Determine emotional diversity to predict life satisfaction

Hi all

I have a sample of 76 participants who answered a survey about which 30 emotions they have experienced during the week and at which frequency (from 1-5 where 1 = never and 5 = all the time). The 30 emotions are divided into 8 different groups based on arousal level and valence (negative/positive emotions in this case).
The same participants also answered a survey about life satisfaction.

I would like to do a regression model to see if a participants emotional diversity can predict life satisfaction and thought I might could do the following:

Calculate a diversity score for each participant that describes the overall emotional diversity for each participant where the highest possible score represents a high frequency of emotions experienced in all 8 groups and a low score represents a low frequency of emotions experienced in only a few of the groups (e.g. a participant that only experienced low aroused negative emotions during the week).

And then use the diversity score as a predictor variable and the life satisfaction score as outcome in the regression model.

But, I can't figure out how to calculate the diversity score. If I just take the sum of the values of all 30 questions, then someone with a high frequency of only positive emotions and no negative emotions could get the same sum as someone with low to medium frequency of both positive and negative emotions.

Any ideas of how to do this?

Thank you so much!
Last edited:


TS Contributor
Seemingly, you differentiate between low to medium frequency vs. high frequency.
It is not totally clear from the description if you want to base the diversity measure
on the 30 single emotions, or on the 8 groups. If it is the former, you could maybe
just count the number of emotions with frequency > medium, and use that count
variable as your predictor.

With kind regards

Thank you so much for your reply.
I ended up using the Shannon diversity index to determine the diversity of the 30 emotions for each participant.