Tag: self-awareness

Part 1 of 2

The short version: Of the four popular personality tests, the Big Five has the strongest scientific support, CliftonStrengths is a practical tool for work, and the MBTI and Enneagram work better as conversation starters than as facts. Whichever you take, you can paste your results into an AI and ask it to challenge them, not flatter you, to understand yourself more clearly.

Most of us have taken a personality test and thought, “That’s me.” Then we closed the tab and never used it again.

I think that’s a missed opportunity. A good result gives you words for patterns you already live in. It can also help you understand the people around you, who are running on a different setup than you are.

There’s a newer, more fun reason to care, too. With AI, information about yourself is worth more than gold. The better your data on how you think, decide, and react, the more useful the answers you get when you ask AI for help, whether that’s with a hard conversation, a career move, or a habit you can’t seem to break. It can also just be interesting- finding out why you do what you do is one of the more exciting puzzles you’ll ever get to work on.

In the past year, I took four different personality tests to understand myself in a more data-driven way than just vibes and emotions. At first, they seemed to contradict each other. I’ll get to what that taught me in Part 2. First, the basics.

Not every test deserves the same trust. Here’s what four popular ones are, where they came from, and how much weight I give each. Then I’ll show you how I use AI to get more out of the results.

Four tests, from most to least scientifically supported

The Big Five is the one researchers rely on. It started with a simple idea: the traits that matter most in people should show up as words in everyday language. In 1936, psychologists Gordon Allport and Henry Odbert pulled over 4,500 trait words out of the dictionary. Over the next few decades, researchers kept statistically sorting those words, and the same five groups kept appearing: openness, conscientiousness, extraversion, agreeableness, and neuroticism. In 1981, Lewis Goldberg gave the set its name, the “Big Five.”

It doesn’t sort you into a type. It gives you a score on each trait, so you’re a little high here and about average there. One study of 559 people, reported in Scientific American, found it was roughly twice as accurate as a Myers-Briggs-style test at predicting life outcomes.

It isn’t perfect, though. The five-trait structure replicates unevenly across cultures, and openness in particular doesn’t hold up well in some Asian samples. That’s worth remembering if you grew up between cultures.

CliftonStrengths comes from Gallup, built on the work of psychologist Don Clifton. It launched in 1999, when about 18,000 people took it. Gallup now reports over 38 million completions. It measures talents, not personality: out of 34 themes, you get your top five, with names like Learner or Responsibility. It’s a practical tool for work conversations. It tells you what you naturally do well, not what’s true about your whole personality.

The MBTI (Myers-Briggs Type Indicator) is the most famous. Katharine Briggs developed her own typology during World War I, then adopted Carl Jung’s theory of psychological types after his book reached English readers in 1923. Her daughter, Isabel Myers, built the first assessment in the 1940s. Neither woman had formal training in psychology. The test sorts you into one of 16 types across four pairs, such as introvert or extravert. Tens of millions of people have taken it, and a lot of large companies have used it for things like team building, leadership development, and coaching.

I have friends who use the MBTI actively to understand the people they meet, and it’s really cool to watch. From a four-letter result, they can pick up a surprising amount about someone and explain why that person acts the way they do. They treat it as a starting point for curiosity, not a final verdict, which makes for very interesting conversations.

MBTI’s biggest limit is the sorting itself. Most people sit somewhere in the middle on each pair, and a slightly different day can flip a letter. The company that sells it says it describes preferences and shouldn’t be used to predict job performance or in hiring. I agree with that. I treat it as a conversation starter.

The Enneagram has the murkiest history. A figure called the enneagram came from the teacher G. I. Gurdjieff. Oscar Ichazo built the nine personality types on it, and the psychiatrist Claudio Naranjo brought them to the United States around 1970. Each type has a core fear and a core desire, which makes it feel deeply personal. That’s also why it’s hard to test. Many psychologists consider it unproven, and in a 2006 poll of 101 doctoral-level psychologists, it was among the tools at least a quarter of them called discredited. I use it as a reflection prompt, not as a fact about me.

What the common results look like

Because the MBTI has been given to so many people, we know roughly how types are distributed in the United States. ISFJ is the most common, at around 14 percent. INFJ is the rarest, at around 1.5 percent. Those figures come from the test publisher’s data and summaries of it, so I’d read them as approximate.

One caution. A rare result feels flattering, and that’s exactly when to be most skeptical. Being uncommon on a test doesn’t make you better, just different from most of the sample.

How I use AI to read my results

A single result is a snapshot. The useful part starts when you compare several, and that’s something AI is good at, if you set it up right.

The risk is that AI tends to agree with you. If you paste in your results and ask, “What does this say about me?”, you’ll often get a warm, flattering summary. That feels good, but teaches you very little. AI amplifies what you bring to it, including a more flattering picture of yourself than the facts support.

So I ask it to push back. Here’s the prompt I’d start with:

Here are my results from [list the tests and your results]. Please don’t flatter me. First, tell me where these results agree with each other. Second, tell me where they seem to conflict, and what could explain the conflict. Third, tell me what each test is likely to miss about me. Fourth, describe how a coworker or family member might experience these traits, including the downsides. Finally, ask me three questions that would help you test your own conclusions.

Then answer those three questions honestly, and keep going. You can ask it to see the world through a different result, such as “How would someone with the opposite profile read my last email?” Seeing your own habits from the other side is where most of the learning happens for me.

A few ground rules. Don’t paste anything you wouldn’t want stored somewhere. Treat the output as a hypothesis to test against your own life, not a verdict. And don’t use any of this to label other people. A test result is a starting question to ask someone, not an answer about them.

What to take from this

You don’t need to believe in any of these tests to get value from them. Use the more rigorous ones as an anchor, the others as prompts, and AI as the person across the table who is willing to say, “Wait, those two don’t match.”

In Part 2, I’ll do this with my own results. Four tests disagreed with each other at first, and sorting out why taught me more than any one result did. I’ll also cover what it changes about how I work, and how I show up with friends and family. You can see my results now on my Personality page.

If you’ve taken any of these, try the prompt above and let me know what surprised you.

Frequently asked questions