Lattice's Natural Language Processing engine leverages artificial intelligence and machine learning technology to provide accurate and rapid sentiment tagging from comment responses. Engagement survey sentiment highlights whether survey comments are positive, neutral, or negative, which can reduce the amount of time admins spend identifying focus areas to improve employee engagement.
Note: The engagement score and sentiment score are unrelated. Engagement scores are based on the question responses (Strongly disagree to strongly agree). While sentiment scores are based on language provided in the free-text comments.
Sentiment scoring
One of three faces indicates the overall sentiment: happy, neutral, and unhappy. The faces correspond to the following numeric ranges out of 10:
- 1-3.9: Negative Score
- 4-7: Neutral Score
- 7+: Positive Score
Overall sentiment
Once employees start to submit survey responses and the anonymity threshold has been met, survey reporting will show an overall survey sentiment calculated from the free-text comments.
Sentiment for questions, themes, and comments
In the List view of the Theme, Questions, and Comments sections, each row will have a calculated sentiment. Hovering over the sentiment provides more information on the exact score out of 10.
Filter comments by sentiment
You can filter comments by sentiment. To filter comment responses:
- Navigate to the Comments tab.
- Under the Filter by section, navigate to Sentiment and select a sentiment.