Personality tests are one of the most popular online quizzes because everyone seems to want to know how they measure up to friends and neighbours.
But many of these tests are fake and have no scientific data to back up the findings.
And psychologists are quick to point out that it’s difficult to shoehorn someone into a specific personality stereotype.
Advances in data technology have allowed a team of researchers to analyse a quiz complete by 1.5 million people in a bid to better identify personality types – and reckon the numbers reveal four distinct types of people.
Four different personalities
They are dubbed average, reserved, role model and self-centred.
- Average people are the most common personality category, who are neurotic and extrovert, but low on openness.
- Reserved personalities are stable, agreeable and conscientious, but not open, neurotic or extrovert.
- Role models are good leaders, dependable and open to change, scoring high in every trait except neuroticism.
- The self-centred score high as extroverts nut low in every other trait.
Underlying personality traits
The underlying traits that mix to give personality types are:
- Neurotics, who have emotional outbursts like anger, worry and sadness as well as showing interpersonal sensitivity.
- Extroverts are dominant, talkative, sociable and enjoy being with others.
- Openness is displaying acceptance of new art, ideas and values.
- Agreeable people tend to go along with others without asserting their opinions or ideas.
- The conscientious are careful, punctual, follow rules and work hard
“Machine learning and data science are promising but can be seen as a little bit of a religion,” . “You still need to test your results. We developed a new method to guide people to solve the clustering problem to test the findings,” said report authors Luis Amaral and William Revelle, of NorthWestern University, Chicago, USA.
Their algorithm first revealed about 16 personality clusters using traditional clustering methods. The researchers then imposed additional constraints, winnowing down the clusters to four distinct personality types.
“The data came back, and they kept coming up with the same four clusters at higher densities than you’d expect by chance, and you can show by replication that this is statistically unlikely. The methodology is the main part of the paper’s contribution to science.
“We know teen boys behave in self-centred ways. If the data were correct and sifted for demographics, they would turn out to be the biggest cluster of people.”