Analytics: Correlation between pillars
We usually use this report to understand how we can improve a given pillar by using the correlation that exists between them.
What do you mean?
First, it is important to understand what is a correlation between the pillars. When we talk that two pillars correlate, it means that for that company/team they usually have the same behavior of variation of growth or fall. The closer this variation between two pillars, the greater the correlation that exists between them. To summarize, if the relationship with the manager has similar growths and falls to the feedback, it means that these two pillars have a high level of correlation between them.
I understand what correlation means, but I can't understand why it is useful to help improve the pillars of engagement.
It is quite simple. Imagine that you want to improve the pillar of "Flying the flag", how do you do to work something that will help improve the pillar of "Flying the flag", which also normally improves your rate of eNPS?
One of the possibilities is to create an action plan based on the answers to the questions of the pillar itself. Another possibility is to take advantage of the pillars with correlation with "Flying the flag".
Let me give you a practical example: let's use the image below to conduct a study.
Vertically and horizontally we have the pillars of engagement and when the line meets the column we enter the level of correlation between those two pillars.
Below we have an example where the pillar "Flying the flag" has an average correlation with "Alignment" and very low with "Well-being", "Happiness", "Satisfaction" and "Relationship with the team".
Understanding the mathematics of the pillar correlation report
To measure the correlation between the pillars of engagement we use the Pearson correlation coefficient, represented by values between -1 and 1 .
To simplify the interpretation, the platform presents the correlations in correlation bands, with the following intervals:
- Above 0.75: Very High
- Above 0.5: High
- Above 0.25: Medium
- Above 0: Low
- Above -1: Very low
Note that even the values in red (0.95) are very close to 1, so on the platform we have some scenarios where the "Very High" also appears in red.
The calculation uses scatter plots between each engagement pillar.
Note that as the score of one pillar increases, the score of the other always grows proportionally, characterizing a perfect linear correlation between these pillars. When this happens, we say that one pillar has a very strong correlation with another.
But what does that mean?
It means that to increase the score of the pillar "Flying the flag", I can work the "Alignment" of the company. Of course, because the correlation is average, the score of "Flying the flag" will not be so impacted by the improvement of alignment, but in cases where the correlation is very high, the impact of the score will be much greater.
Whereas looking at the lower pillars, we see that these pillars won't have a strong representation in the rise and fall of the note of the pillar "Flying the flag".
It doesn't stop there. Have you ever stopped to think that this says a lot about the company's culture?
In this example, the alignment has a relationship of importance for people to feel promoters and proud of the company. Remembering that this is the case for this company, each company has its level of correlation between these pillars.
I found it very interesting, but where do I access it?
In the left side menu, you now have the Analytics option.
Do it differently! Let's Question, Act, and Evolve. Talk to our team right now using the side chat or take the opportunity to read more about the platform.
What do you mean?
First, it is important to understand what is a correlation between the pillars. When we talk that two pillars correlate, it means that for that company/team they usually have the same behavior of variation of growth or fall. The closer this variation between two pillars, the greater the correlation that exists between them. To summarize, if the relationship with the manager has similar growths and falls to the feedback, it means that these two pillars have a high level of correlation between them.
I understand what correlation means, but I can't understand why it is useful to help improve the pillars of engagement.
It is quite simple. Imagine that you want to improve the pillar of "Flying the flag", how do you do to work something that will help improve the pillar of "Flying the flag", which also normally improves your rate of eNPS?
One of the possibilities is to create an action plan based on the answers to the questions of the pillar itself. Another possibility is to take advantage of the pillars with correlation with "Flying the flag".
Let me give you a practical example: let's use the image below to conduct a study.
Vertically and horizontally we have the pillars of engagement and when the line meets the column we enter the level of correlation between those two pillars.
Below we have an example where the pillar "Flying the flag" has an average correlation with "Alignment" and very low with "Well-being", "Happiness", "Satisfaction" and "Relationship with the team".
Understanding the mathematics of the pillar correlation report
To measure the correlation between the pillars of engagement we use the Pearson correlation coefficient, represented by values between -1 and 1 .
To simplify the interpretation, the platform presents the correlations in correlation bands, with the following intervals:
- Above 0.75: Very High
- Above 0.5: High
- Above 0.25: Medium
- Above 0: Low
- Above -1: Very low
Note that even the values in red (0.95) are very close to 1, so on the platform we have some scenarios where the "Very High" also appears in red.
The calculation uses scatter plots between each engagement pillar.
Note that as the score of one pillar increases, the score of the other always grows proportionally, characterizing a perfect linear correlation between these pillars. When this happens, we say that one pillar has a very strong correlation with another.
But what does that mean?
It means that to increase the score of the pillar "Flying the flag", I can work the "Alignment" of the company. Of course, because the correlation is average, the score of "Flying the flag" will not be so impacted by the improvement of alignment, but in cases where the correlation is very high, the impact of the score will be much greater.
Whereas looking at the lower pillars, we see that these pillars won't have a strong representation in the rise and fall of the note of the pillar "Flying the flag".
It doesn't stop there. Have you ever stopped to think that this says a lot about the company's culture?
In this example, the alignment has a relationship of importance for people to feel promoters and proud of the company. Remembering that this is the case for this company, each company has its level of correlation between these pillars.
I found it very interesting, but where do I access it?
In the left side menu, you now have the Analytics option.
Do it differently! Let's Question, Act, and Evolve. Talk to our team right now using the side chat or take the opportunity to read more about the platform.
Updated on: 26/05/2021
Thank you!