Metrics are calculations performed on the terms being analyzed. The following metrics are available:
- Score: The score metric computes the dataset score for each term. Dataset scores are defined with the dataset and may be NPS, MEAN, or TOP[N] Box. If not score field/method is defined for the dataset, the score is the % positive sentiment for the term.
- Score Difference: The “absolute impact” metric computes the absolute impact each term has on the overall score. This is calculated (termscore – overallscore)
- Sentiment Index: The sentiment index metric uses a -100 to 100 scale to compute a sentiment index for each term.
- Impact: The “relative impact” metric computes the relative impact each term has on the overall score. This is calculated (termscore – overallscore) * (termFrequency / overallN)
- Sentiment Grade: The sentiment grade metric converts the % positive satisfaction for each term into an academic A-F grade where terms with 90%+ positive sentiment have grade A, 80% have grace B, etc.
Metric values are color coded to highlight values. The method used to color code metrics varies by metric. Below are some basic rules for how color coding is performed.
- Score: The overall score is computed. Values greater than or equal to the overall score are colored green. Values below the overall score are colored red.
- Score Difference: Differences of zero or greater are colored green. Differences below zero are colored red.
- Impact: Impacts of zero or greater are colored green. Impacts below zero are colored red.
- Sentiment Index: Sentiment Index values of zero or greater are colored green. Sentiment Index values below zero are colored red.
- Sentiment Grade: Grades of A or B are colored green. Grades of C are colored yellow. Grades of D are colored orange. Grades of F are colored red.