Here at SciShow, we like  to get to know our viewers.

And we know that the people who  find our videos are pretty special. So we decided to do a little market research to get to know you all better.

We found that SciShow fans are extremely likely to identify with the following statements: You have an analytical mind, though  you also might space out at times You pride yourself as an independent thinker, and don’t accept other people’s  statements without good proof You love variety and tend to rebel against too many restrictions and limitations You don’t always reveal all of yourself to others, and finally, You have a great desire for other  people to like and admire you That sounds a lot like you, doesn’t it? Except, we’re just kidding. We didn’t actually do any research, and we made up all of those statements.

But we’re willing to bet that some of you really did feel a connection to them. Why is that? Personality tests and profiles  take advantage of a weird psychological tendency that  also benefits everything from horoscopes to fortune  tellers to Buzzfeed quizzes.

Here’s how it works, why we fall for it, and what to look for so you don’t get duped. [♪INTRO] Before we can get into why these  profiles are so easy to believe, let’s step back and get into the  history of personality tests. The first modern standardized  personality test was the Woodworth Personal Data Sheet, also know as WPDS, which came out in 1917. It was designed to predict  which soldiers were vulnerable to shell shock, or what we now call PTSD.

However, right when it was set  to launch, World War 1 ended, so this particular test never got passed around at scale among soldiers. The first version of this test consisted of around 125 yes or no questions. These included things like,  “Have you ever fainted away?”, “Are you bothered by  fluttering of the heart?”, and “Do you think you have hurt yourself  by going too much with women?” All very relatable questions, if you ask me.

And while this version of the  test never got widely used, Woodworth’s later work on personality  testing started a craze of people wanting a test to  tell them about themselves, and the WPI WPDS helped set the standards for personality testing going forward. Academic institutions,  employers, and psychologists soon began devising and using tools  to assess people’s tendencies. These early tools have roots in capitalistic, industrial interests, and were obsessed with the concept of adjustment in  early personality test design.

In this context, adjustment  refers to how adaptable people are to the norms of the era, including the rules and decorum expected of a booming workforce. Basically, they wanted to find  tests that would weed out the people who couldn’t get in line, or worse,  wanted to get on the picket line. Because we all know, a unionized  workforce is bad for business!

From there, these tests found broader popularity, including glowing coverage of  a widely used test called the Humm-Wadsworth Scale, in a  1942 Reader’s Digest article. The story, titled “Fitting the Worker to the Job,” explained to readers that these tests were used by companies like Lockheed Aircraft as a so-called scientific way to  fit people to their jobs. There were questions like “Do you  like to pass along a good story?”, “Do most of the people you meet interest you?”, and “Do you like to see the villain punished?” The results supposedly  revealed personality traits, like a tendency to be overly emotional or whether you’re more of a lone wolf than a team player.

But, there’s often a difference  between something seeming sciencey and being based on, like,  actual high-quality research. And in 1949, a researcher  and college professor named Bertram Forer~ expressed his doubts about all this personality testing hullabaloo. He even said that most of the  time, personality profiles weren’t any more specific than saying that  a person has opposable thumbs.

Meaning that the profile might  say things that are true about whoever’s reading it, but they’re  also true for like, everybody else. Which is when Forer decided  to experiment on his students. He passed out a personality questionnaire in his intro to psychology  class, and told them all that he’d be analyzing the results and creating unique personality profiles for each of them.

Everyone received their  assessments in the classroom, seated far enough apart that they  couldn’t read others’ papers. The students then reviewed and  rated their personality summary for accuracy, and even scored each  bullet point in the list of thirteen. Only one student rated their  results below a 4 out of 5, so basically the entire class felt  that their scores were pretty accurate.

But you may have anticipated the twist here. Those assessments were not in  fact tailor-made for each student, and instead, they all got the exact same list. Actually, you did too, because all  those statements from the start of this video were basically  paraphrased versions of what Forer told his students.

Sorry, by the way. Forer made a list of general, vaguely flattering, and universally relatable statements. So why did everyone believe that their list was so perfectly tailored to them?

Well, that’s the Barnum effect. The Barnum effect was named after PT Barnum, the iconic and problematic showman  known for his ability to captivate, and often manipulate, an audience. The Barnum effect is the phenomenon where if you give someone a personality test, they’re pretty likely to believe that  the results are true and accurate, regardless of how hard the  profile-maker actually tried.

There’s something about  taking the test itself that makes an audience more likely  to believe the end result. All of these different generic  personality assessments are called Barnum profiles, and  they’re all show, no substance. Other experiments have been done after Forer dunked on his intro class, with similar results.

Humans just have a tendency  to fall for Barnum profiles. It’s still not entirely clear why this happens. Some researchers think that the  broad and highly generalizable statements do apply to all or  most people, just like Forer said, and the reason so many people  put stock into them is that they’re written to be pretty one-size fits all, but that’s not obvious to the test-taker.

Multiple studies have also  shown that some individuals will even believe that a Barnum  profile applies more to them than to the general population,  which means that even though the statements are really vague and non-specific, the recipient doesn’t see it that way. It could also be that people want  to believe favorable statements about themselves, which is why  even the more negative parts of a profile usually come with a compliment  or positive spin at the end. Like this morning, CoStar told  me your temper gets the better of you sometimes, but deep  down you have a heart of gold.

There could also be confirmation bias at play. This is when someone affirms  statements or evidence that support their pre-conceived notions, while  overlooking what doesn’t fit. It’s pretty likely that some or  even all of these reasons are part of what makes a Barnum profile so believable.

Once you know what to look for,  you can see how our tendency towards buying into the Barnum  effect has been around in many forms for ages, in everything from  horoscopes and fortune telling to how ChatGPT and other Large  Language Models tap into the Barnum effect to make its vague  statements seem believable. And the likelihood of somebody  believing that their Barnum profile is a perfect fit for them might  be related to a few key traits. For instance, studies have looked  at how self-image might influence whether people believe these  vague personality descriptors.

The idea is that if you  have a weaker sense of self, you tend to look to other  people to tell you who you are, and one of those other people  might be your local astrologist, or a litany of online personality quizzes. People with an external locus of control, which means they don’t feel  they have a lot of control over their own lives, also  tend to accept Barnum profiles. Moreover, if you give someone a personality test and they like what they got,  they’ll end up believing that the entire test itself  is a really accurate test.

Like, let’s say I’ve invented  “Madelyn’s Ultra-Awesome Personality Matrix” and gave you that test, and your results say you’re  smart, and funny, and nice, and really pretty, and you have great hair. Then when you go tell your friends about my test, you’ll be like “that personality  test was amazing and super accurate, and totally worth it, you should give it a try! It only cost me eight bucks!” Hang on, this is giving me  an idea for a side hustle… Some research also suggests  that people tend to accept their Barnum profiles if the assessment  tool uses lots of data, or at least, if the recipient thinks they did.

For instance, you’re more  likely to believe your horoscope if it’s based on the specific  date and time of your birth, rather than just the month and year. Or, you’d be more likely to love  your Ultra-Awesome Personality type if the test is fifty questions  long instead of just ten questions, even if the actual results  are the same either way. Which is good news, because my new and improved questionnaire is eight hundred  questions long.

Bring snacks. There’s still a lot to tease out  around the various types of Barnum profiles and the factors contributing  to people’s acceptance of them. Despite Forer’s early work exposing this effect, personality tests are still super popular today, and even claim to be based on lots  of their own data and research.

Have we cracked the code on making accurate personality tests in this modern era? It’s… complicated. One of the most common  personality tests out there is the Myers-Briggs Type  Indicator assessment, or MBTI, which has been around for over half a century.

It’s a proprietary questionnaire that  sorts people into personality types based on four binary categories:  introversion vs. extroversion, sensing vs. intuition, thinking vs.  feeling, and judging vs. perceiving. The assessment groups people into  one of the two tendencies for each category, so someone can be INTJ, or ENFP, which means there are sixteen possible  personality profiles in total. It’s gone through many iterations over the years, and the current version available  online is 93 questions long.

Remember what I said before  about the illusion of more data giving you more accurate  results? I know you remember. You have a great memory — of course you do.

They even offer training  courses so you too can become certified to administer the  test and interpret the results. Their big thing is using MBTI in the workplace to understand the dynamics on a team, so you can pair the right personalities  together for the best outcomes. So you’d think that if this  company has poured so much time and effort into refining  their test, it’s gotta be good.

Unfortunately, that’s not really what the evidence from non-MBTI-affiliated studies says. See, when researchers want to  see how well a certain assessment tool, test, or survey actually  works, one thing they’ll do is have the same people take the same test multiple times. If they get the same score each time, we’d say that tool has good  test-retest reliability.

And in studies of MBTI where participants took the assessment multiple times, up  to half or even more test takers received a different result for  at least one of the four letters. Not a problem with the Madelyn’s  Ultra-Awesome Personality test, I assure you. It’s not clear what percentage  of people who take the MBTI assessment accept the  accuracy of their results.

But the internet is rife with  people singing its praises, both about how well their profile fits them, and how well MBTI scores fit their friends. Whether or not the data supports  it, MBTI is an extremely popular and well-liked system to try to  categorize people’s personalities. Which brings us to the question of whether or not it’s even possible to use a survey or questionnaire to accurately  evaluate someone’s personality.

Like, is this whole thing a losing game? Well, researchers don’t all think  so, and there have been quite a few attempts to make a statistically  validated personality testing tool. One example is the Big Five  Personality Trait model, which gives people a low, medium,  or high rating in five dimensions: extroversion, neuroticism, openness, conscientiousness, and agreeableness.

These tests, and newer variations  that include subcategories of these five, do seem to show  better test-retest reliability. One major reason that newer  tests based on the Big Five are more reliable is that they’re  based on accumulating data from multiple long-term  studies from the 1990s onwards. And they’re rooted in the principle  that, if a personality trait exists in humans, languages  will adopt words to describe it.

The more cultures and languages  there are that talk about, say, extroversion, the more likely it is that that’s a valid metric on which to assess people. So these Big Five studies  looked at the actual language that participants and cultures  all over the world use to describe personality, in multiple languages. However, there’s a problem.

Pretty  much all the data we have on the BIg Five comes from what  researchers call “WEIRD countries”. That stands for Western, Educated, Industrialized, Rich, and Democratic. Researchers have come up with  this framework to basically remind themselves that people in these  highly educated, industrialized societies are so different from the  vast majority of human history that they’re not a great proxy for  understanding everyone, everywhere.

Which means that if you  only study WEIRD countries, you can’t assume that your results  are universal for everybody. And yeah, it seems that the evidence says that the Big Five doesn’t really work  for non-WEIRD populations. So there’s definitely a lot of work  left to do if we want to figure out how to understand personalities  for everyone, WEIRD or not.

Plus, there’s a concern that  even the best test out there might be flawed just by the  nature of what a test is. When you have to answer  every question from a list of predetermined options, it’s  called a forced-choice measure. These tests are easy to administer and to grade, but the downside is that they’re  really rigid and can flatten nuance, including how people’s  personality traits can change due to the passage of time and other variables.

We’ve all stared at a multiple-choice question and wished there was an option to check “other.” So trying to make a questionnaire-style  test that can accurately gauge anybody’s personality  might be kind of impossible. In any case, if you’ve ever found  yourself putting a lot of stock in your horoscope or MBTI or  any other personality test, it’s not a unique intellectual  failing on your part. It’s just human nature.

Personalities are so much more varied and dynamic than we’re led to believe, and  it’s hard to reduce them to types. After all, I’m far too complex to  be summarized by one single test. My Ultra-Awesome Personality  Matrix report told me so! [♪OUTRO]