Relative Difference is the metric used to compare frequencies across two language sets. It tells you how prevalent a particular language element is in one data set, versus another.
It is calculated by dividing the normalised frequency of a particular word, topic, phrase, grammar or emotional element in the primary language set against the normalised frequency of the same element in the comparison language set.
Understanding infinite (∞) relative differences
When a language element is statistically significant in one language set and completely absent from the other, the relative difference will be shown as ∞.
Please be careful to understand that an infinite relative difference doesn’t necessarily equal a relevant insight. For example, names and other source specific words will commonly surface as having infinite relative differences.
For example, comparing a Harry Potter book to The Great Gatsby would almost certainly surface 'Hermione' as having an ∞ relative difference, but this doesn't tell us much about J.K. Rowling's writing style.
To discern whether a discovery you’ve made is interesting or insightful, it’s helpful to ask yourself whether you personally find it surprising. If yes, then you’re likely on to something... 😮 💡