For years, researchers have been working away at just what makes a person who they are and defining each distinct personality type into a unique category. With personality quizzes everywhere online, it is something that we take for granted, we each have a certain personality type we just don't all know what that is. Now, <a href="https://www.nature.com/articles/s41562-018-0419-z">a new study</a> has taken a look at some of the largest online data pools, mostly from personality quizzes, and determined there are four distinct "types" of personalities.
<iframe width="560" height="315" src="https://www.youtube.com/embed/SwhJ4UUCKAA" frameborder="0" allow="autoplay; encrypted-media" allowfullscreen></iframe>
Researchers used a new methodology for the study that had yet to be applied to personality research: a sorting algorithm. With the help of computers, the study is described in detail in a new paper in Nature Human Behavior. The study is rigorous and replicable and could play a large role in moving the stereotyping of personalities into a more serious scientific realm. According to the researchers, personality "type" may not be the right term, for the purpose of this study personality "clusters" would be a more accurate term.
The paper was co-authored by William Revelle of Northwestern University and raises some new questions about the previously established literature surrounding personality such as the widely popular Myers-Briggs Type Indicator. Revelle is not afraid to voice his opposition to the Myers-Briggs Type Indicator and he is not alone. Many researchers studying personality today seem to opt for a more fluid definition of personality defined as a set of continuous dimensions.
Related coverage: <a href="https://thegoldwater.com/news/29932-Researchers-Say-AI-Can-Smell-Illness-In-Someone-s-Breath">Researchers Say AI Can Smell Illness In Someone's Breath</a>
The latest study using a new algorithm and the world's largest pool of personality data identified four dominant clusters in the overall distribution of personality traits. Revelle refers to them as "lumps in the batter" and proposed the analogy of how people tend to concentrate in cities in the United States. If you divide the country into four regions, north, south, east, and west, and study how the population is distributed you will probably find the highest concentration of people live in dense cities such as Houston, Los Angeles, Chicago, or New York.
Revelle added, "But to describe everyone as living in one of those four cities is a mistake. What we're describing is the likelihood of being at certain parts of that distribution; we're not saying that everyone is in one of those four categories." The researchers pulled from a pool of publicly available data from online quizzes taken by around 1.5 million people across the globe. The data was then plotted with the help of an algorithm into the Big Five basic personality traits: neuroticism, extraversion, openness, agreeableness, and conscientiousness.
Currently, the Big Five is the professional standard for psychologists and those who study personality. Once the algorithm had been applied to the data set, researchers were left with four "types" of personality. Revelle explains it wasn't as simple as that at first since he and the study's co-author Luis Amaral initially found 16 distinct clusters prompting instant skepticism. "That was ridiculous," Revelle said of the first 16 clusters. He admitted that at first he didn't think there were any types at all lurking in the data and set to work with his fellow researchers to refine the results.
Related coverage: <a href="https://thegoldwater.com/news/35157-Blaire-White-Easier-to-Come-out-as-Transgender-than-Conservative">Blaire White: Easier to Come out as Transgender than Conservative</a>
"These statistical learning algorithms do not automatically produce the right answer," Revelle said. "You need to then compare it to random solutions." It was that second step that made all the difference. They soon found that by imposing extra constraints to narrow down the results, they ended up with four distinct personality clusters. Here are the four personality types the refined data yielded:
<b>Average:</b> Those who score high in neuroticism and extraversion but low in trait openness. It is also the most common category with a higher likelihood of women being in it than men.
<b>Reserved:</b> This personality type is more emotionally stable and tends to be less open or neurotic. They typically score lower on extraversion but tend to be more agreeable and conscientious.
<b>Role Models:</b> This category is for people who score high in every trait except for neuroticism. The likelihood of someone fitting this trait increases drastically with age.
<b>Self-Centered:</b> This may seem like an obvious one, but people in this category tend to score very high in extraversion, but low in trait openness, agreeableness, and conscientiousness. Revelle claims most teenage boys fall into this category but may eventually mature out of it. Similar to the Role Models category, the number of people who fall under Self-Centered will decrease drastically with age.
Related coverage: <a href="https://thegoldwater.com/news/34670-Facebook-s-Secret-Ranks-for-Users-Exposed-in-Invasive-User-Score-Scandal">Facebook’s Secret Ranks for Users Exposed in Invasive User Score Scandal</a>
The researchers used one particular data set initially, but after the first analysis, they then replicated the same result on two other independent data sets. This proved their methodology is replicable, at least when it comes to similarly large data sets. Thankfully, with the rise of the internet, such large data sets are increasingly common. Amaral added, "A study with a data set this large would not have been possible before the web."
Tips? Info? Send me a message!