due to the input text variable being modified or the input settings modified.

Step 2 - Calculate sentiment

1. Select the text variable from the Data Sources tree.
2. Click in the '+' next to your selected text variable > Numeric Transformation > Sentiment.
3. A new variable with a Numeric structure will appear in your Data Sources tree that represents the sentiment score.

Step 3 - Compare sentiment by entities

1. Create a crosstab by dragging the entity list categories variable from Step 1 above onto a Page.
2. Drag the sentiment score variable that was calculated in Step 2 above into the Columns of the crosstab table.

Next

How to Calculate Sentiment Scores for Open-Ended Responses

How To Automatically Classify Unstructured Text Data Into an Entity List