Predicting Character Strengths with Social Media


Social media is widely known to produce unprecedentedly large amounts of autobiographical data to study. One way the scholars in this research paper have decided to look at the usefulness of Social Media is to use it to predict character strengths based on 24 different categories.
Active Users:
  • Twitter:
    • 330 million monthly active users
    • 500 million daily tweets


Research has mainly focused on the big 5 personality traits, however, this study uses the 24 character strengths which are described in the following picture:


24 character strengths table


 Modeling Equation Used:

Here we have some word clouds of words with high global positivity factor versus low global positivity factor:


Here's a table the authors use for describing the words used and the correlating character strength:


We can use information like the table above to realize when we might be feeling angrier. For example, you might be using words like: "hate, fuck, hell, stupid, mad" if you are angry. Other words in the following tables are great to look out for to determine where you fall in the different categories:



The Takeaway

Studies like these show us how we can predict our emotions from just the way we write. Although this is much more difficult to do than in person (body language, facial expressions, etc.), we can use our knowledge of how words are usually used to predict more accurately.

The authors' main takeaways were as follows:

"The linguistic cues in the present study provide support for the potential contribution and association of character strengths to both the hedonic (e.g., happiness or positive affect) and the eudemonic (e.g., a sense of meaning and purpose) aspects of well‐being (Wagner et al., 2019)."


How This Data is Useful:

If we can accurately predict one's character strengths from their writing, we can begin to understand an underlying theme in their personality that when studied across time can possibly predict personality changes and mood disruptions. With this, we can monitor someone and see what resources they would most likely benefit from.

I believe this is the future of mental health care. The ability to look at someone's profile and understand potential issues before they present themselves physically or negatively outwardly. This could be a powerful tool in helping those who actually need help receive the best kind of care for them personally. This both makes mental health care more accessible, but more effective and scalable. As always continue to think differently.

 - WPD

wpd signature

Work Cited:
1. Pang, Dandan, Johannes C Eichstaedt, Anneke Buffone, Barry Slaff, Willibald Ruch, and Lyle H Ungar. "The Language of Character Strengths: Predicting Morally Valued Traits on Social Media." Journal of Personality (2019): Journal of Personality, 20 May 2019. Web.

Older Post Newer Post

  • ugclaeocqh on

    Muchas gracias. ?Como puedo iniciar sesion?

  • pOSMZkAjfehLDmzW on


  • ofBspvkxZjH on


Leave a comment