The Temperament of a Kenyan Online Influencer

If you’ve been to the Kenyan Twittersphere, you must have read quirky tweets from Muthoni Maingi.  When she’s not overseeing the Enterprise and Consumer Business Unit at Safaricom, her energies are spent entertaining 16.5 k followers on Twitter under the nom de plume @NonieMG. Today, our universes collide.  In her characteristic witty remarks, inquired if I can look through her ‘soul’ with a data science lens 😉

Challenge was accepted. In the process of studying handwritings, Graphologists discovered that humans release their emotions  while writing, thus making it possible to identify the emotional state of a writer from the curvature, depth and size of each letter used. Much in the same manner we are able to peer into Muthoni’s soul by  analyzing her choice of words, punctuactions, grammar, sentiment et ceteta from her tweets.

Straight into the engine room, I wrote a code in Python to download her latests 33,266 tweets. After that, I feed the tweets to IBMs Waston Personality Insight tool. IBM Watson is an Artificial Intelligence system developed by  IBM’s DeepQA research project. Initially, it was built to automatically answer question but it was further developed to play chess and do other amazing stuff like predict diseases as well as predict human personalities.

Before proceeding, we should look at the personality types that Watson predicts. In the psychology field of personality theory, five distinct factors that define personalities namely; Openness to Experience, Conscientiousness, Extraversion, Agreeableness, and Neuroticism.

  • Openness to experience: – appreciation for art, emotion, adventure, unusual ideas, curiosity, and variety of experience.
  • Conscientiousness:- a tendency to be organized and dependable, show self-discipline, act dutifully, aim for achievement, and prefer planned rather than spontaneous behavior.
  • Extraversion: – energy, positive emotions,assertiveness, sociability and the tendency to seek stimulation in the company of others, and talkativeness.
  • Agreeableness:- a tendency to be compassionate and cooperative rather than suspicious and antagonistic towards others.
  • Neuroticism: – the tendency to experience unpleasant emotions easily, such as anger, anxiety, depression, and vulnerability.

Without further ado, let’s have a look at how Muthoni fared on the Big Five personality traits.  The IBM Watson Personality Insight tool calculates a percentile to show how one compares to the rest of the population in each trait. it further define sub traits for each of the traits.

Big Five Personality Traits Predictions for Muthoni Maingi

Voila! As you can see in the diagram above non of the percentile are above 50%, meaning in the five personality traits, she compares lower than the average person. The odd factor is the lowest score is on extraversion which would be assumed to be high for an online influencer.  Let’s compare her to a celebrity – Oprah Winfrey.

Oprah Winfrey Personality Test

There’s definitely a stack difference between the two. Since the trait percentile score are made of aggregate of other sub-traints it would be good to dive into the sub scores to understand the nuances. A good trait to begin with is Neuroticism which she sored the highest.  Neurotcism defines the tendency to experience unpleasant emotions easily, such as anger, anxiety, depression, and vulnerability. Under that, she scores 92 percent for self-consciousness (an acute sense of one’s self-awareness,appearance, or actions), yet score 9 percent on Immoderation (the quality of being excessive and lacking in restraint; overindulgence) 🙂

The devil is in the details 😀 . Another good example is under extraversion. In it, she scores 90 percent on excitement seeking but 14 percent on assertiveness (confident and forceful behaviour). I have attached the output file for thorough indulgence.

Personality prediction output:

NonieMG Tweets:

  1. Well articulated data story. You demystified it and made it relatable to day to day.
    You can also find the OCEAN model in its original format at and run your own personality analytics. It also comes with complete with embed code of ots algorithm for deployment in various use cases e.g HR for interviews , imsuarance premium , Credit score referencing amd other relevant personality assements ensembles.



    1. Hi Tim,

      Thank you for the kind words. I’ll definitely check the OCEAN model in depth for future applications. Didn’t know they utilize personality traits for credit scoring.

      Chris Orwa



  2. Hi Chris, enjoyed this analysis, can you email me on with a phone contact for a follow up conversation, i



    1. Hi Irungu,

      Glad you enjoyed the article. Response is in your inbox.

      Chris Orwa



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: