I'm Angela Duckworth.

I'm Mike Mann.

And you're listening to No Stupid Questions.

Today on the show, how accurately do we see ourselves?

When you ask people, are you an average driver?

Oh my gosh, everyone's a great driver.

Mike, we have an email from a listener named Haji, and I'm going to read it to you.

Okay.

It begins, I found the ongoing discussions about the Big Five personality series fascinating.

Oh, nice.

I mean, we went through each of the Big Five personality traits in a series recently.

And apparently Haji was one of the, I think we had 50,000 listeners take the survey.

So Haji says, after taking the survey, I couldn't help but wonder how others might rate me if asked and vice versa.

What factors lead to the variations between an individual's self-perception and how they're perceived by their family, colleagues, or friends?

Terrific question and something I have been thinking about, I think, since my very first day of graduate school.

I love this because I think we lie to ourselves all the time, probably not on purpose.

But I often wonder, like the way I perceive myself, I'm sure is massively different than how other people perceive me.

Be specific.

I just think it's probably true that family and close friends are both more accurate in their perception of us, but maybe also more judgy.

Not because they're judgmental, but because they have seen a broader perspective of our personality.

So they'd be more critical, you think?

Yeah, but critical in an observant way, not necessarily critical as in a negative way, right?

I think it's really hard to be honest with ourselves all the time.

Okay, I've been collecting data on self-report questionnaires like No Stupid Questions listeners took.

And by the way, by popular demand, we just kept the survey up.

So if anybody wants to see what they think of their own personality traits, they can do that at Freakonomics.com slash big five.

But I've been collecting data that is like that survey, self-report.

Right.

Answer a bunch of questions about yourself.

And then you get friends or teachers or parents to answer the same exact questions.

But now the pronouns are not first person singular, they're third person.

The technical term for this in psychology is called informant reports.

You know, is Mike Maughan talkative?

Does Mike Maughan like to go to parties?

Did you know my mom once observed me talking to someone?

This is not a great moment in my life, but I walked back over and she said, who put a quarter in you?

Because I was incredibly, maybe overly talkative.

Were you a jukebox?

Yeah.

Maybe not as solicitous of asking questions of the other person as I ought to have been, as I was taught.

Yeah, I love that expression.

Had not heard it.

But anyway, the question is really rich and deep.

And I mentioned that I've been thinking about this question since my first days of graduate school.

That is because in my very first study as a graduate student, I gave questionnaires to middle school students.

Actually, eighth grade is about the first grade where somebody can kind of reliably answer these questionnaires.

Before that, kids tend to be very egocentric.

So when you ask them, like, are you talkative?

They don't do what they're supposed to do, which is to compare themselves to other kids.

They answer the questions, but there's a lot of noise and not a lot of signal.

But I gave these self-report questionnaires about personality, in particular self-control, to these middle schoolers.

And then I also gave the same exact questionnaires in their informant report version.

So now, you know, does this child control themselves when they need to, etc.

To the teachers of these kids and then also their parents.

And Mike Mond would not be surprised that when you get all the data back, one way you know how much signal there is and how much noise is whether the scores for that particular way of measuring self-control correlate with the others.

And in fact, parent ratings of self-control are the least correlated.

Really?

Yeah.

I mean, it's not that there's no correlation, but they don't predict outcomes as well as...

Wait, wait, wait.

Let me clarify.

So parents are the least able to accurately determine?

That would be my reading of the data.

Okay.

I just want to make sure I was understanding.

It's like you would expect them to know the kid best, right?

They've changed their diapers.

They've seen them for much longer than their teachers.

I mean, who doesn't know you but your mother?

Yeah.

Is it that we all think our kids are the best?

I can't remember whether the averages were higher because that's one way you would know.

Like, oh, the parents are rating these kids like four out of five and the teachers are rating them like two out of five or something like that.

But they just didn't relate to other things very well.

And I found that over and over again.

And one parent mailed back the survey.

And instead of checking off the boxes, like very much like my child, you know, not at all like my child, they wrote in the margin, how am I supposed to answer this question?

I only have one kid.

I have no idea how they compare to others.

Oh, interesting.

And I do think that one of the reasons why parents struggle is because, you know, they want to think their son is the next Tom Brady or something.

But they also just have a very limited frame of reference compared to teachers, right, who see, I don't know, dozens of kids a year, sometimes hundreds, and then hundreds of kids, if not thousands over the course of their career.

Well, and it's hard to judge yourself often going to Haji's question, because, I mean, it's the first time I've ever been a fifth grader or the first time I've ever been new at this company.

I don't even know how to rate myself against everybody else.

It's sort of like when they ask you at the hospital, rate your pain on a scale of one to ten.

I don't want to say that I'm a ten.

I don't know what it's like to give birth.

It's so hard to compare.

Right?

Yeah, it is.

It used to be like, how's your pain?

And then words would come out of people's mouths, but you're like, what?

It's better to say like, okay, scale from one to ten.

But what is a six?

So now the better scales that your doctor can ask you are scales where the number comes with a description, cannot walk up a flight of stairs, you know, five, cannot pick up a half gallon of milk, seven, cannot get out of bed.

By the way, they still aren't perfect.

This is called the vignette approach.

But to this day, there's always a little bit of signal.

And there's also this kind of noise that comes from a stoic person, Jason's grandfather, who was affectionately known as Pothead.

I don't even know what his real name was.

Wait, Pothead?

I can't believe.

Maybe his name was Ted and it was like Paw Ted.

Oh, Paw Ted.

It sounded a lot like a marijuana user, Pothead.

No, yes.

That's what I thought you said.

He drank vodka, but no marijuana so far as we know.

And Pothead is just like this extremely high pain threshold.

There was these stories of his like breaking a bone and still doing that.

Like, so even with the little vignettes that are supposed to help you with these reference points, like, I think one of the main lessons of measuring personality or anything else is that my six and your six are not the same six.

Even when we add words and descriptions, I think you put your finger on one of the many reasons why when somebody else rates us and we rate ourselves, those ratings aren't always going to line up.

They may have different standards than we do.

You've mentioned that parents have a lack of ability maybe to see as objectively.

Are there people or groups to whom we can turn that help us gain a better sense of who we are?

Like in my world, we often would do these 360 degree feedback surveys.

And that's where your manager, your direct reports and your peers all give feedback on a set list of questions.

And that's meant to give a well-rounded view of you as a person and maybe point out blind spots or areas that are going well.

I learned this lesson my very first year of graduate school when I was collecting that data from parents and teachers and then the students themselves.

In general, 360 is good because even though the parents have their biases and they struggle with knowing what an average fifth grader is like, and even though the teachers may also have biases, like rules of thumb about what girls are like or what boys are like.

And then also there's this problem called halo bias, where if a person has one positive characteristic, like they have a good sense of humor, you tend to attribute other things like they must be smart, they're probably a good athlete.

I bet they have a lot of friends, right?

Right, for sure.

But here's the thing about 360 assessment, and here's why it's a good thing.

It's called the principle of aggregation, and it's so non-intuitive and it's so powerful and it's so important.

When you take all these imperfect signals, so everybody has a little signal, but they have a lot of noise.

When you just average them together, what happens is not what you would think.

Instead of getting more noise, you actually get a lot of signal because the noise kind of like cancels out.

And so 360 assessment is essentially what I did in my first year of graduate school.

I just averaged together all the ratings, and even though each one was imperfect, their collective signal was a lot stronger than any one element alone.

So I think employers should keep doing 360s.

I think that's also something we should do when we hire people, right?

Like, don't call one reference.

Right, oh my gosh.

Such a rookie error.

I make it all the time, though.

Well, also, don't just call the references that they give you.

You know what question that Luis went on, who was the last chair of the board that my nonprofit Character Lab had?

The question that he likes to ask when he's doing reference checks, would you hire that person again?

That's a great question because at the end of the day, it says so much.

Do you think that your close friends, if they filled out the Big Five personality inventory, let's say you averaged together like your eight friends, do you think that if the scores came out differently for the eight friends who know you best, would you think that that is better evidence of who you really are than your own ratings?

Like, they came out differently than my scores, not differently than each other.

Yeah, and by the way, when you take these ratings, they very often, like, don't line up perfectly.

Right.

But let's take the average of them.

Right.

Let's use the principle of aggregation.

And let's get the Mike Mon friend score and say they thought you were like really high in openness to experience and you thought you were lower than they thought.

Who would you believe?

I think that I would want to believe them because I would think that maybe I'm engaged in I don't know what it's even called, but a self-perception bias.

Is that a thing?

Sometimes it's called self-serving bias.

Well, because I think to some extent we answer these things aspirationally or we all have selective memory or we have a willingness to kind of glorify the past, forget the things that were hard or bad.

And I wonder if in our own taking of these questionnaires, we tend to focus on the most proximate experiences that are coloring things or, you know, I'm thinking of my best self when maybe they're thinking of my whole self.

Like everyone who takes psych one, you know, intro psych is taught that we can be biased to say we're better than average.

Right.

And we often teach these introductory psychology students that when you ask people like, are you an average driver?

Oh, my gosh.

Are you better than average?

Worse than average?

Everyone's a great driver.

Except for me, by the way, I would raise my hand and be like, I'm a terrible, terrible driver.

You don't drive.

Yes.

Then again, I'm not really a driver.

Okay, fair enough.

But everybody can't be better than average, so you must be deluding yourself.

Right.

Well, a few years ago, there was this meta analysis.

So these scientists took all of the data that had been gathered that they could get on like self-report personality scores, like the ones that we've been collecting on No Stupid Questions from our listeners, and then informant report about that person.

And if there really is this strong self-serving bias, the average scores for positive traits like conscientiousness or agreeableness, they should be higher for self-report than for informant report.

Uh-huh.

So I read this, and I was really surprised.

What they found from, you know, over 30,000 data points across 150 samples is that there is really not much difference between the average scores of people's self-report, big five personality, and informant report.

So there wasn't strong evidence.

There was one exception, which we can talk about, but in general, they're like, nah, maybe people are not as prone to self-serving bias as we teach in Psych 1.

Interesting.

Is that true maybe just of rating your personality versus other factors?

Because I would imagine there are other areas that we are pretty self-deluded.

I don't know.

This study was only on personality, but it's a good question.

You know, the big five, as you recall, it's extroversion, openness to experience, agreeableness, conscientiousness, and then neuroticism.