All Collections
Learn about Chattermill
Using Reports
How we calculate Positivity & Negativity Index
How we calculate Positivity & Negativity Index

Understand how we analyze theme mentions

Billie Bradley avatar
Written by Billie Bradley
Updated over a week ago

Negativity Index

Negativity index measures the number of negative theme mentions within a set of responses based on your filter selection (e.g. NPS responses in the past 30 days). We then normalise this number per 100 responses (similar to how you calculate a percentage).

Unlike a percentage, this number can be more than 100 because one response could have many theme mentions.


In the example above, the negativity index is calculated as follows:

We count all negative mentions from responses from the last 30 days, and then divide them by the total number of responses and multiply by 100.

The calculation becomes a bit more complicated when you introduce breakdowns.

If we take the example of the Build Attributes category, here we have counted all negative theme mentions in responses that contain the category Build Attributes. This is then divided by the total number of responses (for all categories), and multiplied by 100.

Another example would be if you wanted to add a breakdown by theme:

In the screenshot above you can see we've broken down by themes in the Build Attributes category, within this category the theme "Designing Process" exists.

To calculate the negativity index for the theme Designing Process, we first count all the negative mentions of the theme (ignoring positive or neutral ones). We then divide this by the total number of responses (for all categories and themes) and then multiply this by 100. As a simple example, if among 1000 responses we had 200 negative mentions of the Theme Design Process, the Negativity Index for this theme would be

(200/1000) * 100 = 0.2 | ➡️ | 0.2 * 100 = 20

The Negativity Index of the Design Process Theme would be 20 mentions per 100 responses.

Positivity Index

Positivity works in exactly the same way as negativity index, but instead you are counting the number of positive theme mentions per 100 responses.

Did this answer your question?