I Will Never Have The Ability To Clearly Explain My Beliefs About Growth Mindset


A lot of the comments I’ve gotten about Tuesday’s post on growth mindset have been pretty similar. They’ve argued that yes, innate ability might matter, but that even the most innate abilityed person needs effort to fulfill her potential. If someone were to believe that success were 100% due to fixed innate ability and had nothing to do with practice, then they wouldn’t bother practicing, and they would fall behind. Even if their innate ability kept them from falling behind morons, at the very least they would fall behind their equally innate abilityed peers who did practice.

I will call this the Bloody Obvious Position, since it’s hard to believe it isn’t true. I once tried to imagine a world without it as a thought experiment, but it was pretty weird and I wasn’t serious.

The Bloody Obvious Position was what I was trying to get at with my post on basketball, and with my terrible ad hoc graph:

Nevertheless, some people thought I was denying the Bloody Obvious Position. Other people thought I was accusing Carol Dweck of denying the Bloody Obvious Position (see eg here). This despite my making sure to say:

I want to end by correcting a very important mistake about growth mindset that Dweck mostly avoids but which her partisans constantly commit egregiously.

I believe the Bloody Obvious Position. Dweck believes the Bloody Obvious Position. I acknowledge that Dweck believes the Bloody Obvious Position. There are a lot of growth mindset partisans online who don’t believe the Bloody Obvious Position, and I satisfied my urge to yell at them, but now they’ve been yelled at, and the more important issues debated by reasonable people still remain.

So where do I disagree with Dweck? I interpret Dweck as making the following statement:

The more important you believe innate ability to be compared to effort, the more likely you are to stop trying, to avoid challenges, to lie and cheat, to hate learning, and to be obsessed with how you appear before others

Call it the Controversial Position. This is not the same thing as the Bloody Obvious Position. In the Bloody Obvious Position, someone can believe success is 90% innate ability and 10% effort. They might also be an Olympian who realizes that at her level, pretty much everyone is at a innate ability ceiling, and a 10% difference is the difference between a gold medal and a last-place finish. So she practices very hard and does just as well as anyone else.

According to the Controversial Position, this athlete will still do worse than someone who believes success is 80% ability and 20% effort, who will in turn do worse than someone who believes success is 70% ability and 30% effort, all the way down to the person who believes success is 0% ability and 100% effort, who will do best of all and take the gold medal.

And this is why I deny that I’m secretly agreeing with Dweck, or strawmanning Dweck, or whatever. I don’t believe the Controversial Position, but I think Dweck does. For example, here she writes: “The more a player believed athletic ability was a result of effort and practice rather than just natural ability, the better that player performed”.

There is nothing in there about “the more a player realizes that, no matter how important innate ability is, effort matters too.” Her statement says that it’s entirely about what degree a player attributes success to effort versus innate ability. The natural conclusion there is that the player who believes success is 0% innate ability and 100% effort will do the best.

Her studies reflect this as well. The most common design uses the IAR, a test where children are asked to attribute different things to effort versus ability. Those who attribute too many things to ability are classified as “helpless” and “fixed mindset”. There’s no question about “Okay, some things are due to ability, but if you work hard that still helps, right?” Nor have I ever seen any of the literature claim “it’s important to believe effort matters a little, but after a certain point more effort-attribution doesn’t help”, or “Maybe there’s an L-shaped relationship between belief-in-importance-of-ability and success.”

I’d like to be able to teach my children that success is X% innate ability and Y% practice, for non-zero values of both X and Y. I think growth mindset theory claims that if some other parent teaches their kids the same thing for a lower value of X and higher value of Y, their children will be more honest, harder-working, and more successful. And the parent who says it’s 0% innate ability and 100% practice will do best of all. If growth mindset people don’t believe that, I can only confess I have never been able to infer that lack of belief from their writings.


Worse, we can distinguish between a Sorta Controversial Position and a Very Controversial Position:

SCP: The more children believe effort matters, and the less they believe innate ability matters, the more successful they will be. This is because every iota of belief they have in effort gives them more incentive to practice. A child who believes innate ability and effort both explain part of the story might think “Well, if I practice I’ll become a little better, but I’ll never be as good as Mozart. So I’ll practice a little but not get my hopes up.” A child who believes only effort matters, and innate ability doesn’t matter at all, might think “If I practice enough, I can become exactly as good as Mozart.” Then she will practice a truly ridiculous amount to try to achieve fame and fortune. This is why growth mindset works.

VCP: Belief in the importance of ability directly saps a child’s good qualities in some complicated psychological way. It is worse than merely believing that success is based on luck, or success is based on skin color, or that success is based on whatever other thing that isn’t effort. It shifts children into a mode where they must protect their claim to genius at all costs, whether that requires lying, cheating, self-sabotaging, or just avoiding intellectual effort entirely. When a fixed mindset child doesn’t practice as much, it’s not because they’ve made a rational calculation about the utility of practice towards achieving success, it’s because they’ve partly or entirely abandoned success as a goal in favor of the goal of trying to convince other people that they’re Smart.

Carol Dweck unambiguously believes the Very Controversial Position. In a quotation which I admit I am mangling and ellipsis-ing heavily to remove extra verbiage, but which I think preserves the meaning of her claim:

[People with fixed mindsets] are so concerned with being and looking talented that they never realize their full potential. In a fixed mindset, the cardinal rule is to look talented at all costs. The second rule is don’t work too hard or practive too much…having to work casts doubt on your ability. The third rule is, when faced with setbacks, run away. They say things like ‘I would try to cheat on the next test’. They make excuses, they blame others, they make themselves feel better by looking down on those who have done worse.”

Can we all agree this is a much stronger claim than “ability matters, but effort also matters?”


I was not intending to “debunk” growth mindset, or even present a pure polemic against growth mindset. I admitted that many of the studies around it were very good, and that I don’t have good answers to them. My bias is against the theory, but I tried not to just follow my bias. I tried to treat it on the level of “there’s a lot of good evidence for growth mindset, now what’s the best evidence we can find against it?”

So I guess I should probably come out and say what I believe about each position.

I believe the Bloody Obvious Position is bloody obvious.

I believe the Somewhat Controversial Position is probably not a good way to parse things. Part of it is that we might be confusing explicit versus implicit beliefs. Maybe a particular geneticist is very aware of research showing how important genetics is to success, and would give a very high estimate if asked, but in her own life, when she fails, lack of effort is still the first explanation to immediately leap to mind. Or maybe some teacher is very on board with growth mindset and things IQ is a racist construct, but is convinced that she can’t do physics because she’s just “not a math kind of person”. The research I’ve seen hasn’t really distinguished between explicit and implicit beliefs. The priming experiments sure seem more likely to affect what immediately comes to mind than your stable, well-reasoned beliefs about how the world works (even though a few priming experiments have checked stable well-reasoned beliefs to see if the intervention worked!)

If you put a gun to my head, I’ll say it certainly works in the lab, and give you about 50-50 odds that it matters in real life. The studies don’t show any real-life correlation between growth mindset and any measures of success. Many people have pointed out that this could be confounded – dumb people might preferentially believe ability doesn’t matter to make themselves feel better about not having it, and smart people might preferentially believe effort doesn’t matter because they rarely have to use it. But if you accept that, some of the rest of it starts to look confounded. If fixed mindset = smart people, than might the reason they react poorly to challenges and failure be that they have no experience with them? Might it be that the more challenges and failures you’ve encountered before, the better you are at dealing with them? Certainly that is how I interpret this Dweck paper, even though she thinks it is purely a mindset effect. Another way of explaining the ecological results without bringing in that particular confounder would be if growth mindset helped in some situations, but fixed mindset helped in others. For example, a person with fixed mindset risks not trying hard enough because they think there’s no point. But a person with growth mindset risks the opportunity costs of prolonging their inevitable failure instead of (in the Silicon Valley term) “failing fast” and pivoting towards a higher-payoff activity.

The Very Controversial Position is also well-supported, also contradicted by ecological data, and really really doesn’t match my experience. The people I know who are most interested in issues of innate ability don’t behave at all like Dweck’s subjects. In fact, I wonder if a lot of the “life-hacking” movement might be ability-mindset people trying to figure out how to succeed more by improving ability – certainly the people who practice dual-n-back every day because they think it increases IQ fall into this category, but so do nootropics users, people who follow special diets to increase energy, and “try this one weird trick to improve your motivation”. And these same people seem interested in things like spaced repetition software, which might be thought of as sort of prosthetic ability-enhancers. On the other hand, I’ve also met people who say “I could succeed if only I put in some effort, but I have some mental block / depression / ADHD / low conscientiousness score that makes it impossible for me to work that hard, so better go eat worms”, then sabotage themselves at every opportunity.

And yes, it’s a sin to privilege your own experience and priors over the results of good studies, but sometimes it’s necessary. And it’s another sin to prefer the results of broad ecological studies to controlled experimental trials, but sometimes that’s necessary too.

I deny the claim that I don’t disagree with Dweck on anything of substance. I don’t absolutely disagree with her on anything, but there are a lot of things I doubt, or that I expect to capture true insights without being the best way to express them. Growth mindset makes some surprising and genuinely controversial claims, and I’m not yet at the point where I can feel sure about them either way.

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

277 Responses to I Will Never Have The Ability To Clearly Explain My Beliefs About Growth Mindset

  1. Anthony says:

    Side note:
    I’d like to be able to teach my children that success is X% innate ability and Y% practice, for non-zero values of both X and Y.

    I don’t think that X+Y=1, or some other fixed number.

    Your “terrible ad hoc graph” explains it much better – your innate ability tells you what the bounds of the graph are, and your effort determines where on the graph you end up.

    It’s actually more complicated than that. For single tasks, my explanation above is reasonably correct. For “life in general”, you (any one person) have a whole bunch of different ability graphs. The ranges are often correlated, especially for similar types of tasks. The range boundaries are fuzzy, though the bottom end tends to stretch towards zero, and the relationship between effort and position on the graph is not linear. And there’s more an element of luck over larger domains and longer time spans.

    Also, Those ability graphs have corresponding “success graphs” which are wildly different – being the best juggler in the world will probably allow you to make a good living for ten or twenty years, enough to remain comfortable for the rest of your life even if you coast after you’re no longer getting paid. But being in the top 1% of jugglers in the world means you’ll be a little more popular at parties, and nothing more. Being in the top 1% of computer programming skill can lead to a pretty good life. Being in the top 10% of sales and marketing skill is probably about as good, if you’re selling the right things.

    There’s also different measures of success. Assuming what RoissyInDC said about himself is true, he’s spectacularly successful at getting pretty women to have sex with him, but he’s probably below median (for DC) financially. If he were willing, he could probably put his talents into a more remunerative use, and become pretty successful financially. (But it might interfere with the flow of women through his bedroom.) A top level politician can have a very significant influence over what people do – tremendous amounts of power, while not becoming terribly rich. Ralph Nader has had a very big influence on the U.S., but he’s not rich, and he might never have been able to crack the top 1% even if he’d tried.

    • Lemminkainen says:

      A minor nitpick, but Ralph Nader owns about $3 million dollars in stocks and mutual fund shares, which fairly comfortably puts him into the top 1% by net worth.

      • g says:

        Not according to this article which claims the threshold is over $8M. Though I’ve seen another that cites other data to claim it’s more like $1.5M.

        • Who wouldn't want to be Anonymous says:

          A 1% threshold of $8M is much closer to my Fermi estimate. In case nobody noticed, there are a lot of closely held small/medium size businesses.

          Besides, $1.5M is chump change; if you’re consistently middle/upper-middle class and you can’t accumulate that in your career you weren’t trying hard enough. [See what I did there?]

          The rule of thumb I have always heard is that “rich” is starts at $5M.

          • Deiseach says:

            The rule of thumb I’ve always heard is that “rich” is The Other Guy.

            I’m doing okay, you’re comfortably off, he’s a bloated plutocrat 🙂

          • Who wouldn't want to be Anonymous says:

            In my mind, the difference between “doing okay/well” and “being comfortable/rich” is the difference between being a well payed wage slave, and having assets. This is why talking about income disparity can be less informative than talking about net worth. The reason we talk about income almost exclusively, though, is because the numbers are easy to get from the census. Net worth is a lot harder to figure out.

            If you have a well paying job, the second you retire, get fired, irritate the board of directors, injure your knee, or drop a record that doesn’t go platinum… you’re destitute. Sure, making a few hundred thousand, or a few million dollars a year will allow you to lead a banging life style that will be the envy of everyone at your high school class reunion, but the income doesn’t last. Especially at the higher rungs, you’re only making that kind of money for a few years; certainly not long enough to build the asset pool you need to keep leading that kind of life during retirement. At the lower rungs, populated by educated professional, the income is hopefully steadier for longer. Unemployment and social security benefits may help pay the mortgage on a middle class house, but it’ll barely make a dent in an upper class one. The disability, unemployment, retirement risk is much higher the further you go up the income scale. So when a doctor/lawyer/executive/rockstar tells you “I’m doing alright for myself” what they are alluding to is that are able to afford the lifestyle they live right now but that their whole life is perched on a precipice.

            And everyone seems to want the safety net to not catch them out of spite. Whether you do or should feel bad for them is a different issue entirely.

            On the other hand, if you have assets things are pretty good. We’ll talk about cash because that’s easy.

            If you have cash a well diversified portfolio, as a rule of thumb you can spend about 5% of the value of your principle per year and have it stay inflation adjusted indefinitely. You can bump it up to around 7% if you don’t care about your kids inheriting. So Ralph Nader can burn $150k a year, every year until he is pushing up daisies. For reference, that is about the starting pay for a fresh law school graduate at a “Big Law” firm. Ralph wont be able to spend any more in the future, but the lawyer will be making about twice that in a couple years. Big Law is pretty exclusive. Maybe this would be better: according to the Bureau of Labor Statistics (if I am reading right) the 25th and 50th percentiles for psychiatrist pay are around $120k and $170k.

            The doctors and lawyers among the baby boomers would have had no trouble racking up a couple million to retire on (assuming they cared and weren’t terrible at managing their finances, which becoming a doctor or lawyer kind of indicates against). Now, with the insane cost of school, I don’t really expect Scott to get there. Saving early is the most important part of saving long, and paying the first half a million you can afford in debt service… well, it pretty much wrecks the long term picture.

            If Ralph stopped working (is he even working?), he would be petty comfortable. Comfort implying having enough for security but not enough for caviar-as-a-staple-menu-item.

            The $5m figure I quoted would produce an income of around $250k, which is about the breaking in point to being able to really afford big house(s) and fancy car(s), and jetting off to exotic places. Without having to work. Unless you want to so you can buy even more houses and cars. Or horses.

            Owning a business is somewhere in between. You’re working like crazy and in constant danger of ruin–but if you don’t trash your balance sheet too bad on the way down you can recover a decent chunk of change to cushion yourself. And if you are really lucky, you might even be able to take your hand off the tiller for a little while and have some leisure.

            … I’m probably rambling. I do that a lot. I’m not even sure were I was going with this any more.

          • Harald K says:

            is he even working?

            He’s 81, so it would surprise me if he held a regular job.

    • Nathan says:

      The best juggler the world has ever seen quit juggling to lay concrete.

      • Mike says:

        True story. This is probably the article you’re thinking of.

        • Thanks for the link. The world’s greatest juggler is running a construction company, which is not quite the same as him just laying concrete. Also, part of the problem with juggling is that the vast majority of people can’t appreciate the difference between merely hard and extraordinary juggling.

          • Harald K says:

            Circuses focus on hitting the sweet spot of doing things that look extremely impressive compared to how hard they are.

            Not that what they do isn’t hard, but most of what they do at the olympics is probably even harder.

            Getting so good that your audience can’t even appreciate how good you are, would probably be disappointing to a circus artist.

    • Deiseach says:

      Assuming what RoissyInDC said about himself is true, he’s spectacularly successful at getting pretty women to have sex with him

      Can he get them to stick around, though? (I don’t know the person in question or anything about him, so I don’t know what his aims or goals are). If you want a stream of ‘here today, gone tomorrow’ bed companions, I think you can get them if you want them and try the tricks, but getting someone to stick around for any kind of deeper relationship needs different approach.

      Someone could be very successful at having short affairs where they charm the object of their attentions into bed, keep them happy enough to return for a while, then dump them without too much recrimination, yet still feel like a failure because all they can manage to have are shallow affairs and they fail at getting a long-term relationship.

      • anon says:

        That sounds like a failure in the same way that not improving his financial situation is a failure. Or not becoming an influential politician, or not learning juggling.

    • William O. B'Livion says:

      Being in the top 1% of computer programming skill can lead to a pretty good life.

      Been thinking about this off and on since last night. Actually, in a different way a lot longer because we homeschool my daughter right now.

      This gets back to one line in my comment yesterday that (at least IMO) needs more focus:

      One thing that bothers me is that you use the word “success” but never tell us what that is.

      The top *80* percent of “computer programming skill” can lead to a good life. Heck, I’ve got a Bachelor of Fine Arts degree, I was a Marine, and my slightly-more-than-moderate skill with a computer has gotten me jobs on 3 continents (and almost a 4th, and I’ve still got at least 25 years of work life). I make enough that my wife can stay home and raise our daughter. We’ll soon (depending) be buying a house.

      Given that I’ve rarely worked more than 50 hours a week (more than a few jobs were contractually bound at 40 a week, although one was a year of 72 hour weeks), I don’t have a degree or much formal training in my field, I’m not doing bad.

      There are a lot of things I do that I’ll *never* be “successful” at by any objective measure. I do “classic” nordic/crosscountry skiing, and it’s clear that I don’t have a particular “talent” for it, but I can usually do the “intermediate” trails without breaking too much skin and no bones. I shoot IDPA. I’ll never be nationally ranked, but *my* skills are improving.

      So what does it mean to be successful in this arena?

      • Anthony says:

        The top *80* percent of “computer programming skill” can lead to a good life.

        I was comparing over the population of all people, not all programmers.

        • William O. B'Livion says:

          Yesterday I came across a *bicycle mechanic* job that had:
          – wheelbuilding experience a plus
          – Linux and programming experience a plus
          among the bullet points. And no, I’m a lousy wheel builder (and I doubt it pays what I make)

          I have a friend of mine who used to work in IT. He went back and got a degree in Accounting and does accounting stuff. He uses a *lot* of programming when he does his Excel Wizardry. I’m willing to bet that his “better than 80% of the world” programming skills add a significant bump to his value.

          If you’re in India or China having even what we in the industry would consider *minimal* programming skills will get you a HUGE bump in quality of life.

          So yeah, while I wasn’t thinking of “top 80 percent” of all people, I don’t think it changes things much, given that “good” is clearly a relative statement.

          I have a good life compared to a significant number of the people who live within a mile of me–I live in a standalone house with a garage, a yard and an extra bedroom for an office and a stay-at-home spouse. Many of them live in smaller apartments, sometimes two kids to a bedroom, no garage etc.

          But I rent a house in a lower-middle class neighborhood, so compared to a cow-orker who owns a house in a upper middle class neighborhood, and a weekend home up in the mountains, who’s got a better life?

          Then again both of us are living better than about 90% of humanity right now, and that is EXACTLY where I’m going with this.

          To be successful in the *world*, by strictly economic tests, just being born in the Anglosphere/EU qualifies you, no effort or talent needed (Justin Beeber, call your office)

          If you define “successful” as “top 1% in your field” you’re going to need both innate talent and hard work.

          If you define “successful” as “successful in life” then a mix of moderately hard work and *self discipline* is enough *depending on your degree of success*. I’m moderately talented with computers, and I’m a moderately hard worker (he says while f*ing off at work), but I lack the self-discipline in both finances (have trouble saving) and study (to really improve) to be *more* successful. OTOH, I make in the top quarter of incomes in the US while living in not horribly expensive area.

          So am I successful?

      • Shenpen says:

        It’s probably what you know, not how well you know it. You can be very good at typical Javastuff, but will not get you these kinds of excellent jobs. Dynamics-NAV used to have such a boom in Europe about 6-8 years ago that headhunters were basically begging people with 1 years of C/AL programming experience to work for them, now the market seems pretty quiet. Headhunter based jobs are often based on believably lying on your CV about the latest technology. Jobs based on professional connections are more based on actual evidence of problem-solving skills.

        • William O. B'Livion says:

          “…will not get you these kinds of excellent jobs.”

          The window I can look out of from my cube has a view of the Front Range here in Colorado. I can see the storms move across and down into the plains.

          So can the guys working in the building I can see out of a different window. When the rain lashes the glass, when there’s the rattle of ice or the whiteness of snow I turn back to my monitor and type. They often have to quit for the day, or they’re working in 20 degree weather, wet, hot whatever.

          My wife used to work for a startup in Silicon Valley (actually a couple of them) and through hard work twice went from doing Customer Service to doing QA work, and in one place QA lead. VERY minimal programming skills (a touch of SQL really) but focus and time management skills. Not what most of us would consider “highly successful”, but for a woman who grew up living with her mom, brother and grandparents in a 2 bedroom house in a lower working class neighborhood, yeah, that’s success.

  2. Alex says:

    Beyond the positions that you’ve outlined, Growth Mindset ideas, whether very controversial or slightly controversial or whatever else, also get mangled when taken beyond the social sciences and into reform movements in other academic disciplines. If you meet an academic who wants to reduce the emphasis on traditional academic measures (however one defines those), odds are that they can either quote Dweck or quote some mangled telephone game version of Dweck. In some ways I’d like to argue with them, but I realize that arguing with them won’t work, and even if it could work, they’ve read more of Dweck than I have, so they can win the “Who can cite more studies?” game.

  3. NonsignificantName says:

    I’m not sure mindhacking really counts as ability mindset though? By that logic, growth mindset to playing the violin would be to get up on stage without practice and work really really hard at playing the violin. That said, there’s probably an interesting difference between, say, “efficiency” and “effort” attitudes towards growth, where some people keep at the same approach for a long time while even if it doesn’t show results immediately, while others try to find clever shortcuts to self-improvement, but I think that’s probably orthogonal to Dweck’s thing

  4. moridinamael says:

    Ultimate question: Should I stop telling my two-year-old daughter “Good job! You are working so hard at those legos!” and “What a lovely painting! You must be working very hard at painting!” Am I harming her? Am I helping her? Neutral?

    ETA: When I was a kid I was told I was “smart,” and when things stopped being easy, I imploded and exhibited all the worst self-sabotaging behaviors, so I’m always reticent to tell my daughter she’s “smart” or “talented” because … that didn’t help me.

    • Scott Alexander says:

      I intend to continue using effort-praise as per growth mindset, on the grounds that it might help and I see no reason to think it hurts.

      • Anonymous says:

        Actually, there was a recent study done on EXACTLY THIS and the conclusion was that it varied by age: praising the attribute worked better for 5-year-olds, while praising the action/effort worked better for 8-year-olds. I will try to dig up the study.

    • dipitty do says:

      I’d smack myself if I told my kid she was doing a good job with her Legos. Legos are enjoyable all on their own.

      If you are genuinely impressed by how tall her tower is, say so. If not, let her work in peace. Not everything our kids do has to become a metric for how much we approve of them.

      • moridinamael says:

        Thanks for the parenting advice.

        • dipitty do says:

          I cannot tell whether you are being sarcastic or not. She’s your kid. If it makes the two of you happy to tell her she built a great tower, do so. If it doesn’t feel right to you, don’t. Ultimately, it’s none of my business.

          I think parents are under constant pressure to BE MORE DO MORE. Praise, extra-curriculars, all of it. The constant pressure to do everything perfectly.

          I am under constant pressure from [relative] to praise my kids more. I have three, they only raised one, which means they basically have no idea how much work I am already doing when they tell me I need to DO MORE. Sometimes you can’t do more. Sometimes you need to do less.

          So if you feel like you’re doing too much or the wrong kind of praising, it’s okay to scale it back. And if you’re happy with what you’re doing, then keep doing it. She’s your kid. You know what she needs better than we do.

          • “Sometimes you can’t do more. Sometimes you need to do less. ”

            That’s part of the argument of Bryan Caplan’s book on why people underestimate the optimal number of children to have. He thinks modern parents badly overestimate how much time they have to put in in order to be a reasonably good parent.

          • Null Hypothesis says:

            I do like Bryan Caplan. His claims range from reasonably to completely insanely overstated, but he always makes a decent case for at least some fractional merit in his contrarian positions.

            At the very least I always feel a lot more balanced after reading or listening to him – he’s willing to go into detail on so many things that are against the grain, and we need that.

    • Shmi Nux says:

      In general, don’t substitute existing intrinsic motivation with extrinsic. “Nicely done” is probably OK, “you are working hard” when she is simply having fun is probably counterproductive, since she didn’t feel she was working hard, and now you are adding pressure, and suddenly Legos are a chore.

    • Deiseach says:

      No, that’s fine, but if there is one thing she’s particularly good at, or talented at, it’s no harm either to tell her she’s good at it. I think kids have a realisation about things they like or are good at, and they like recognition that yes, this is something they have a knack for. It shows that their parents are paying attention to them as people, and acknowledging their interests, rather than ticking off the positive-reinforcement behaviour boxes.

      But, eh, I know nothing about raising kids. Keep doing what you’re doing; to quote the TV version of Alien Nation: All you can do is love them, teach them right from wrong, and hope they don’t grow up to be axe murderers 🙂

      • moridinamael says:

        Thanks, this is honestly not something that occurred to me.

        I don’t think I’ve screwed anything up yet, because when a kid is two, almost everything requires effort.

    • Richard says:

      I’m essentially with Deiseach on this one. You should probably try to get an accurate picture of how hard she actually is trying, at least as she gets older.

      I have 3 kids and have failed on this particular bit rather badly. Two kids are very smart and one only somewhat above average. Praising the smart kids for effort when they didn’t put any in was extremely counter productive as they stopped trying at all. Not praising the normal kid when he was putting in a lot of effort and failed to meet the standards of his brothers was probably even worse.

      Correcting for those mistakes when they reached their teens turned out to require an insane amount of effort on all parties involved.

      • Deiseach says:

        Praising the smart kids for effort when they didn’t put any in was extremely counter productive as they stopped trying at all.

        It also makes the parent or teacher look stupid; “Wow, this is so good, you must have worked really hard on that!” when the child knows they didn’t work that hard, or as hard as their sibling/classmate, makes the adult look like (a) they don’t know what they’re talking about (b) it’s easy to fool them (c) adults lie to kids (d) they’re not being sincere, so why should the kid believe their judgement when they say it’s good work if they can’t believe their judgement when they say it must have taken a lot of effort?

    • This is a good idea, even under the Bloody Obvious Hypothesis.

      A parent (or a teacher who is very close to the student) can also distinguish between really-trying-to-do-better-taking-on-new-challenges vs busy-work-pretending-to-try-for-praise which is the failure mode of the growth mindset.

  5. onyomi says:

    I really like your point about the biohack movement, as I think I have long had more of a fixed mindset, in that I lack faith in my ability to achieve better results just by “trying harder,” yet I attempt to game the system by doing things I think will improve my ability (though I strongly recommend against the pharmaceutical route).

    That said, your summary of the VCP *does* very strongly resonate with my experience in academia, which is a culture which values “genius” over hard work, and is full of people desperately trying to look smart. Of course, one is rewarded in academia in proportion to how much other academics believe you are smart (as opposed to hard working, experienced, or, even having contributed substantively to a field), so this may be a perfectly rational calculation.

    I also notice in myself a possible negative consequence of the “fixed mindset” which may accord with the SCP or even the VCP: I have a tendency to either try to prove myself in esoteric areas where the competition is weak, or else to not try very seriously when I engage in an area where the competition is strong, because that protects me from the conclusion “I gave it my best, but my best was not enough.”

    If you don’t put yourself out there fully, your natural ability can’t be found lacking, and this seems to me a plausible negative consequence of the fixed mindset.

    • Scott Alexander says:

      I think your academia example mixes “ability” and “innate ability”.

      If a physics professor discovers lots of important new science, no one will hold it against her that she did it by not being very smart but working extra hard.

      But if a physics professor just looks like she’s not really all that good at physics, no one will say “Well, I bet you’re trying really hard!”, they’re going to say “Why did we hire you as our physics professor again?”

      Growth mindset seems to apply most naturally to people who are in the process of learning, not to evaluating what people are expected to already know.

      • onyomi says:

        While I certainly think people do care about acquired abilities in academia, I also think they fetishize innate ability more than most. Middle-tier schools and below are looking for competence, but top-tier schools flatter themselves by thinking they deserve genius. And they don’t care if the work-ethic and actual contributions to the field aren’t really there; they’ll hire the person who seems to have “genius” over the person who seems to have worked very hard.

        Some people in academia are (correctly or incorrectly) perceived as “work horses,” and some are perceived as “natural talents.” The best have some of both, but if you can only give off one vibe or the other, you want to give off the natural talent vibe.

        Also, in your physics professor example, even if the new science she discovered was *actually* the result of hard work, I think the culture of academia is such that it interprets new and innovative findings as the result of genius, almost by definition.

        What’s more, I think academia tends to be deterministic even about effort. For example, there are always certain professors in any field who just seem to be wildly productive. They may or may not be innovative, but they somehow come out with a new book every year. But I think people perceive the “motor” which drives this productivity as itself being innate.

        • Irrelevant says:

          I think academia tends to be deterministic even about effort. For example, there are always certain professors in any field who just seem to be wildly productive. They may or may not be innovative, but they somehow come out with a new book every year. But I think people perceive the “motor” which drives this productivity as itself being innate.

          And do you have reason to believe they’re not correct about that? Perpetual hellbent effort certainly appears to be innate in the “by the time you’re asking the question the answer is fixed” sense, if not the “hard-coded in your genetics” sense.

          • onyomi says:

            Actually, no. I mostly sort of agree with them (hence my interest in “biohacks,” as opposed to “trying harder”), but I worry that that belief, even if accurate, might possibly be detrimental.

  6. Zubon says:

    Compare it to the “10,000 hours to mastery” factoid. Yes, there are some areas where you need 10,000 hours to master something. There are areas where lifelong learning is also the case, rather than having a “master” tier. And then there are areas where the learning curve is short or talent dominates. Running speed is one where talent dominates: you can become faster, but you very rarely see very slow runners become very fast runners.

    It would be interesting if competing in events where talent dominates correlates with fixed mindset.

    I’m not sure how one believes the Bloody Obvious and Very Controversial Positions at the same time, because the latter implies that believing the former is undermining yourself. It’s not logically inconsistent, if you believe you would be doing better if you did not know that talent mattered, but around here we tend not to believe in Noble Lies, especially ones you need to somehow convince yourself of.

    • Steve Sailer says:

      The Dan Plan website was set up by a guy who started about four years ago to take up golf and try to make the pro tour, after reading a Malcolm Gladwell book, by engaging in 10,000 hours of directed practice. He’s whittled his handicap down to 3.1, which is very good, but he still has about 8 or 10 strokes to go be pro-good, and his trendlines over the last two years are not promising:


      • Gbdub says:

        Golf seems like the perfect example – it has often been discussed how diligently and extensively Tiger Woods practices, and strongly implied that this, just as much as anything innate, is what created his success.

        But many golfers practice equally hard, and basically none reached his level of success. Also, Tiger himself isn’t as good as he was 10 years ago, despite 10 years of hard work. Maybe he just has a fixed mindset and it’s hampered his game 😉

        • Steve Sailer says:

          East Asian parents have been trying out Gladwell’s 10,000 Hour Law on their children to get them to become professional golfers for some number of years now. It seems to work pretty well on girls, mostly because there isn’t much competition in women’s golf.

          At my local driving range, a Korean-American boy around ten became a legend for practicing huge numbers of hours. He made the pro tour and won some tournaments, but now he appears to have given up golf even though he’s still in his 20s. When other Korean parents ask his parents how to do it, they tell them: “Don’t.”


          • onyomi says:

            Does Gladwell really claim that 10,000 hours of practice will automatically make you one of the best in the world at x (i. e. good enough to make a lot of money at x)? I always interpreted it as “10,000 hours of practice is enough for most people to be ‘world-class’ at x, as in, ‘the pros could practice with you and you wouldn’t be getting in their way.'”

            Like, if I practiced the violin for 10,000 hours (starting from 0 now), I’d expect to be able to play most any piece with reasonable skill and expression, but not to be playing at Lincoln Center. I think everyone knows that that takes innate talent+10,000 hours?

            And even though that’s true, I still find the 10,000 hour rule encouraging because it means I could, *if I really wanted to*, become decently good at all kinds of things I have no obvious natural talent for.

          • Steve Sailer says:

            Gladwell writes a lot of extreme, absolutist things in the New Yorker that go over well when he’s making a $50k speech to sales conventions. He then tries to backpedal when he gets called on them, but he’s put too much in writing to get away with weaseling out:


            Gladwell’s decision in 2009 to enter into a spat in the New York Times with Steven Pinker over Pinker using my analysis of NFL quarterback draft picks to dismiss Gladwell’s contention that nobody can predict with any accuracy how college quarterbacks will do was the straw that broke the camel’s back in terms of Gladwell’s long term reputation.

            Keep in mind that Gladwell is pretty sincere about most of the stupid things he says — if he were cynical he would not have publicly called out Pinker, who probably has 60 IQ points on him, in an empirical dispute where Gladwell was obviously wrong but Gladwell didn’t realize it. Instead, a lot of the blame should go to The New Yorker, which has a vaunted fact-checking and editorial infrastructure, for letting Gladwell get away with obvious mistakes for so many highly profitable years.

        • http://bloggingheads.tv/videos/34558

          It’s also possible that Woods practiced so hard he injured himself enough to cut his career short.

          • Steve Sailer says:

            Starting early in life appears to be a big benefit in becoming a great golfer, suggesting that nurture is more important relative to nature in golf than in many other fields. Both Woods and Mickelson were swinging golf clubs by age 2. The pros can execute an unbelievable variety of shots, a facility that comes from spending enormous numbers of hours with clubs in their hands. The 10,000 hour rule is close to a requirement in golf. I wouldn’t be surprised if Phil Mickelson put in 25,000 hours before winning his first major championship.

            In contrast, one explanation given for Greg Norman’s somewhat disappointing career, despite being the best athlete on the tour, was that he hadn’t started golf until he was 16 and that’s just too late. Back in the 80s there were a couple of good touring pros, Larry Nelson and Calvin Peete, who hadn’t taken up golf until they were around 21, but that seems rarer today.

            There used to be a lot of talk about how famous team sport athletes like Michael Jordan were going to compete on the Senior/Champions tour for 50+ golfers when they retired from “real” sports, but that has turned out to be much harder than was assumed. I believe QB John Brodie is still the only player from a big sport to have won on the Senior Tour.

            It appears that Jordan Spieth, the 21 year old leader in The Masters, took up golf at age 9. That’s probably about average for modern tour pros. Spieth is just better at golf than any other American of his generation.

          • Steve Sailer says:

            “It’s also possible that Woods practiced so hard he injured himself enough to cut his career short.”

            By the account of his old swing coach, Woods’ worst injury came not from practicing golf but from working out to join the Navy SEALs in his early 30s. That sounds crazy, but it fits with how the wiry Woods blew up to look like GI Joe in the middle of his career:


      • One of the ways 10K hours gets misunderstood is that masters have taken 10K hours (not exactly true– there’s a lot of variation) doesn’t imply anyone can become a master at an arbitrary skill by putting in 10K hours.

        This is from The Sports Gene, both the idea that 10K is very much an approximation, and that there are innate physical requirements for elite abilities for various sports, and those requirements are frequently not at all obvious.

    • Hari Seldon says:

      I think sports is a great place to elucidate the concept Scott is talking about. When we are talking about the importance of practice vs. genetics we are skipping an important point. It matters immensely how much competition is involved before we call a person skilled.

      We rank the level of skill achieved based almost solely on how that person compares to others with the same skill. Sports gives us a great place to look at the difference.

      For highly paid skills like basketball, there is a lot of competition. Innate genetic ability quickly rises to the top. Height and wingspan trump just about everything. If you are 5′ 10″, no amount of practice is going to get you into the NBA. If you are 7 feet tall, average athletic ability is all that is required.

      For universal sports like running, genetic ability rises to the top for the same reason as basketball. The number of competitors is extremely high. It isn’t necessarily highly paid, but running is a near universal experience. You know as a child if you are fast or not.

      For obscure winter olympics sports, mere exposure to the activity may be enough to warrant a medal. You may be genetically in the bottom half of the bell curve for a sport like bobsledding, but simply living in a place where a bobsled track exists puts you at an advantage over 99% of human beings. A little practice is probably enough to put you into the elite.

      Innate ability is the most important factor. But if there aren’t many people exercising their innate ability, practice can take you to the top.

      • And yet, younger siblings are overrepresented among runners. One hypothesis is that they have to run faster to keep up with older siblings, in a way first borns don’t have to run to keep up with their parents.

        • Douglas Knight says:

          Do you have a source for that?

          • I believe I read it in Top Dog (Po Broson and Ashley Merryman) but I don’t have the scholarly citation.

          • Douglas Knight says:

            I do not see it in that book, nor in NurtureShock. Top Dog does say that among pairs of brothers that both played professional baseball, the younger ones attempted to steal more often and had longer careers. Also it claims that younger siblings are “more likely to participate in dangerous sports.”

    • Harald K says:

      “very rarely see very slow runners become very fast runners.”

      A pet peeve here: It depends a lot on how you measure it. Do you measure it as growth mindset skeptics (read: IQ fans) do it, in terms of standard deviations? Then it’s probably quite common for very slow runners to become very fast runners with practice – provided you look on the whole population. If you look on the population of people who already practice running a lot (i.e. runners), then I’m sure it isn’t.

      If you measure in meters per second, then what you can get by practice is probably low even for the entire population. When judging stuff like where to place the emergency exits in the top athlete centre, you can and probably should disregard their greater running speed.

      In the real world, it’s often right to measure things by meters per second, rather than the always popular bell curve. Sometimes the top 0.1% don’t deliver appreciably more in absolute terms than the top 10%. There are times it’s appropriate to care about normalized, relative difference between individuals, but mostly that’s not what you really care about.

  7. dipitty do says:

    Dweck is obviously just trying to make “innate ability” unmentionable in conversations about ability, by making everyone afraid that even thinking about it will suddenly make them incompetent. No child should ever be taught that innate ability exists; thus the future generations grow up without even knowing about it. Then through the magic of trying really hard, we’ll all actually perform equally well.

    Everyone knows that innate ability matters at least a little bit, but we’re never supposed to say that, for fear that someone might use that information to explain racial IQ differences. Which is why ordinary folks, not on board with the hush-hush agenda, keep confusedly responding, “Huuh? But of course innate abilities matter, too.”

    • Emile says:

      Dweck is obviously just trying to make “innate ability” unmentionable in conversations about ability

      “Bob believes X” -> “Oh no, Bob is trying to make non-X unmentionable!” … I’d rather see less of that kind of hyperbole (no, adding “obviously” doesn’t make the paranoid reading more convincing.

      (and could we try not to drag race & IQ into this?)

      • dipitty do says:

        Too late. Race and IQ have been in this for decades.

        And I don’t really care what kind of hyperbole you’d rather not see.

      • Anonymous says:

        @Emile, I agree! It’s silly, and makes legitimate criticisms of Dweck seem that much less so.

      • Irrelevant says:

        I’d rather see less of that kind of hyperbole

        And I’d rather it were hyperbole, but here we are.

        Belief in innate ability gaps in particular has been dubbed the “Voldemort View” for a reason, and that’s far from the only political topic where the argumentation strategy of attempting to make the opposing view unmentionable is in common use.

        • Anonymous says:

          Belief in innate ability gaps in particular has been dubbed the “Voldemort View”

          Maybe you mean, belief in ability gaps between racial groups?

          • Irrelevant says:

            No I did not. Though I was referring to the context of education, which was implicit in the first post of this thread but which I failed to spell out.

          • Douglas Knight says:

            But that is what the phrase means (even more clearly here, which is probably Irrelevant’s source.)

          • Irrelevant says:

            If you read what you linked, it’s more general than racial gaps. The Voldemort View is what Gladwell is quoted as calling “IQ Fundamentalism”: the theory that the dominant factor in the performance difference between low-achieving and high-achieving students is that the high-achieving students are actually more intelligent.

            Race is simply the area where the problem is the hardest to obfuscate.

      • ADifferentAnonymous says:

        I’d agree that the comment is not clearly true and is generally unkind in its tone, such that is rather not see its like. But on the other hand, growth mindset theory has probably at least gotten a sizable boost from the general air of “boo innate ability” in progressivism, arising largely from, yes, race and IQ.

        • Emile says:

          It may have got such a boost, but that’s still a far cry from “making ability unmentionable” being Dweck’s goal in the first place (as opposed to improving how we raise our children).

        • social justice warlock says:

          Innate ability being very important is fatal to political egalitarianism on all fronts, not just race. If there are natural superiors and inferiors, the superiors can of course act benevolently towards their inferiors (or not,) but that’s charity, not solidarity.

      • dipitty do’s claim doesn’t follow from the fact that Dweck makes the arguments she makes. But it might still be true. One would need to know more about Dweck, and about the people who accept and push her views, to tell.

        • Steve Sailer says:

          Dweck hardly needs to make anything unmentionable since it already is. How much money does Malcolm Gladwell making on corporate speaking gigs versus Charles Murray?

  8. Consider a very smart child who calculates (A) how important effort is to academic success, and (B) how much effort is “appropriate”. This child knows that it takes him very little effort to do extremely well at school. Giving this child a growth mindset will indeed cause him to increase his estimate of (A) and this might make him work harder, but it will also cause him to lower (B). Rather than downplaying the importance of innate ability, this kids parents should tell him that although right now school is extremely easy for him, someday he will compete with people at least as smart as he is, so he needs to push himself to study advanced material and learn how to tackle difficult problems.

    • dipitty do says:

      Or just let the kid study material appropriate to his ability/skill level because that’s where he happens to be performing, like everybody else.

      • But doing school the kid will be mostly studying material well below his ability level.

        • dipitty do says:

          The simple answer is that then you don’t waste the kid’s time with school.

          The real answer is that most schools have some sort of program in place to give kids level-appropriate materials, outside of k-2 math. Many districts will let talented highschoolers start doing college-level work if they want to.

          • Anonymous says:

            Many districts push back against this type of accommodation. Hard.

          • Ptoliporthos says:

            I’d second that comment about pushback from the schools. Even schools that provide this accommodation have teachers who resent it and try to undermine it.

            When I was growing up my school system had a gifted and talented program that took kids out of the regular classroom one day a week to an enrichment program at another school in the district. Without fail the regular classroom teachers would present new material on that day, and give a pop quiz on it first thing the next morning. I’m not sure why they had so much resentment about the program, or why they thought the solution was to take it out on the kids, but that’s just how it was. Perhaps they read “Harrison Bergeron” and thought it sounded like paradise.

          • You can avoid that set of problems by having entirely separate schools for the gifted, but that approach has problems of its own. (Oh those merry days at St Nietzsche’s)

          • William O. B'Livion says:

            Even schools that provide this accommodation have teachers who resent it and try to undermine it.

            They should be given the opportunity to find a career more in line with their goals, desires and abilities.

          • Null Hypothesis says:

            Perhaps they read “Harrison Bergeron” and thought it sounded like paradise.

            But a paradise, of course, no better than any other.

          • Harald K says:

            Many districts push back against this type of accommodation. Hard.

            And why do you suppose that is?

            Maybe because of how it affects the others? “Math isn’t my thing, it’s little Einstein’s thing over there”.

            Maybe because parents are sometimes overly optimistic on their kids’ behalf?

            It could also be that they’re just some sort of Randian hobgoblins trying to keep your kid down. I’m not ruling it out, but I think we should look at all possibilities here.

          • Irrelevant says:

            Randian hobgoblins do not try to keep your kid down. Randian hobgoblins trade your kid for a changeling of equal market value.

          • Other says:

            If you interpret “Randian hobgoblin” to mean “Rand’s acolytes” as opposed to interpreting it to mean “the hobgoblins of Rand’s stories.”

      • This is hard to do when you have a class of 25+ kids, but is getting much more possible by using online content that has built in assessments to see what kids have mastered.

        North Carolina just passed legislation to give credit for demonstrating content mastery, instead of mandated seat-time before you can get credit. You have to have this before you can let kids go at their own pace.

        The future of education is moving towards meetings kids where they are and moving them forward at the pace that is right for them. This is not really the Growth Mindset, but it is not the Fixed Mindset. Especially if you were fixing by demographic characteristics.

        Several things make this hard besides the fact that it is not what we are used to. The role of the teacher changes. Teachers are still important because you have to know what you are doing to select high quality content and organize a classroom. There are huge social implications. The kids who may learn the fastest and get the furthest may not be the same social class that is now in the advanced courses. This is a huge culture change.

        • Other says:

          It’s also not Fixed Mindset. Especially if you were fixing by demographic characteristics.

          Who’s doing this?

          You’re taking Dweck’s strawman and making it even weaker. Even the claims of phrenology (as practiced by Franz Gall and people who actually read him, as opposed to what random armchair-phrenologists who neither read nor wrote about their practices may have done) didn’t go so far as to say that the size and shape of your skull in particular locations fully determined your intellectual abilities with respect to what the brain regions behind that portion of skull were capable of doing. Merely that the two were strongly correlated. And phrenology comes a lot closer to being a theory of Fixed Mindset than any school of biodeterminism or racism today.

          Nobody is looking in the mirror saying “I’m Kenyan; therefore, I have the ability to run fast” or “I’m Asian; therefore I have the ability to be good at economics.” A few people might be looking at themselves in the mirror and saying “I’m a girl; therefore I will never be good at physics” or something similarly ridiculous. More people are looking at other people and making these sorts of stereotypical judgments, but that’s irrelevant to a conversation about mindset, which is about the stories people tell to themselves about why they succeed or fail.

    • What I remember my parents telling me was that what mattered was not how I did compared to the other kids but compared to what I could do. I don’t remember any warnings about future competition.

  9. Steve Sailer says:

    Scott says: “give you about 50-50 odds that it matters in real life.”

    This is a general insight that I’ve been arguing for awhile starting in the late 1990s when I reviewed Arthur Jensen’s “The g Factor:” the things where we argue the longest and best over whether the glass is full or empty, such as nature v. nurture or (later) Tom Brady v. Peyton Manning, tend to be those where the glass is roughly half full and half empty.

    • HeelBearCub says:

      the things where we argue the longest and best over whether the [measured quantity is above or below a threshold] … tend to be those where the [measured quantity is not obviously above or below the threshold]

      I’m not sure if you meant to make the statement above, which is basically trivial, or something else, but it reads the way I re-wrote it.

      Can you expand on your point?

      • The point is that we argue the most over the things that matter the least. Argument happens when it’s not clear which of two options is better/more accurate. But if it’s really not clear which of the two is better, then probably they’re about equal and so it won’t matter which we pick in the end. In cases where one option really is much better than the other…well, in that case it’s usually obvious that the better option is better, and so no one argues about it in the first place.

        Or at least that’s how I read it.

      • There’s also a tendency to argue that the glass is 100% full or 100% empty, or some other very high percentage, or at least to act on the implications of such a belief even if it’s not explicitly stated, rather than acknowledge even a small effect from the ‘wrong’ side.

      • Steve Sailer says:

        Can you expand on your point?


        In 2002, Steven Pinker told me: “mental effort seems to be engaged most with the knife edge at which one finds extreme and radically different consequences with each outcome, but the considerations militating towards each one are close to equal.”

        It’s not necessarily trivial. For example, the biggest NFL draft hot stove league argument of all time was over who should be the #1 choice in the 1999 pro football draft: Peyton Manning or Ryan Leaf. I favored Indianapolis picking Leaf (who turned out to be a disaster), but I was willing to admit there was a lot of evidence for picking Manning (who turned out awesome). That’s what made the argument fun. The right answer wasn’t obvious but it wasn’t unimportant either (relative to NFL history).

        • Steve Sailer says:

          This is related to the argument in the New York Times between Pinker and Gladwell over drafting NFL quarterbacks that proved the downfall of Gladwell’s reputation as a serious intellectual.

        • HeelBearCub says:

          I agree that in any situation where a small change in absolute quantity creates a large value difference, and that value curve looks asymptotic, their will be a great deal of effort expended on discussing the “true” measure of the value. Perhaps Leaf vs. Manning is an example of that. There might be an asymptote that looks like x = NFL level QB, but that curve must level back off, as the difference between (say) Trent Dilfer and Peyton Manning would attest.

          Brady vs. Manning (and most GOAT type arguments) are not examples of that, I don’t think. Here you are looking at an arbitrary number of data points near the end of the distribution and saying “pick one and only one and call it the best one”. That’s more of an arbitrary and artificial asymptote, not a natural one.

          • Steve Sailer says:

            Brady v. Manning is an oddity in that both have continued to perform at an extremely high level for an absurdly long time, so that any kind of big money decision that had to be made about which one was worth investing a massive contract for has turned out to be surprisingly unimportant because both would have been fine. But that’s very rare. Old sports pages are full of debates over who is better that seemed pretty equal at the time, but in hindsight didn’t work out that way.

            A more common case would be the discussion in the late 1940s of trading Joe Dimaggio for Ted Williams because the right handed Dimaggio was better suited for Fenway Park and the left-handed Williams for Yankee Stadium. Which one was better? There was a lot of evidence on either side at the time.

            But that trade would have been a disaster for the Red Sox to make because Dimaggio retired after 1951, while Williams played through 1960. So, even seemingly matched arguments can prove to be very important.

  10. Andy Harless says:

    “The more a player believed athletic ability was a result of effort and practice rather than just natural ability, the better that player performed”

    I think the literal interpretation of this statement, given the inclusion of the word “just,” refers to the player’s confidence in the proposition that effort matters at all (vs. “just” natural ability, i.e. 100% natural ability with no other ingredients), rather than her belief about the degree to which effort matters. It’s a weaselly statement that begs to be distorted the way Scott is distorting it, so that readers will believe it has some actual substance. Which (since I know little about Dweck other than what I’ve read in this and the previous blog post and in Noah Smith’s response & comments thereto) leads me to suspect that Scott is attacking a straw woman because the actual women is made of something even less sturdy than straw. But then I’ve long suspected (since spending a year a psych major in college many years ago) that the field of psychology is made almost entirely of jelly.

    • Deiseach says:

      What’s the “better” judged against here, though? The player who practiced more and worked on his weaknesses and developed his skills played better next year than he did last year – he improved his own performance – or he played better than the ‘gifted but lazy’ player who had more innate ability but frittered it away?

      I think that matters, because if we’re talking about ‘you can improve what you have by putting in effort’, I think everyone agrees; if we’re talking about ‘less talented but works hard can beat more talented but lazy’ then it happens at times; but if we’re to take it that ‘hard work alone beats talent plus hard work’, I don’t think so – unless we mean things like defenders being instructed that since they’re going up against the star player on the opposition team who is world-class and there is no way they can beat them on ability, so take them out of the game by clattering into them with hard tackles every chance they get.

      • HeelBearCub says:

        Athletics is a little bit of a strange beast, though, in that innate ability is static under some measures, and dynamic in others.

        An individual players innate ability is fixed and “work” will determine their net ability (this is the obvious position).

        A team’s innate ability is fixed once a season starts, but dynamic between seasons. In addition, there are a variety of rules that are intended to level out the NET ability of teams coming into the season.

        This means that strategically a coach/GM will want to do two things:
        a) Get players to do the maximum amount of “work” (where work is defined as effort which results in increased ability)
        b) Find players that have more untapped innate ability, which will mean means a greater output potential due to work

        Bottom line is that a coach is going to tell players that effort is the only thing that matters, but GMs will act like innate ability matters more.

  11. Sammy says:

    I’d just like to say that my personal experience is similar to Scott’s – that as long as you believe the Bloody Obvious position you’ll work as hard as you otherwise would (because of your conscienciousness score or whatever. Come to think of it, has anyone done a study on the conscieniousness of fixed vs growth mindset personalities? I predict that there would be a small or no difference), regardless of your factual beliefs about innate ability vs intelligence.
    I’d heard of the growth mindset and its promotion as a really excellent way to think before and noticed that I was very, very confused when I read your last article on how the evidence for it working was so strong. I’ve always dismissed people who say that effort accounts for most of the difference in people’s acheviments (and doubt I could make myself believe it if believing it actually did make me into a better person) and I do sometimes attribute my own failures to a lack of intelligence. The idea that believing that would actually reduce my level of effort didn’t really occur to me until I read these last two articles.

    The ability ceiling (sub-) position you mentioned pretty much sums up my implicit beliefs about the importance of intelligence vs ability in a high-IQ environment. I’m reminded of the quote by Warren Buffet:

    You don’t need to be a rocket scientist. Investing is not a game where the guy with the 160 IQ beats the guy with 130 IQ.

    • Douglas Knight says:

      A study of UCL students found a correlation of 0.29 between mindset and conscientiousness. That’s higher than the correlation between mindset and anything else, but not that high. It is lower than the correlation between conscientiousness and grades.

  12. Steve Sailer says:

    A massive problem in 21st Century intellectual discourse is that so many people feel we must censor frank discussions of social science to not disillusion children. What we need is a grown-up space where people like Jason Richwine, James D. Watson, and Larry Summers can discuss the findings of social science without losing their jobs to encourage the others.

    In contrast, while Victorian England had all sorts of censorship on sexual matters, it also preserved a space for adults to discuss sex because it was important to running the Empire. For example, the explorer and sexologist Richard Burton got knighted by Queen Victoria three years after quietly publishing his translation of the Kama Sutra.

  13. Sniffnoy says:

    Man, the Very Controversial Position really does not sit well with me. In particular, why would having a fixed mindset cause one to abandon success as a goal?

    Here’s a version of the Very Controversial Position that makes more sense — we can call this the Cynical Controversial Position. The Cynical Controversial Position states that, in fact, everyone has abandoned success as a goal; everyone is just trying to look good. It’s just that, if you believe effort matters more, then the best way to look better is to actually become better; while if you believe ability matters more, then that’s not an option, so you spend your effort on things that only improve the appearance of success. (Although it’s probably worth noting that this still requires believing that effort matters when it comes to the appearance of success!)

    This is not a position that I would actually endorse. But I think it rescues some of the things in the Very Controversial Position while having fewer totally unexplained bits. And since a lot of this is working at a correlational rather than individual level, perhaps some of the results could be explained by the Cynical Controversial Position being true for such-and-such a fraction of the population.

    • Viliam Búr says:

      why would having a fixed mindset cause one to abandon success as a goal?

      Because people are stupid and our instincts are often miscalibrated for the environment we are living in. Also, we live in a social context, which sometimes gives us benefits for lying, and sometimes we overdo it, start lying to ourselves, and then sometimes we fail in some way we don’t see because we have blinded ourselves.

      Specifically in this case it may go like this: “People believe that ability matters, right? That means I could be refused an opportunity even before I have a chance to try working, if someone gets an impression that my ability is not enough. On the other hand, I could be given an opportunity even without doing any real work, if someone gets the impression that my ability is great. Seems like projecting an image of great ability could bring benefits in itself.”

      So far, it’s reasonable. But at some moment the tradeoffs start coming — you should not do X, because it could harm your image of having great ability. And sometimes people protect their image so much it actually makes them fail. Because people are stupid.

      The dangerous values of X are: trying to solve a really difficult problem; starting to learn something completely new; doing something that someone else near you is already good at; etc. All of these can make you fail or appear worse than someone else in the eyes of observers. Yet they are the way to grow, so if you protect your image, you sacrifice your growth.

      It is easy to see why protecting the image too much may hurt the actual skills. But I think that in real world for usual human goals image also matters, and nerds may underestimate this part. There are many people who got laid or got rich because they succeeded to impress the right person at the right moment, even if their strategy did not lead to having optimal skills in the long term. Yet they are successful and happy.

      And some other people were trying to do the same thing, but they were doing it wrong, and they failed.

      • Harald K says:

        I agree with you, but I’d go even further: I’d say it isn’t necessarily even stupid, or miscalibrated instincts. Maybe self-handicapping actually makes sense when we do it. It’s not optimal for society, but it may well be sensible for the individual.

  14. Irrelevant says:

    I’ll reiterate my question from the previous thread in hope it gets better visibility here:

    The “Very Controversial Position”, as a factual claim, seems to be getting the cause and effect wrong. Instead of people with fixed mindsets then being unethical, failure averse, self-sabotaging, and highly protective of their self-image, it would make more sense that people who are highly protective of their self-image will be failure averse, even to the point of cheating and self-sabotage, and also will attest to believing that ability is the main thing that matters if that’s what protects their egos at the moment. We should be able to check this by finding the risk-averse self-sabotaging people and seeing if they’ll say it’s about effort instead when that’s the better ego defender. Has anyone tried that?

    • Sniffnoy says:

      Yes, this also makes more sense than the VCP.

      Edit: OK I guess this pretty similar to what I was saying when I said “maybe the Cynical Controversial Position is true for a portion of the population”.

    • Irrelevant says:

      Also I just realized how closely this echoes The Last Psychiatrist’s old wardrum of the problem being narcissists.

    • Scott Alexander says:

      Really? Most of Dweck’s studies are done after someone has failed. If a person was just self-esteem-protecting, wouldn’t they say “I failed because I wasn’t really trying” rather than “I failed because I wasn’t smart enough”?

      • Irrelevant says:

        I confess I stopped at your summaries rather than digging into the studies themselves, but they seem only tangentially related to the claim of the Very Controversial Position.

        With respect to the studies, the first seems to mostly be looking at children’s ability to do what they thought the researchers wanted, and be capturing the difference in effort between people to whom it’s suggested that they learn during the test and people to whom it’s suggested that they’re supposed to know it already. And in the second, which as far as I can tell doesn’t appear to have checked the abilities of the students outside the test, it groups people so strangely that I’m not convinced her “exhibits learned helplessness” category wouldn’t be better labeled “accurately believes they are unintelligent.”

        In short, I don’t think Dweck is studying what she thinks she’s studying.

        • Arceris says:

          I would agree with @Irrelevant’s analysis, however I would extend it with an additional alternate reading. Perhaps there is an unexplored “moral” dimension here, which might turn the explanation on its head.

          For instance, if one has been raised with a traditional “protestant work ethic” (or something similar – Asian work ethic, etc), where hard work has been linked to a moral imperative, and you’re asked the question “did you fail because of lack of effort or lack of innate ability”, I think there would be a strong impulse to answer “innate ability”. Because your moral code is wrapped up in effort, it will be harder for you to admit failure on that score, than a simple lack of ability. People don’t like to admit moral failings, or lose face, and once a loss is preordained (you failed the problem set), I think that people will do a calculation and choose the lesser loss of face based on their experience/outlook (I know I do).

          While I don’t believe the above is a dominating effect, it would be another confounding possibility in her results.

  15. Deiseach says:

    But a person with growth mindset risks the opportunity costs of prolonging their inevitable failure instead of (in the Silicon Valley term) “failing fast” and pivoting towards a higher-payoff activity.

    I think that is the major justified criticism of the popular version of “growth mindset”, and it’s exemplified by the saying “Throwing good money after bad”. Someone who has been drilled that you’ll succeed if you only work hard enough, and that you can go as far as you want to go, no limits, just put in the time and the effort, may respond to “Well, goodness, looks like my business selling flood insurance in the middle of a drought isn’t doing as well as I’d hoped. Guess instead of working twelve hours a day cold-calling and doorstopping people, I’d better ramp it up to fourteen hours a day!”

    No, sometimes you have to walk away and give up. And sometimes you’ll lose even when you’ve done all you can, not because you weren’t hard-working enough but because the other person was better due to natural ability (and/or a bit of luck).

    We’ve all seen talented people who skated by on pure talent and didn’t do the grind of practice which, when they hit the patch when their talent wasn’t firing on all cylinders that particular day, would have got them through it. Nobody denies that, no matter how talented you are, you still need to practice and learn.

    But drumming it into people that yes, you too can be a movie star/sports star/billionaire if you just work really, really hard and keep at it is equally unrealistic. Sometimes second or third place is as good as you’re going to get, and it’s not the worst to get, either. Making people feel guilty for not working hard enough, making them drive themselves into physical and mental breakdowns by putting in too much effort, is as cruel as telling them “You’ll never get anywhere, you’re too stupid/ugly/lazy”.

    • onyomi says:

      “…drumming it into people that yes, you too can be a movie star/sports star/billionaire if you just work really, really hard and keep at it is equally unrealistic.”

      Yes, as I mentioned in the other thread, given that most jobs are not especially fulfilling, but most people in developed countries can at least expect to make enough money to enjoy some leisure time, take care of a family, etc. IF they work hard, the “follow your dreams” message may be increasing the probability of unhappiness in a large proportion of the population for whom a job that is BOTH exciting and fulfilling AND reasonably remunerative is simply not an option.

      In fact, by promoting “follow your dreams” we may be giving a slight boost to the naturally talented and driven, who would probably have done so anyway, at the expense of slightly harming the great mass of people who are neither especially driven nor unusually talented in any one area.

      Instead of “work hard and you can do anything you want,” maybe the better message is “work hard and you have a very good chance of finding a tolerable job that makes you enough money to enjoy what you really love in your off time” (and this message also has the advantage of being the truth, while the “follow your dreams” thing seems, at best, an expedient lie).

      I’m going to sound like my anarcho-socialist labor movement friends (with whom I rarely agree) for a moment here and say that the “work hard and achieve your dreams” message seems almost like a capitalist’s convenient way of distracting people from the fact that it’s harder and harder to just get a regular job that pays the bills. Nowadays, if you can’t pay the bills it’s not because the labor market sucks, has less and less need of non-creative workers, etc. but because you haven’t adequately followed your dream, pursued your passion, etc. etc.

  16. pxib says:

    Two comments:

    1. Fundamental attribution error: My situation is based on my specific circumstances. My effort is an example of those circumstances. It is easy to believe that effort (rather than character traits) are responsible for my own successes, and that lack of effort is responsible for my own failures. Easier, at any rate, than believing they were predestined at birth. Otherwise the attribution error wouldn’t be fundamental.

    2. Innate ability includes effort, only it doesn’t feel like real effort. It feels like fun. It’s stimulating and engaging and worth thinking about and practicing. The effort still matters, but there’s less mental cost to engaging in it. This tiny advantage has the opportunity to compound over the entire experiential life of talented individuals. Hour by hour, day by day, year by year they are enjoying what the untalented consider “hard work”. Nothing in their mind rebels and wishes they were doing something else. There is nothing they would rather do. In a very real way, ability and effort are the same thing.

  17. Bill Openthalt says:

    I keep coming back to the political incorrectness of innate ability. The growth mindset is attractive because it means all people are innately equally capable, whatever their race, gender, or class. It is attractive because it means everyone can easily become more successful simply by changing a mindset. It is attractive because it vindicates affirmative action, gender quotas, extra public money spent on low achievers, lowering school standards to lower the number of kids failing, etc. We’re ideologically committed to equality, and hence studies ‘proving’ equality are attractive and people design studies to prove just that, like cultures that believe in innate inequality will produce studies confirming that belief.
    Confirmation bias in action.

    • Scott Alexander says:

      Yes, but everyone has confirmation bias. I am very reluctant to say “It is very suspicious that your facts support your narrative” until there’s good proof that the position is wrong. After all, my facts support my narrative too!

      • Bill Openthalt says:

        Granted, but in the case of the growth mindset supports the current social consensus — equality of (intellectual) ability independent of gender or any (real or perceived) divisive attribute we attach to individuals.
        The reason why their ‘facts’ are suspicious is that they are not supported by reality — we know it takes talent (of the physical or intellectual kind) to excel at anything. We also know it takes a lot of effort to excel at anything, but that it takes less effort when one is talented. Precocious readers, for example, simply haven’t had the time to put in the required effort to explain the chasm in reading proficiency. Unfortunately, it’s not easy to devise an experiment (like Dweck’s) to formalise the evidence that children who teach themselves to read at age 4 are ‘more talented’ than kids who are still functionally illiterate at age 12, their parents’ efforts notwithstanding. Dweck has no facts, just some nice-sounding explanations of short-term effects, explanations that support the current social consensus and hence are morally difficult to impeach.

        • Emp says:

          “we know it takes talent (of the physical or intellectual kind) to excel at anything”

          What’s the definition of ‘talent’ here?

          Is there any method of identifying beforehand who a talented person is?

          The only thing that is actually known for a 100% fact is that some effort is required (even if only effort so minimal as 2,500 IQ guy bothering to learn the rules of chess and showing up at the board).

      • Steve Sailer says:

        Dweck is restating the conventional wisdom. There’s a huge market for motivational speakers.

        This isn’t to say what motivational speakers tell you is 100% wrong. There’s a lot of truth to the conventional wisdom, just as there’s a lot of truth to the much more unpopular counter-idea. The truth likely lies in between the two straw persons.

  18. Anr-X says:

    I feel like we need to give more attention to the social component, here.

    Like, it seems that the outcome Dweck’s quote is presenting is actually very largely due to a ‘people thinking you have low ability is super bad, so if you have low ability you must hide it’ type mindset.

    But, ‘ability is important’ and ‘people thinking you have low ability is super bad, so if you have low ability you must hide it’ don’t actually have to show up together.

    If I believed ability was important, and it was desirable to figure out my abilities so that I could make the best decisions about my life from there and that people would generally support me in doing so, then it seems like that also solves the issue Dweck is worried about. Because the issue she’s worried about doesn’t come from me thinking I might have low talent, it comes from me prioritizing *looking* like I have high talent.

    So it seems to me that the important thing is in fact acceptance – both socially and personally, having the idea that it’s *OK* to be low talent. It’s not a sin to be concealed, it’s just useful information to have. And I can go forward based on that, whether by deciding to go focus on another area instead, or deciding ‘well, this area is important to me so I will work hard to maximize my capabilities, even as I know they won’t be as high as other people’s’, or deciding ‘well, I guess I’ll never be elite in this area, but I enjoy it and this level is fine with me, so I can just keep casually working on it’. Like Scott wrote about in one of his talents posts – accepting that you are not going to be the saves-the-world-with-math person, that’s the way things are, it’s OK, and you can go forward with your life from there.

  19. Paul Torek says:

    Let’s play a game called spot the contradiction! Some quotes from the previous post:

    Good research shows that inborn ability (including but not limited to IQ) matters a lot, […]
    [Dweck’s claim is] that belief that ability might matter is precisely what makes people fail.

    Did you spot the contradiction between Scott’s claim, and Dweck’s? I hope not, because it isn’t there.

    Let’s play Spot the Implication! Quotes from the present post:

    The more important you believe innate ability to be compared to effort, the more likely you are to stop trying, to avoid challenges, to lie and cheat, to hate learning, and to be obsessed with how you appear before others

    Call it the Controversial Position.
    SCP: The more children believe effort matters, and the less they believe innate ability matters, the more successful they will be. This is because every iota of belief they have in effort gives them more incentive to practice.

    Given that the Sorta Controversial Position is supposed to be a mild version of the Controversial Position, you would kinda expect an implication from the latter to the former. Did you find it? Hint: it isn’t there. Can you think of an alternative explanation, not involving incentives as such, whereby CP could be true even while SCP is false? Why yes, yes you can.

    Let’s play Spot the Implication, again!

    [People with fixed mindsets] are so concerned with being and looking talented that they never realize their full potential. — Dweck
    Carol Dweck unambiguously believes the Very Controversial Position. [which claims, among other points, that] Belief in the importance of ability […] is worse than merely believing that success is based on luck, or success is based on skin color,

    Hmm, fixed mindset. Could luck be fixed? Is skin color fixed?

    Edit: almost forgot this:

    There is nothing in there about “the more a player realizes that, no matter how important innate ability is, effort matters too.” Her statement says that it’s entirely about what degree a player attributes success to effort versus innate ability.

    Well, her statement only mentions one factor, which absolutely does not imply that she thinks there’s only one factor. Yes, there is room for more-complex hypotheses, which would require more-complex experiments. A scientist has to start somewhere.

    Hey, if I make three logic errors in one day, that’s a good day. But Scott just made three in a very short space, and, well, he’s smarter than me. Maybe his problem is that he knows it 😉

    • Douglas Knight says:

      Are you sure you read the post?

      • Paul Torek says:

        Maybe I’m being paranoid, but it seems to me that every time Scott rephrases Dweck’s position, he makes it worse. For example, implying that skin color doesn’t count as a fixed factor in Dweck’s sense, so therefore she must think that belief in ability is even more disheartening than belief that people of your race are totally being kept down. I’ll admit I haven’t read Dweck, but “fixed mindset ” sure sounds like it would include skin color.

        Which of my pairs of Scott’s statements did you think I failed to understand?

        • Douglas Knight says:

          Thanks for the improved tone.

          You understand the first pair just fine. But then you put nonsense in Scott’s mouth. Scott not only doesn’t say that they contradict, he explicitly disclaims it. Maybe his disclaimer next to the statements is too subtle and his extensive disclaimer too far away. But those statements you quote are not part of the discussion of growth mindset! They are in the section on Scott’s biases. There is no point in worrying about the local logical structure if you don’t understand the global logical structure.

          • Paul Torek says:

            Hmm, you’re right, there is a subtle disclaimer shortly after the two statements, showing he doesn’t see any contradiction. I guess that was one of my three daily logic mistakes.

    • Scott Alexander says:

      I don’t know what you’re doing. Could you explain this using short words?

      • Paul Torek says:

        You have either strongly suggested (first two of the four cases above) or outright stated several logical relationships between statements/positions, which don’t obtain. For example in the first case, you say that a reason you’re biased against Dweck’s thesis is that you think inborn ability matters. But there is basically zero evidential relationship between “inborn ability (in fact) matters” and “belief that ability matters has various bad effects.” Now, I can see why it’s somewhat uncomfortable to hold (as I basically do) both positions. But unless you’re accusing yourself of wishful thinking – which I don’t think you’re very guilty of in general – I don’t see why thinking that ability matters would bias you against Dweck’s claim.

  20. Paul Torek says:

    OK, time for me to be constructive. Most of the problems Scott is raising just vanish, it seems to me, if we remember that Homo Sapiens != Homo Economicus.

    We can, and IMO should, hold that

    Good research shows that inborn ability (including but not limited to IQ) matters a lot, and that the popular prejudice that people who fail just weren’t trying hard enough is both wrong and harmful.

    While simultaneously putting substantial to quite high probability on

    The more important you believe innate ability to be compared to effort, the more likely you are to stop trying, to avoid challenges, to lie and cheat, to hate learning, and to be obsessed with how you appear before others.

    Human beings are heavily influenced by emotional states. Emotions are not divorced from cognition. But neither, contra Hume or at least some of his intellectual descendants, is there a clean division of labor between passion and reason. Part of “emotions not divorced from cognition” is that beliefs – or maybe better, aliefs – matter to emotions. Part of “no clean division” is that reason does not automatically adjust to some clearcut set of goals. It does not go like this: “well, I want to be financially successful, five utilons per standard deviation above average, but I also want to protect my self-image, six utilons per standard deviation above average; and I believe that success is 0.5 ability, 0.3 effort and 0.2 unpredictable environmental variations; therefore the best strategy is …”.

    I think, given the enormous brainpower on this blog, I can leave the rest of what I would say unsaid; you’ve already figured it out.

    I think, given the careful and thoughtful reasoning practiced by readers of this blog, that I can leave constructing the rest of the story as an opportunity for further self-improvement.

    • Paul Torek says:

      I should add that Scott is extremely right to highlight the difference between explicit and implicit beliefs (aliefs), in my view. The implicit ones are probably driving the phenomenon Dweck observes, which is why asking about incentives for effort looks like a way of barking up the wrong tree. Incentives engage explicit beliefs, mostly. Well, at least the way most people around here seem to think about incentives, i.e. rational choice.

  21. Chris Thomas says:

    “I will call this the Bloody Obvious Position, since I’m not sure anyone in history has ever denied it.”
    And then:
    “There are a lot of growth mindset partisans online who don’t believe the Bloody Obvious Position…”

    Which is it Scott!? Which is it!? Or do these sneaky partisans simply lack belief, but refrain from denying?

  22. darxan says:

    Scott says:

    I will call this the Bloody Obvious Position, since it’s hard to believe it isn’t true. I once tried to imagine a world without it as a thought experiment, but it was pretty weird and I wasn’t serious.

    From the post in question:

    “Child prodigies” are autistic types who don’t understand the unspoken rules of society and so naively use their full powers right away. They end out as social outcasts not by coincidence but as unconscious social punishment for this defection.

    Are sick burns on one’s more musically talented brothers in thought experiments also no vice?

  23. AR+ says:

    If you put a gun to my head, I’ll say it certainly works in the lab, and give you about 50-50 odds that it matters in real life.

    If you keep saying stuff like that, eventually there is going to be enough predictions for somebody to conduct a test of how well your predictions for how you’d act when somebody puts a gun to your head match up to how you act when they actually do.

    • Null Hypothesis says:

      “Yes, it’s loaded. Now, first question: ‘How much do you predict your predicted answers in this situation will align with reality?’ And remember, this is for posterity, so be honest.”

  24. The deep irony of this exchange is that Noah Smith argues for Dweck’s theme of growth mindset, but then uses his brilliant talents polemically, and when challenged just gives up. Perhaps because he thinks his talent can normally carry him through. While Scott is digging deep and trying to get to the bottom of things with far more writing and detail and citing studies. Perhaps due to his belief in effort over ability.

    Perhaps the real winner here is Angela Duckworth. Grit is the right framework to explain both Dweck’s results and why this exchange went the way it did. I enjoy this blog because of the intellectual grit. Keep up the good work!

    • Scott Alexander says:

      I would prefer people not bother Noah more than they already are.

      I once heard someone ask if grit was just the, uh, gritty reboot of Conscientiousness. I’ve never heard a good answer for that one.

      • Douglas Knight says:

        I believe that there are different studies putting the correlation between grit and conscientiousness at 0.3 and 0.7. There are even studies concede that it is a subcomponent of conscientiousness, but claim that it has better predictive power. I have to wonder if these studies are all defining grit the same way.

  25. TomA says:

    I think that you’re trying to imbue this topic with more structure and coherence than is justified by the available evidence. In life, you play the cards that you are dealt, regardless of what your genetics has provided as innate ability. And propagating the habit of hard work and practice is logically an enhancement to species success (and the corresponding evolutionary reproduction-based persistence).

  26. I haven’t read through the thread yet, but I think the obvious guess is that true beliefs pay. Success in most activities depends in part on ability, in part on effort. If you believe it depends entirely on effort, you will waste a lot of time and energy attempting the wrong things, things you are not going to be able to do because you don’t have enough of the relevant ability—learning to be a good singer in my case, for example. If you believe it depends entirely on ability, you won’t put out much effort and will achieve much less than you could have.

    • Richard says:

      So, if I am unpacking this correctly:

      Forming a correct model of reality and acting accordingly has a high chance of success.

      I’m willing to accept that as an axiom really.

      For this to track with the research, the studies in question are then not measuring growth mindset as such, but rather how correct the people in the study assess the length of the dotted line in Scotts graph, and people who judge it to be shorter than it really is perform worse than people who judge correctly and also worse than people who believe it is longer because in both the latter cases they will reach point 2, while the first will stop somewhere along the way.

      Also, people who overestimate the length of the line will waste effort trying to improve after reaching point 2, but this is not shown in the research because the studies don’t even try to measure that.

      That seems like a scary elegant explanation, so I’m probably missing something?

      • Anthony says:

        An other issue is that “movement along the line” isn’t necessarily linear with effort.

      • Null Hypothesis says:

        “Study doesn’t align with reality because it doesn’t consider or measure opportunity cost.” is almost always a valid explanation.

    • Paul Torek says:

      It sure would be nice if true beliefs pay. But at best that is only true most of the time, and in this case, probably not.

      • Jaskologist says:

        True beliefs pay. Religious beliefs pay in terms of greater life expectancy, better mental health, and more pro-social behavior. Therefore religious beliefs are true.

  27. dhill says:

    Yesterday I experienced growth mindset failing. I noticed that my usually slacking friend works hard on Friday evening. We had a little chat and we came to conclusion that we’re blaming ourselves too much. We would do better job just relaxing before noon and accepting that open-space environment is breaking our flow. We could do things that don’t need so much concentration and leave those mind intensive tasks for the time, when it’s quieter. Denying this reality is counterproductive, it creates anxious mind at the end of the day and makes it hard to work when the environment gets better.

    • For my equivalent … .

      A fair while back, I assigned myself two hours a day of work on my current writing projects. After spending part of today doing taxes, I didn’t feel up to anything creative. So I spent my two hours locating people online with expertise relevant to the various chapters of my current nonfiction book project (Legal Systems Very Different From Ours) and emailing them to ask if they would be willing to look over a chapter and tell me what I got wrong, or if they could recommend someone else. It was something I had been planning to do for quite a while, and didn’t require me to be in a productive mind set.

  28. Johannes says:

    The dangers of the growth mindset have been hinted at but I think they can be as grave as the ones of the fixed ability mindset.
    The simple graph with the thresholds and effect bars given a certain ability level pretty much says it all.
    The difficulty is to find out if one is A, B or C in a particular field. And the job of the teachers is to help students to find this out, not to impart A with a mindset he could compete with C (with the abilities and thresholds as in the picture), no matter what.

    It also seems to me that the anecdotal evidence suffers from bad control groups. E.g. students at elite universities might want to appear lazy geniuses but here we have both a high-ability subset and probably also one where a lot are efficient workers so they do not have to put in so much busywork. There is also the converse that the growth mindset might someone make to fake busywork to appear busy.

    In any case, especially in the education of children/teens to me it seems that background and thresholds are extremely important. One does not have to be a genius to do very well in middle school despite being a lazy bastard because the thresholds are so low. Many students will experience a shock at university because they never had to put in any effort into e.g. maths to excel and now find themselves as one rather average bloke among apparent geniuses.
    Conversely, although it is certainly worthwhile to put a lot of effort into education methods there are things, say calculus or reading Vergil in Latin, that will be hard or too hard for a considerable percentage of students, no matter how hard they work.

    And this can be extremely frustrating for students if they are told they can do everything with hard work. I remember a student who came to my highschool in 11th grade (we had 13 grades in Germany 25 years ago and one had to chose two “majors” or advanced subjects for the last two years with twice as many lessons, grades counting double or triple etc.) from a different school. In maths he had had a “B+” or so from his former school and he wanted to enter advanced maths for the last two years. He almost failed the first semester (still in the general maths which was basically introductory calculus) and barely received the passing grade to enter the advanced class at the end of the year.
    The teacher (who was not very good IMO) encouraged him hesitatingly and said he could do it, if he tried hard. I was not in this class later on but the guy always hovered between passing and failing and I am sure he worked hard. He passed the final exams barely but his grades were rotten.
    The problem here was 1) that the level at his former school was so low that he thought he was reasonably good at maths and 2) that while he was shown the limits of his ability at our harder school the teacher was afraid to discourage him from doing the advanced math class because he apparently believed in some vulgar version of growth mindset.

    • Anonymous says:

      The problem I have with your story about your classmate is that it is ignoring curricular gaps as an explanation. (When schools are “low-level” that usually means they are teaching less.) Math builds on itself. Your classmate may have had the same innate ability as you or anyone else in your classes, but unless taught the material explicit, he would struggle. Likely, so would most of your classmates. The proper response on the part of the teacher would have been figure out where the gaps were and remediate them, instead of just hoping they would go away on their own.

      • Johannes says:

        you are right generally, but he did not move there from far away, just had attended another school in a neighboring town and in principle they should have covered the same material. Additionally, our first semester of the 11th grade did have quite a bit of repetition of pre-calc stuff we had done the year before. So he had a chance to catch up. He simply was not all that bright, to the extent that I am not sure if our school where he was really challenged was rather frustrating for him or if it was still a better experience than getting As or Bs for doing well at a considerably lower level.

        Anyway, anecdotal evidence is not worth much. My point is that there are some hard thresholds and cases were people should be discouraged from doing A and encouraged to do B. And everybody, even comparably gifted students should be given a chance for hard stuff so they can find out how good they really are and at which subjects. In practice, “given a chance to try hard stuff”, usually means more or less gentle forcing.

        It all depends on the balance. Sure, students should learn to work hard. But if the average student has to put in 2 hours of homework a week for a particular class and I need 6 hours and a remedial teacher to help me, this (unless it is for catching up after missing something) is both frustrating as well as inefficient. In such a case, shouldn’t I better drop the e.g. Latin class I am not gifted for and rather do maths or French or history. This seems fairly plausible to me but flies in the face of the growth mindset. I should believe it depends only on hard work and work my ass off, probably hating Latin and missing the French or Physics I could have learned in the same time instead.

        • Two math classes in the same school can be different levels of quality by far. And two math classes by the same name may teach totally different things. For example, I’ve taught middle school algebra, high school algebra, and college algebra. All classes were called “Algebra.” But, the middle school algebra was very rigorous. Students were expected to understand concepts and how to compute, and were prepared with higher order thinking skills so they could go on to Honors and Advanced Placement math and science. High school algebra was arithmetic with some variables in it. It was remedial rote computational skills. Almost never did students go from this to Honors math or science, even if they started the same as the kids who took algebra in middle school. College Algebra is middle school algebra for kids who took algebra in 9th grade and therefore did not really learn algebra.

          Common Core Math was supposed to make all Math classes the same when they had the same name, so the kids who take it in middle school and high school get the same curriculum. But it has not played out this way.

  29. W0 says:

    Don’t mix up the two different questions:
    1. How much does effort matter vs. innate ability?
    2. What is the consequence of believing a certain answer to 1.
    It is not necessary true that believing what is true is most successful.

    • Irrelevant says:

      It is not necessarily true that believing what is true is most successful.

      I disagree. Understanding a truth includes knowing its importance. All supposed examples of truth you’d be better off “not knowing” are just truths it is being claimed are less harmful to undervalue than overvalue.

      • Jon Miller says:

        Even if you’re entirely correct about this, I still think W0 is drawing a useful distinction. His (1) and (2) are logically distinct questions, even if their answers are connected in actual fact; it is important for the purpose of discussion to acknowledge that, in principle and as a matter of logic, the answer to question (1) doesn’t entail an answer to question (2), and vice-versa.

        This is worth noting because Scott does not draw a distinction between (1) and (2), and his Bloody Obvious Position is a view about (1), while his Controversial Position is a view about (2).

  30. Having thought carefully about the matter and reflected on my own experience, I don’t believe either the Bloody Obvious Position or either Controversial position. All three seem to me to be rather (um) obviously inadequate.

    I now label and summarize the Amplification Position: Innate ability amplifies effort. To the degree you are talented at something, your effort to learn it will be rewarded with rapid acquisition of skill and insight. The more talented you are, the faster you will improve. If you are not talented at all, you will find that large amounts of effort produce zero or only tiny improvements.

    Scott’s horrible graph fails to match reality because difference in starting competence are not the main consequence of innate ability – speed of learning is. We are fooled by this because, though everyone starts from zero, very rapid learning looks like a higher baseline even from the inside.

    Predictive application of the Amplification Position is rendered more difficult by two related but separate truths. One is that IQ measures something (Spearman’s g, whatever) which will amplify effort either in combination with or as a substitute for more domain-specific talents like musical or mathematical ability. A sufficiently high IQ is difficult to distinguish from having domain-specific talent at everything; this is why “polymath” is a thing.

    The other is that fields of effort vary substantially in the extent to which talent can substitute for skill or vice-versa. Some fields (mathematics, music) require pretty substantial talent to even get out the gate. In others (writing, physics) mastery can readily be driven by either talent or skill. (On the one hand you have talent monsters like Feynman; in the other, skill monsters with notoriously poor intuition like Murray Gell-Man.) In still others, talent is very helpful early on but high-level mastery seems to be mostly effort (software engineering).

    This model is deeply rooted in my personal experience, both of my own growth and watching differently-talented peers. I have myself had the experience of being a talent monster, in particular at music – I’m one of those deeply annoying people who can approach a new instrument, take a short time to learn its mechanics, and almost instantly start making actual music because my ear is so good. I have failed at another heavily talent-intensive field where I had the talent: mathematics. I have achieved high-level mastery through effort in two other fields – software engineering and writing – mostly through continued investment in effort, though I believe talent gave me a fast takeoff in software engineering.

    I don’t think my position can be adequately summarized by either “ability mindset” or “growth mindset”. The implication is: notice what you’re talented at, then invest heavily there.

    I agree that Dweck’s phenomenon of people with an ability mindset falling out does occur – I have written about it elsewhere as The Curse of the Gifted and for a time I was stuck in it myself. But I think she is overinterpreting her evidence to reach an oversimplified monocausal conclusion. And yes, I do think the widespread uncritical embrace of that “growth mindset” is driven by a political desire to avoid taboo, uncomfortable truths about IQ and innate ability.

    But these are both only marginally relevant to my central point, which is that effort often pays off but ability ceilings are also real in heavily talent-intensive fields.You need to figure out where your talent is and invest there; you also need to have a realistic appreciation of the relative importance of talent in the field(s) you choose..

    • Emile says:

      The Amplification Model makes sense; another model I find plausible: people don’t vary as much in raw ability as they do in “what kind of things they find rewarding”, i.e. is reading fun or a chore, is partying fun or boring, though the relationship of that and ability could go both ways (we enjoy what we are good at / improve at easily; and we get better at what we do a lot because we enjoy it) making it pretty hard to measure.

    • Emile says:

      And yes, I do think the widespread uncritical embrace of that “growth mindset” is driven by a political desire to avoid taboo, uncomfortable truths about IQ and innate ability.

      Possibly, but I don’t think the direct causal effect is very strong. Other reasons for that view being popular (some are probably wrong):

      – it’s mostly true (as you said, cases of “smart therefore lazy” are not rare)

      – people care about the welfare and performance of children and therefore care about theories that might improve things

      – people who make a living writing books and giving lectures a) will tend to prefer theories that mean more books and lectures can solve problems and b) have a disproportionate influence on public opinion

      … those causes seem strong enough that “fear of uncomfortable truths about IQ” isn’t really needed as an explanation, except maybe indirectly as it may make alternative theories less popular (and even then, I haven’t seen a huge backlash against Judith Harris The Nurture Assumption).

      • Deiseach says:

        I think a lot of it is that humans like fixes. Putting in hard work and slog and getting success out is a fix. Making yourself and others smarter is (not yet) a fix.

        So a solution that says “Telling kids they’re talented is not alone the wrong way to go, it messes them up; what gets you places is putting in effort” appeals to us. In theory, everyone can always put in more effort and so there are no rungs on the ladder that are out of reach.

        Saying “You can practice all you like but if you’re short and fat with little stumpy legs you’ll never run the 100m faster than Usain Bolt” is a hard physical fact we can’t get around, so we’re forced to accept it. But saying “You’ll never get on as well as Joe Soap because your IQ is 120 and his is 140” is unpalatable to us; it offends what we think of as fairness because we can never improve our brain up to be 140 IQ merely by effort, so Joe starts off with what is an unfair advantage (or appears to be so).

        We like fixes and we like (what seems to be) fairness, so “effort counts more than talent” appeals to both instincts.

      • Stezinech says:

        those causes seem strong enough that “fear of uncomfortable truths about IQ” isn’t really needed as an explanation, except maybe indirectly as it may make alternative theories less popular”

        + 1

        I agree, it seems more about invoking wishful thinking than the mentality of taboo. People have a natural tendency to over-believe things that are optimistic, and popularizers capitalize on this. Perhaps they even half believe it themselves.

        On the Amplification Model, there’s a similar academic theory called Experience Producing Drive Theory:


  31. Jon Miller says:

    This is a great post (as always), but the distinction you raise in Section I between the Bloody Obvious Position and the Controversial Position merits further clarification.

    The Bloody Obvious Position (BOP) is a view about the relative contribution of ability and effort to outcomes. The Controversial Position (CP) is a view about how beliefs about the relative contribution of ability and effort affect other states of mind and behaviors (at least some of which in turn affect outcomes).

    BOP and CP are not logical opposites; it’s clear you understand this, because you say that Dweck accepts BOP and also accepts CP.

    However, there seems to be an inconsistency between your initial definition of CP and your later discussion of it.

    Here is your initial definition of CP: “The more important you believe innate ability to be compared to effort, the more likely you are to stop trying, to avoid challenges, to lie and cheat, to hate learning, and to be obsessed with how you appear before others.”

    Here is you later discussion of CP: “According to the Controversial Position, this athlete will still do worse than someone who believes success is 80% ability and 20% effort, who will in turn do worse than someone who believes success is 70% ability and 30% effort, all the way down to the person who believes success is 0% ability and 100% effort, who will do best of all and take the gold medal.”

    And: “The natural conclusion there is that the player who believes success is 0% innate ability and 100% effort will do the best.”

    Some thoughts:

    1. Your initial definition of CP doesn’t directly mention outcomes, so it doesn’t necessarily imply that, the more a person believes ability matters, the worse their outcome will be. Your definition of CP mentions the effect of belief in innate ability on various behaviors and attitudes, but it does not explicitly state that these behaviors and attitudes necessarily lead to worse outcomes. This is a minor problem and an easy fix, but still worth pointing out.

    2. Your initial definition of CP doesn’t say that a person who believes innate ability matters more will have a worse outcome than another person who believes innate ability matters less–which is what you assume in your later discussion of CP.

    If CP is revised in line with the first suggestion, such that it draws a direct connection between beliefs about innate ability and outcomes, it is still only a view about the connection between one person’s beliefs about innate ability and that person’s outcomes. It says nothing about how well that person will do compared to other people.

    So which is it: Is CP a view about the effects of a person’s belief about innate ability on that person’s behaviors, attitudes, and (hence) outcomes, or is it a view about the effects of a person’s belief about innate ability on how that person’s outcome ranks compared to the outcomes achieved by other people?

    3. Sometimes you focus on CP as a view about the effects of beliefs about innate ability on behaviors and attitudes, and at other times you focus on CP as a view about the effects of beliefs about innate ability on outcomes. Of course these could be and probably are connected in reality, but it’s worth noting in your discussion that these are two logically distinct positions, and a person could affirm one and not the other. Since Dweck seems to believe both, it’s reasonable to define CP so that it includes both, but it’s probably worth noting that there are other possible views out there.

  32. In the talent vs. nurture wars, people’s positions on the issue depend on the said skill. As I explain here, people are more open to the idea of nature superseding nurture if the involved skills aren’t intellectual, such as athletic ability, versus an intellectual skills such as academic achievement, in which case people want to believe practice is more important than biology.

  33. kernly says:

    A big thing that’s being missed here is that people don’t typically do things in order to get better at them. They do things because they enjoy them or because they get respect or money for doing them. Sure, if you think that doing something will eventually make you really good at that thing you’re more likely to pursue it – but it’s a much, much smaller chunk of people’s motivation than is being assumed here.

    Even if you don’t think you’re going to get much better at your job or hobby, the driving factor isn’t you getting better later, it’s the enjoyment or money you get out of your hobby or job. If you’re enjoying/being rewarded for what you’re doing, believing that you’re gonna get a lot better at it doesn’t matter that much. You might even be *less* likely to continue because you’re more likely to experience negative reinforcement in the form of unmet expectations.

  34. Furslid says:

    The growth mindset you describe really grates on me. This is because it claims that inaccurate beliefs are better than accurate beliefs. It’s not axiomatic that for the typical person having accurate beliefs is good, but I think it’s a fundamental theory and well proven in almost every area.

    I may emphasize effort to myself and others, because effort is controllable in ways that ability is not, but I can’t knowingly hold false beliefs, won’t try to reason towards specific beliefs, and won’t advocate beliefs I don’t hold to others.

    • Not a big fan of it either. Practice allows one to live up to their cognitive and physical potential, and that is it.

    • swanknasty says:

      If we know that people can dramatically improve their skills in various narrow ability domains, and most jobs involve practice in narrow ability domains, I’m not sure why people treat the discussion as settled.

  35. Jim Stone says:

    I’ve seen growth and fixed mindsets at work in others, and in myself at different times in my life. And I think people can benefit from switching from a fixed to a growth mindset.

    But I think you’ve convinced me that the mindsets are largely independent of how much one attributes performance to innate ability vs training.

    There might happen to be a correlation between innate-ability-attribution and a fixed mindset. But there’s no necessary connection. And the correlation, if it exists, seems very plausibly due to contingent cultural narratives.

  36. I work in education, primarily evaluating grant-funded programs, and helping K-12 educators move into the “data-driven” world. Right now, we see K-12 educators getting professional development in Growth Mindset. This is like the flavor of the day. It doesn’t sink in or have an impact on anything. It looks like all their other fad trainings, like Who Moved My Cheese training, etc. They are also getting some training in how to have a process for using data-driven decisions, which includes assigning someone to be your data analyst for the day, and someone else to take notes, etc. They don’t realize that using data and growth mindset are any different than getting trained on how to feel more like a leader, etc.

    Current education trends, like Growth Mindset, are being fueled by how you can spend money and what grants will fund. But this is really hard for educators because it doesn’t fit in the structure of K-12, and the changes would upset so much of the power structure.

    Historically, education money was allotted for demographic groups. We had the war on poverty, and Title 1 funds are to spend on poor kids. Lots of federal grants were for certain demographic groups. I have been evaluating these for 20+ years and have witnessed the changes. We do independent evaluations for accountability. In the 1990s, we would be asked to do document that a grant for poor kids did indeed serve poor kids. Teachers would get their training on how to empathize about being poor so they would love the children more. There were no goals or objectives other than to serve the poor or whatever group the grant was for. Products were marketed as being great for poor kids, or minority kids, or kids whose parents didn’t go to college, or girls, etc. Lots of trainings were to help teachers have certain kinds of feelings about these groups.

    Then, in the early 2000s, funds were still given to serve demographic groups but our accountability reports would be required to include post-academic data. No pre data were required because it was assumed that we know the academic status of each demographic group. (I know this doesn’t make sense, but this is how it was.) In this era, the products marketed made some academic claims. They were not simply to make poor kids feel better, but they’d feel better and maybe some day know how to read or do math. Very vague claims but with an academic component to them. For example, teachers could get certified by some company to teach an elective course that would teach poor kids, black kids, and kids whose parents did not go to college how to organize their notebook and take notes, with the hope that this would lead to some day being able to learn algebra (with the assumption that those groups cannot learn algebra now, but we might be able to break the curse with this certified notebook organizing training.) We’ve evaluated hundreds of these programs.

    Then around 2005 or so, the funds were still allocated for demographic groups but the accountability reports called for reporting pre vs. post data and having objectives. This was the crazy era. Many of our clients got mad at us because they thought our reports made no sense. Schools would make an objective of bringing students in a program to grade level in reading, serve all the poor and Black kids, then call us to analyze pre vs. post reading data for the kids. We would find that most of the kids served were already at grade level prior to services, and those kids usually lost ground when pulled from the core courses to receive remedial instruction. (This was happening before but no one ever compared pre vs. post data so they could imagine something else was happening.)

    This is when we started doing our Data Academies to teach them how to operate in the world of data. We had to create visual icons to communicate concepts, including the mathematical concepts. For example, educators we worked with thought poor kids being “at-risk” meant that even if they could read, they might forget how any day now. We used haystacks with needles in them to illustrate that we call them “at-risk” academically because if we divide the big haystack of all kids into the poor and non-poor, and needles represent non-readers, there is a greater concentration of needles in the poor haystack. A non-reader in either stack needs help learning to read. A reader in either stack does not need to be pulled from core curriculum to be given remedial work. This was shocking to most of them. They were so used to delivering services based on demographic haystack. We tried to teach them that using data to align services meant looking for those needles, which we can easily do now that we have computers and reading scores.

    We started using some of Dwecks’ Growth Mindset ideas in the “research nuggets” we would include in our Data Academy trainings. This was to introduce the idea that kids are not a certain kind of kid, and that you sort them by whatever attributes you sort by, and then put them on some academic track that you think is right for that kind of person.

    We got a lot of push back. There was huge push back at putting poor or minority kids with very high math achievement scores into advanced math classes. We saw many upper class white kids labeled as “academically gifted” in math, yet their math achievement scores were well below those of some poor and black kids who were receiving remedial work. Some educators expressed something like the fixed mindset, that these gifted kids are inherently better at math, but it is going to come out later. And later these high scoring poor and minority kids will hit a wall and won’t be able to go past it. So, they left everyone where they were.

    Some principals moved kids based on what the data told them. And their achievement gaps between poor and non-poor, white and minority closed in one year. One principal went from one advanced math class in the school to 7.

    We saw principals moved out of schools for allowing high scoring poor and minority students into advanced classes. The social structure of schools is that upper class people get the better more advanced classes, as if they are the most academically able. Using data to provide opportunities is not necessarily the growth mindset but it kills the fixed mindset.

    There has recently been a switch in education again. Nothing changes all at once. The trend now is the next logical step after seeing the pre-post comparisons from educational programs in the 2006-12 or so era, where kids got served based on demographic characteristics and then the accountability reports showed services were not aligned to their need.

    Now, grants are funding moving kids forward from where they are, and rather than sort on to different learning tracks, all kids are to learn the same thing. In the past we taught as if not everyone can learn a subject, especially in math and science. (I can explain where this idea came from. I may at the end.)

    To have an educational program where you give a student a lesson based on what he already has mastered, with the goal of getting him to a certain place, you need to understand data, know what lessons are for (e.g., for teaching how to add vs. for poor people), and you have to believe the kid can learn.

    To support these new educational goals, educators need professional development. In the past, their professional development did not give them any skills and they did not learn to do anything. They came away having been exposed to an idea or attitude. They know they need training on how to use data. Believe me, they don’t know how and school data sets are often difficult to use. But, the people who used to sell “how to feel about poor people” training are now selling “how to have a process for using data” training. Not “how to use data” but how to have a process. Like, “Decide when to meet to talk about it; Decide what rules you’ll have for the meetings; Choose roles. Assign someone as the data analyst for the day…” And, they are getting Growth Mindset training.

    Common Core Math is one rigorous curriculum for all kids. It is not anything new. The changed the order of some things, to make more sense. It is simply what the top track kids always got–conceptual understanding along with the skills for computing. But only a few teachers have experience teaching this. You can’t teach conceptual understanding if you don’t have conceptual understanding. So, they are finding ways to make remedial Common Core Math classes and kids with no advocates get enrolled in these. They used to be able to say that our professional opinion is that they wouldn’t be able to learn the rigorous math. Now, the data shows they would be able to and many already have. They go to their Growth Mindset training, then to their “how to have a meeting about data” training, and then track the kids into remedial versions of what is supposed to be rigorous only classes.

    They expect that using data, and thinking any kid can learn are fads that like everything else they’ve been trained on, will soon be replaced by the next fad.

    Here is why we think only some kids can learn math and science and that teachers can spot them:

    1950s: We needed rocket boys to race the Russians to the moon.
    Wealthy white people wanted their kids to be doctors, lawyers, and businessmen.
    Teachers identified smart lower to middle income white males (because culturally, that is who could work at NASA) to learn advanced math. Black people were trying to get a seat in the front of the bus, not in an algebra class.

    1960s and 70s we integrated the schools. This is when we started the concept of Academically Gifted, testing in 3rd grade and then labeling the kid forever as “gifted.” This is very fixed mindset. We also labeled many many black males as “academically disabled” in a fixed mindset way. The labels came in elementary school and the kid kept the label forever, and got assigned to different learning paths based on the label. The wealthy didn’t really care about math. They still wanted to be doctors, lawyers, and businessmen. Poor white people, both male and female by the 70s, would be identified by the professional eye of the teacher as someone who could learn math. They were looking for rocket boys and girls (fortunately for me). Kids who could learn math were viewed as socially strange. It was socially strange because it was someone other than the wealthy in advanced courses.

    Then, the tech boom happened and engineers make a lot of money. So, the wealthy want the rigorous math and science classes for themselves. There are not enough teachers for everyone to have them. Keeping the idea that teachers can spot those with potential to learn rigorous math and science, they now are hand selecting wealthy white kids (because their parents can make that happen.) But, everyone pretends like we are still just hand selecting the most potentially talented, like we’ve always done. This method requires the fixed mindset.

    In an attempt to break out of this fixed mindset, educators are getting Growth Mindset training. The changes are being driven by how money is allocated, and what federal education grants will fund.

    • Fascinating account. If anything could make me more in favor of a system of school vouchers, where school incentives are driven by what parents want for their own kids … .

      • Besserwisser says:

        Are you supposing that as a solution? From what I’ve gathered, the problem was that rich people were trying to get their kids into courses they wanted them in in the first place. Seems kinda counter-productive to advocate for a system where they can do this more effectively.

        • ADifferentAnonymous says:

          Rich kids learning math isn’t the problem. Poor kids not learning math is the problem. Given a fixed supply of math education, the one implies the other. The market-based solution aims to increase the supply of math education.

          • Besserwisser says:

            Doesn’t this mean people would have to individually pay for their education? This doesn’t seem to help particularly poor people. And even if you somehow create more opportunities for studying math, which I somehow doubt would work the way you want it to, there will still be somewhat of a bottleneck in terms of demand for math-based jobs. I don’t see how a “market-based solution” wouldn’t still favor rich kids.

    • John says:

      I apologize for releasing my mom onto your blog, Scott.

      The haystack analogy is good, though. The at-risk model was absurd. If there’s 100 rich whites and 100 poor blacks, then probably 30 of the whites need remedial help and 35 of the blacks need remedial help. But all the teachers and principles heard was “poor blacks are at risk of dropping out”, so all of their money in programs targetted poor blacks. my mom spent decades trying to make everybody understand that, rather than pulling all the poors and blacks into remediation because they’re slightly more “at risk” (30% vs 35%), you can use data to figure out which kids actually need the help. That is, you can take the 30 whites and 35 blacks, rather than all the blacks and none of the whites.

      They didn’t understand, of course. They’d say their program targetted kids likely to drop out, but when asked how they determined who was at risk, they said they used the list of students on the “reduced-cost lunch” list, or they’d use the bus route info to see which kids were in poor neighborhoods. And we’d say, “no, you’re supposed to target kids who are failing their classes, not kids who are demographically slightly more at risk of failing their classes”. And they wouldn’t understand. “But the research shows that poors and blacks ARE at risk for failing their classes!”

      It was only when we actually sat down and drew pictures of hay stacks, one HUGE haystack labeled “white” with about 30% needles, and one tiny haystack labelled “black” with 35% needles, that some of them understood. But not many, not nearly enough to actually make a difference. Most simply do not understand percents (I’m sure mom [Janet, above] can tell of some of the hilariously depressing surveys we gave teachers with questions like “Do you know what a percent is?” where basically none of them did. How are you supposed to interpret “30% of whites need help, 35% of blacks need help” any way OTHER than “blacks need all the help” if you don’t understand percents, after all?).

      so if there is an upside to the “growth mindset” fad in education, it’s that teachers might have a reason to stop this compassionate racism, without first having to do the impossible and understand exactly what 30% and 35% means.

      • Emile says:

        Great comments, the two of you – reminds me of an old post on LessWrong: http://lesswrong.com/lw/hbk/problems_in_education/

        What I mostly take from this is that okay, the system has a lot of problems, but things do seem to be improving!

        • I would agree with your conclusion. Things were messed up and we couldn’t tell. Now we are going through a phase where we can tell what is wrong and how to fix it. Change hurts, so there are a lot of people feeling confused, angry, and frustrated. But things are moving in a much better direction. And funding is driving the direction of the changes. People are changing even if they don’t want to because they want the grant funds.

          Educators are getting Growth Mindset training, but what they are really trying to do is move away from the Fixed mindset. That does not necessarily mean you have to go to Growth. We coined the term Pro-Equity several years ago to replace the At-Risk mindset. By this we meant, take the kids where they are based on tests of content mastery and move them forward from there. This is to replace pulling poor or minority students from core instruction to get remediation, or not helping non-poor students who need it. This is different than Growth Mindset. But in both cases, you have to move away from the Fixed mindset.

          • Decius says:

            What’s worse is that the fixed-mindset of the educators is being passed down to the students in both gifted and remedial tracks.

            The remedial students learn that they will never be good at math, and they aren’t.

            But the gifted students learn that they will always be good at math, and often fail to practice and thrive.

            The result is that we have much fewer people who are really good at math.

          • ryan says:

            It doesn’t sound to me like you and ma are using fixed and growth mindset in the same way Scott is.

            You two seem to be saying “when dividing children into classrooms, group them by their achievement level.” If a kid has algebra all figured out, put them in a calculus class. If they don’t understand algebra, put them in an algebra class.

            That seems the obviously correct thing to do, but also a wholly separate issue from the psychological effects of telling kids your smart or work harder.

            Regardless, interesting posts, I enjoyed reading them.

      • Douglas Knight says:

        Where do 30% and 35% come from? Did you make them up because they lead to the conclusion you want?

        • John says:

          Yeah those are made up. The point is: 1) there are a lot more whites than blacks 2) blacks perform worse academically, but the effect is pretty small (35% versus 30% as a ball park example) 4) teachers and counselors target 100% of poors and blacks because they hear they’re “at a higher risk of failing”, even though the absolute number of struggling whites is huge compared to the absolute number of struggling blacks (simply because there are more whites) 5) they don’t understand why this is bad because they have absolutely no clue what 30% or 35% mean.

          Teachers who think that “targetting students who struggle with math” means “targetting students of low socioeconomic status” simply cannot understand that the latter is a *proxy* for the former, and only as good of a proxy as the distinction in achievement between blacks and whites is large. Since we’re talking about a single digit percent difference, it’s not a very good proxy at all, but the school system treats them as absolutely interchangeable.

          • When we work with schools, we use their actual data. Depending on the school, the percent differences my be big or small. But the point is that now that we have good data on who has mastered what academic skills, it is possible to provide interventions and opportunities based on the academic data.

          • Douglas Knight says:

            but the effect is pretty small (35% versus 30% as a ball park example)

            It’s perfectly fine to make up ballpark numbers around what you believe. The problem is that your beliefs are amazingly wrong. You talk about data, so why don’t you look at some data?

          • John says:


            The actual numbers aren’t the point. The point is if there are 1000 whites and 100 blacks, 10% of whites are failing and 20% of blacks are failing, then targetting 100% of blacks for intervention will miss 5/6ths of the failing students, while also screwing over the 80% of blacks who got pulled out of their regular class for remediation they didn’t need.

            Most of the people who decide what heuristics should be used for targeting struggling students didn’t understand this.

            What beliefs of mine are false? I work with K12 school data for a living; this is my job.

          • William Newman says:

            For an example of widely-available real-world numbers: a Google search for ‘us sat scores racial breakdown’ gave me http://media.collegeboard.com/digitalServices/pdf/research/SAT-Percentile-Ranks-By-Gender-Ethnicity-2013.pdf .

            (These numbers are for college-bound students, which would not be my first choice for perfectly representative numbers, but I don’t happen to know where to get broader numbers, and I think these numbers at least suffice to illustrate why people are griping about John’s ideas about representative figures.)

            From those numbers (for 2013 college-bound seniors), the cutoff numbers giving proportions of the kind that John has in mind seem to be around 400 or 450. For all six numbers (400 and 450 numbers on the three different subtests, Critical Reading, Math, and Writing) the percentage of blacks below the cutoff is more than twice the percentage of whites below the cutoff; sometimes it is four times the percentage. For example, 8% of whites score below 400 on Math, while 37% of blacks do, and 9% of whites score below 400 on Critical Reading, while 37% of blacks do.

            If instead we want to choose remedial cutoffs so that not only do at least 35% of black test takers fall below the cutoff, but at least 30% of the white test takers also fall below the cutoff, the next highest score tabulated, 500, does the job. At that level, white failure percentages are 39, 36, and 45 for the three subtests, and the corresponding black failure percentages are 76, 76, and 81.

          • As someone has observed, John doesn’t know what he’s talking about.
            The achievement gap is so huge that percentage-wise, more poor whites are proficient than non-poor blacks.

            So if you’re going to spitball with made up numbers, it’s not “10% of whites are failing and 20% of blacks”, but rather “70% of blacks are failing and 10% of whites are failing and the average black failing score is far, far below the average white failing score.”

            Teachers do not, in fact, think that black = low ability or poor= low ability. What they know, based on test scores, is that black of any income means a higher probability of low ability, and low income blacks means a spectacularly high probability of low ability.

          • John says:

            I did NOT mean to imply that the gap is only 5%. It’s huge, by the end of high school. In some counties, the graduation rate gap is >50%.

            My goal was not to imply the gap is small. When I say I used ‘30%’ and ‘35%’ as examples, I meant examples to demonstrate the base rate error being made, not examples of what the achievement gap was (although, in some elementary schools, it is as low as 5%).

            The point I was trying to make was that, if you have an intervention designed to help people failing 8th grade algebra, frequently we would see the project leader or, in cases where kids are referred to the project, teachers use reduced-cost luncg lists and bus routes to determine who to put in the project, when THE GODDAMN GRADEBOOK IS RIGHT THERE AND YOU CAN JUST LOOK UP WHO IS FAILING ALGEBRA, without using “socioeconomic standing” as a PROXY for “failing algebra”.

            The specific problem they are making in assuming poor/minority = fsil algrbra is ignoring the base rate. When I came up with the example to illustrate this, my mental algorithm was to generate a random-sounding number between 1 and 100, a number which did not and was not supposed to correlate with my beliefs about the achievement gap (and I am sorry if it came across that way), then add 5, since base rate failures are made more obvious when the differences between the groups are small.

            let me make this absolutely clear: while the achievement gap varies between elementary school and high school, between rural and suburban and urban, between the South in the north, between the East Coast and West Coast, it is generally fucking huge, more like 35%-40% than 5%.

          • John says:

            I agree that, presently, even if these interventions targetted the correct students rather than selecting them from the pool of poor students, the achievement gap would still exist. (although you would be amazed at the success when schools start using bayesian methods to predict success in 8th grade algebra, and then letting the algorithm do the placement instead of the teachers or counselors. One middle school reduced the racial disparity in their 8th grade algebra vs prealgebra seats to within a couple percent of being proportional to the population’s demographics.)

            But that doesn’t mean this particular issue isn’t huge. If you’re a smart, but poor, black kid, you have a ridiculously high chance of, at *some* point, getting sidelined into one of these dropout prevention, academic “enrichment”, etc programs, and once you’re getting pulled out of class to attend a remedial course for kids who “struggle with reading” or “need to learn better study skills”, your chance of getting identified as “at risk” in the *future* skyrockets, your chance of being put in the top-level classes plummets, and your outcomes come crashing down.

            This doesn’t happen a *lot*, but it doesn’t have to happen a lot to have a *huge* affect on the data. That’s why I think the elementary school data is more illustrative of any actual aptitude gap, and the rest is just this kind of snowballing process by which the reaction of the system to that gap amplifies it to the ridiculously huge gap at the end of highschool.

          • Emile says:

            John: I think everybody understood your original comments and the illustrative nature of it’s numbers, it’s just that some people have an axe to grind and are looking for excuses to bring their favourite topics in the limelight.

            (it’s a somewhat interesting topic too, but not worth hijacking an interesting discussion with nitpicking…)

          • Alexander Stanislaw says:


            From John’s initial comment:

            The haystack analogy is good, though. The at-risk model was absurd. If there’s 100 rich whites and 100 poor blacks, then probably 30 of the whites need remedial help and 35 of the blacks need remedial help.

            There is nothing in there about “suppose the values are X”. I’m willing to accept John’s acknowledgement that his numbers did not reflect reality but it is still a worthwhile thought experiment, but I am not willing to accept your assertion that pointing out this error is an instance of unjustified axe grinding.

          • David Allen says:

            To those criticizing John’s numbers (arbitrary as they are),
            you all seem to be conflating intra-school demographics with more general statistics. I’m willing to bet the actual intra-school numbers are far closer than the ones you propose.

          • Douglas Knight says:

            David, what do you think is closer to the truth, the numbers that first popped into John’s head, or the numbers he produced after censoring himself?

          • Conspirator says:

            the rest is just this kind of snowballing process by which the reaction of the system to that gap amplifies it to the ridiculously huge gap at the end of highschool.

            If true, I suppose this interpretation would countervail against the study finding Scott cites here that unschooled kids only perform one grade level below their schooled counterparts. (Well, I suppose school could be substantially counterproductive for these incorrectly tracked kids… that’s actually a bit scary.)

          • rttf says:

            @William Newman

            Your numbers are worthless. If black kids have been put in remedial classes since elementary school, then of course they’re going to do worse when they’ve finished high school. This is true independent on whether putting them into remedial classes in the first place was the right thing to do.

        • John says:

          accidentally hit submit too early.

          well I don’t exactly agree with some of the claims of the growth mindset, at least teachers who fall in line with the new fad will have a reason to no longer make this mistake, without having to learn about percents.

        • Steve Sailer says:

          “Where do 30% and 35% come from?”

          Right, The Gap that so obsesses Americans really is a lot bigger than that. There’s a good reason everybody is so worked up and touchy over it and has a hard time reasoning calmly about it: it’s pretty big. Not huge, but pretty big. It’s more like, to make up percentiles one standard deviation apart, 16% and 50%.

          • Irrelevant says:

            Note that in a representative school a 16% vs. 50% split would still end up with twice as many white remedial students as black.

          • PGD says:

            I think I’m going to believe the people who actually work with school data every day over Steve Sailer on black-white test differentials.

            Note that the ‘gap’ is endogenous to the kind of practices Janet and John are describing — a gap that expands greatly over the course of schooling, from being fairly narrow in elementary school to very big in HS — is exactly what one would expect to be caused by the practices they are describing.

          • John says:


            PDG, I did not mean to suggest that 30% vs 35% is anything close to what you would see in the real world. The first numbers that popped into my head were “30% and 60%”, but I wanted a stronger example of the base rate mistake.

            The actual gap is much larger than 5% in most instances. (I assume this is what Knight was complaining about as well?)

          • John says:

            @PGD I posted a clarification above; the numbers are much worse than 30% vs 35%. Those numbers were selected to illustrate the problem of ignoring the base rate, not to illustrate what a real gap might look like at a real school.

            depending on what metric you are using (graduation rate, 8th grade algebra, algebra, core curriculum grades, etc) the gap is much larger than 5%, usually more like 20% but in some places as high as 50%.

          • JK says:

            I think I’m going to believe the people who actually work with school data every day over Steve Sailer on black-white test differentials.

            John’s corrected numbers are actually very similar to what Sailer suggested.

            Note that the ‘gap’ is endogenous to the kind of practices Janet and John are describing — a gap that expands greatly over the course of schooling, from being fairly narrow in elementary school to very big in HS — is exactly what one would expect to be caused by the practices they are describing.

            The black-white test score gap is pretty constant from age 3 to age 13 (and beyond). See this meta-analysis.

      • Anonymous says:

        so if there is an upside to the “growth mindset” fad in education, it’s that teachers might have a reason to stop this compassionate racism, without first having to do the impossible and understand exactly what 30% and 35% means.

        So you’re telling me that the people teaching middle school math don’t understand middle school math? That’s concerning.

        • Sorry to upset you. Some of them know middle school math. And those are the ones that teach the top track kids.

          • Shieldfoss says:

            Sorry to upset you. Some of them know middle school math. And those are the ones that teach the top track kids.

            America is weird.

            Is this a funding issue? I realize that basically all problems can be reduced to funding issues if you’re willing to throw sufficient zeroes after the “$1” in front of the check but is there some special issue that needs attention or is it just insufficient money to attract high-skill teachers?

          • Jaskologist says:

            According to my dear aged mother, herself a public school teacher, this is the result of woman’s liberation. In the past, teaching was one of the higher status career paths a woman could take. Now, a smart woman has many other options, which often pay better. Consequently, the girls who go into teaching may be nice, but they are substantially dumber.

          • Alex R says:


            My guess is that it is in part a funding issue (teacher salaries in the US are absurdly low), in part presence/absence of alternatives (you can get a job teaching math without knowing math because there aren’t enough teachers; if you know math, you can get a better, different job), and in part bullshit intolerance (public schools in the US have a special kind and degree of bullshit that teachers have to put up with).

          • mobile says:

            @Jaskologist: Someone took a look at your mother’s hypothesis. Conclusion: average teacher quality is about the same, the variance is lower now.

          • Jaskologist says:

            Interesting. I can’t decide if that’s a confirmation or not. She’s in a good school district, so the net effect of “fewer smart teachers” probably would be a negative impact on teacher IQ there.

    • Null Hypothesis says:

      That was very engaging to read.

      Two questions if you’d indulge them. First is, to what degree did “No Child Left Behind.” (and the mandated testing/funding incentives and all of that stuff that went with it) help (or inhibit) the recognition and/or acceptance of the general uselessness of those fluff flavor-of-the-month teaching methods, and the adoption of more data-based ones?

      The Second Question is more of asking for a counter-opinion from someone qualified to give it – because I’ve had no chance to express it to someone like you, and opinions without pushback aren’t terribly useful. You mentioned the Common-Core Math as being the same advanced thing that is now being given to everyone.

      In my opinion, looking at examples of the curriculum, I ‘get’ what they’re trying to do, but I also thing its a horribly thought-through process. It seems like they’re trying to teach a bunch of different random weird ways to approach and solve math problems. And they do this because kids that are good at math are always finding shortcuts and identifying and exploiting relationships and analogous situations to find answers quicker.

      To me it seems like they have causality backwards – that kids will be good and intuitive at math by being exposed to a bunch of tricks. Rather, the kids that are smart at math will see said relationships and find the tricks themselves. Hell, I’d actually get frustrated if I was being taught from the C.C. curriculum, because when I was in school I could get a problem, and solve it with whatever cool, intuitive method I had figured out. But now I’d have teachers telling me: “No no, you’re doing it wrong. You have to solve it with this intuitive method we’ve taught you.

      But more importantly, I’ve talked with several young kids about some of their homework sets, who struggle with them, and it seems to me that the whole process is a great disservice to them, because rather than focusing on a prescriptive, reliable, fundamental approach to the problems, they spend a bunch of time ‘teaching kids to be intuitive’, and all that happens is the non-intuitive kids don’t see the connections, and get really frustrated at these stupid random math processes they’re being taught.

      And considering that a lot of kids already considered math indecipherable, this seems like it only exacerbates the problem. It seems like its following the logic of “middle class people live in homes. Therefore, if we artificially push poor people into homes, they’ll become middle class.” And I’m very, very worried about the aftermath being equally catastrophic – a generation of non-math-intuitive kids that are still not intuitive, but also have a much more tenuous grasp of the fundamentals and much more animosity towards the subject.

      I’d be very interested to hear your opinions on it. For society’s sake, I’d love to be told I’m wrong.

      • Irrelevant says:

        To me it seems like they have causality backwards – that kids will be good and intuitive at math by being exposed to a bunch of tricks. Rather, the kids that are smart at math will see said relationships and find the tricks themselves.

        I’ll second this from a tutor’s perspective: it’s very possible to introduce “tricks” to students, and I consider that one of the major things I’m being paid for. But the point of introducing tricks is to provoke a realization on their part. If the kid’s reaction to having the trick explained isn’t “oh wow, that makes so much sense now” then I move on to find a different explanation or teach them the slow way. The Common Core sheets seem to instead want to spend a couple hours belaboring each trick in hopes it’ll take in unprepared soil, and everything I know about math instruction says that’s hellishly inefficient.

        Of course, I also don’t see reason to suspect CC will be any worse than the previous system.

        • Null Hypothesis says:

          More importantly, I think they’ve even gotten the concept of these ‘tricks’ wrong. The vast majority of tricks I used in elementary and middle school were… well… for elementary and middle-school problems. I’d see a pattern, use that trick to do the two or three problems it applies to, then forget it and never look back. All in the span of 2 minutes.

          If you asked me how I did it so quickly, I could tell you, and then you could write it down and devote a half-hour to teaching everyone in the class this random, weird, poorly-generalization pattern or method to solve a problem. But that’s missing the whole point.

          Clearly we should be teaching kids how to forget tricks right after learning them.

          • Irrelevant says:

            Less convinced on that, since most of the tricks I’ve seen them attempt to teach look like they’re general enough to be worth knowing. I haven’t been exposed to a large selection of CC worksheets yet, but since the ones I’ve seen online appear cherry-picked for being terrible rather than being good, the rest presumably are in the same vein and likely less obtuse.

            Sidenote: I was a plague on my own math instructors, because I didn’t think I understood anything if I couldn’t do it in my head, and therefore refused to write anything out unless it involved long division. And long division was obviously nonsense.

          • Null Hypothesis says:

            That kind of selection-bias of bad CC problems is precisely why I asked here. I’ve seen the same horrible ones (and, to be frank, some not-so-horrible ones people called horrible).

            However, several of the ones I’ve seen were not online, but in-person from kids. So for some I’ve seen, there isn’t the “CC is pure evil” cherry picking going on, but then again, they were problems they were having trouble with, so still some selection bias there.

            My main concern with the ones I’ve seen is that they take a 3 step problem and turn it into an 8 step problem with pictures and unnecessary diagrams. And I can’t see how you’ll do better teaching a kid an 8-step process when he’s struggling with the 3-step one. Or conversely, why you’d bother teaching an 8 step process to someone whose mastered the 3-step.

            So that’s my general contention with the system from the limited and likely cherry-picked worksheets I’ve seen. If anyone here is in some sort of position where they get to see a full survey of the material and methods, your insight would be greatly appreciated.

      • Anonymous says:

        First is, to what degree did “No Child Left Behind” . . . help (or inhibit) the recognition and/or acceptance of the general uselessness of those fluff flavor-of-the-month teaching methods, and the adoption of more data-based ones?

        It neither helped nor inhibited. Those useless, flavor-of-the-month methods may have used NCLB to justify their existence, but they were rolling on their own steam regardless. And, since those methods DO sell themselves as being “research-based,” the adoption of more “data-based” methods isn’t going to happen. After all, administrators don’t know the difference, and teachers are either on-board (if elementary) or forced to go along with the admin’s decisions (if middle/high school, and they know better).

      • Anonymous says:

        @Null Hypothesis
        In my opinion, looking at examples of the [C.C.] curriculum, I ‘get’ what they’re trying to do, but I also thing its a horribly thought-through process.

        First off, to be pedantic, C.C. is not a curriculum, it is a set of standards. Different publishers are free to create different C.C.-aligned curricula, and these are probably what you’re talking about. But.

        Believe it or not, C.C. math was an attempt to get away from constructivist, no-one-true-method, fuzzy math teaching and return to more traditional methods. I think a lot of publishers instead 1) latched on to the one sentence about students solving problems in “at least two ways” and 2) slap C.C. alignments onto whatever methods they were going to push anyway. Who’s going to stop them? Teachers/admins are already on the bandwagon.

        Anyhow. When you say C.C. curriculum, I think that ^ is what you mean.

        • C.C. math was an attempt to get away from constructivist, no-one-true-method, fuzzy math teaching and return to more traditional methods.

          That’s nonsense. It’s clearly dog-whistling reform math. And it’s also absurd to say that the Common Core isn’t specifying curriculum, when it gives pretty clear instructions on how to teach.

      • NCLB was an attempt to create an infrastructure so that we could determine what is going on. It was rolled out over many years. Step 1 was that states had to list objectives for each subject and each grade. State’s rights in education gave them the right to set the curriculum. But, to get federal funds, they had to decide what they wanted taught in each grade and document it. 12 states already did this. I am in NC, and NC was one of 12. I used to teach in Illinois. They were not. All the little districts decided what would be in their 3rd grade math, for example. It took a couple years for states to standardize their lists of what would be covered in each course and grade.

        Step 2 was that states then had to start using standardized tests that measured the degree to which students had mastered the objectives you set in step 1. Before this, schools gave standardized tests, but they were usually “norm referenced” instead of “criterion referenced.” The difference is that the norm referenced ranks the kids on things like aptitude, (thinking skills) and tells you the percentile they fall into. Criterion referenced tests content mastery. States had to switch to standardized tests that measure whether kids learned what they tried to teach them or not. States had to set a bar for identifying “proficient.”

        You heard a lot of complaining at this point because they changed from everyone teaching whatever they wanted, and standardized tests measuring how smart kids are to the entire state has to cover specific objectives in each class/grade, and they had to test whether kids learned what you were supposed to be teaching them. (in reading and math at certain grades)

        The next step is where the not leaving kids behind comes in. There were 10 or so demographic groups that the national data showed learned less in school than the other demographic groups. These were low-income, certain races, students with disabilities, don’t speak English, etc. Schools had to report outcomes for those groups (percent proficient).

        They had to make plans for raising the achievement of these subgroups, and measure their progress toward the goals you set.

        This may sound rational, but as always, people try to make a financial killing on educational reform and new way vs. old gets people confused. If a Title 1 (low income) school failed to make the progress toward the goal they set for closing the achievement gaps for consecutive years, then they had federal sanctions come in. One of these called for using federal Title 1 funds to tutor all low-income kids in the school with outsiders. Tutoring firms were growing out of the woodwork.

        What didn’t make sense of this sanction was that if any of those subgroups did not make Adequate Yearly Progress, you had to tutor the low-income kids. You had to offer tutoring to low-income kids whether they needed it or not. (This was the old at-risk model overlapping the new data-driven model, and the cat has a mouse head…)

        Here is an evaluation of such a tutoring program. http://webarchive.wcpss.net/results/reports/2006/0609ses_hodge.pdf

        I don’t make this stuff up.

        NCLB gave us the infrastructure for being able to measure effectiveness of the fad of the day. They also started a clearninghouse (What Works Clearinghouse) where programs are reviewed for whether there is evidence that they are effective. It is getting better, but nearly every program I’ve looked up there reports something like, “165 research studies have been done to determine the effectiveness of this program. Two of them were valid, and these showed no effect.”

        NCLB is over now. Because of this infrastructure being put in place, they can tie certain things to getting educational grant funds. They are starting to require that you know the reading scores of kids that you put in a grant-funded reading program. They couldn’t require that until this infrastructure was in place.

      • When NCLB phase of education reform was over, they started the next phase. For the first time in my career in education, it seems as if someone actually has a long-term plan, rather than just flavors of the day.

        They are using money to drive the reforms. And lots of it. They gave out Race to the Top grants to push things forward. In NCLB, a state had to set objectives for what would be taught in a grade/class then test mastery of that. To get a Race to the Top grant (given to states) they had to agree that they would use the Common Core Math and Language Arts curriculum, use a national test that measured learning (not what your state makes), and agree to a bunch of other things.

        Common Core is a list of objectives. It is not curriculum. It says what gets taught when. I don’t know if they realized how severely we track, especially in math. We have always had one set of state objectives for a math class, but might have had 3 different books for standard, remedial, and advanced. We gave one standardized test for, say, algebra, but kids who took it in 8th grade actually covered all the objectives on that test, while kids who took in 9th grade got a watered version of the lessons, often with an inexperienced or out-of-field teacher. Huge percentages of the kids who took the standardized algebra test in 9th grade failed it. But, ..whatever…

        One of the things you had to agree to for a Race to the Top grant is to measure the effectiveness of a school and of teachers by whether kids can demonstrate that they have mastered the objectives. Now we can’t just say, “whatever…” when the 9th grade algebra kids have low pass rates.

        The lessons you see on the web to show how stupid Common Core is are lessons made by people who are teaching math and have always done whatever they wanted in their class, but now they need to try to follow the curriculum objectives because they will be evaluated on whether their kids learn them.

        Common Core objectives might say to illustrate the concept before teaching the rote skill. Many people don’t know that there are concepts in math. They just know how to compute. These teachers may be shown an example of how to illustrate a concept and they think it is a new way to compute.

        I have seen lessons from my friends’ kids classes that say the new way to multiply is to draw a rectangle and divide it up, and count squares and they are to no longer use “the old fashioned way.” The teachers are crying because they feel like it is crazy, which this is. But it is not Common Core. It is lessons being made by someone who doesn’t know what they are doing.

        In NC, as in many states that got Race to the Top money, it was not a textbook adoption year when we got the money. So, we had to switch to Common Core without getting books. The teachers who normally just go through the book and cover whatever… had to start making their own lessons without the help of a book. This is why we are seeing all these crazy lessons.

        Let me tie this back to Growth vs Fixed mindset. In the way back past, before NCLB, teachers taught pretty much what they wanted. Kids were tested so we could rank them as smartest to dumbest. (Fixed Mindset. Kids didn’t learn much because the are stupid.) Now, we have moved to having specific things to teach them and we are measuring the degree to which kids learn (Growth).

        The evaluations of teachers and schools are using growth, not absolute achievement scores. In NC, they use predicted vs. observed growth. A kid who hardly learns anything by 5th grade will be predicted to also hardly learn anything in 5th grade. If he learns more than predicted but is still not proficient, that is a high mark for the teacher and school. This is why “growth” is in the foreground now in education. NCLB was about proficiency. Now we are focused on growth. The super way ahead kids who came in to a class way above proficient have to show growth from where they are or it goes against the teacher and the school.

        • Emile says:

          Thanks, with all that context NCLB makes a lot more sense than what I had heard previously.

        • Null Hypothesis says:

          I really appreciate the long response. That really cleared up a lot of the minutia for me.

          I will say, that while I understand that Common Core itself is not proscribing things like those ass-backwards math worksheets (very good and important to know – thanks), I can’t help but feel the program itself is culpable.

          For instance, the race-to-the-top program is putting a big, giant pile of money on the table to encourage states to conform. But a state can’t NOT take the money, just like they can’t NOT take highway funds. Because their citizens are still going to be paying taxes to cover that huge chunk of money.

          Incidentally, I’m from Washington State, where they spent about $300 million revamping our schools’ curriculum to match C.C…. and then the applications for the grants through the Race-to-the-top program were denied. And I recall that the C.C. standards were heavily criticized by the only mathematician on the board that created the standards in the first place. So we essentially spent tons of money to adjust our math standards to an equally poor standard in hopes of getting back money the Federal government had-or-will-some-day tax away. And in the end, we didn’t get the money.

          Specifics and efficacy of the programs aside though, I really appreciate understanding the timeline and how the whole thing was structured. Next step would be knowing what to do about it, but just knowing how we got here in the first place is a great start.

    • Paul Torek says:

      A non-reader in either stack needs help learning to read. A reader in either stack does not need to be pulled from core curriculum to be given remedial work. This was shocking to most of them.

      Thank you for providing us with so much information – even though lots of it just makes me want to weep.

      We saw many upper class white kids labeled as “academically gifted” in math, yet their math achievement scores were well below those of some poor and black kids who were receiving remedial work. Some educators expressed something like the fixed mindset, that these gifted kids are inherently better at math, but it is going to come out later.

      I’m pro-growth-mindset, and critical of some of Scott’s criticisms. But that’s not fixed mindset. That’s fixed mindset PLUS sucking-up-to-the-powerful, or maybe just plain insanity. It’s hard to tell.

      • Irrelevant says:

        I’m pro-growth-mindset, and critical of some of Scott’s criticisms. But that’s not fixed mindset. That’s fixed mindset PLUS sucking-up-to-the-powerful, or maybe just plain insanity. It’s hard to tell.

        Sucking up to the powerful is a feature of any successful system! Though I’ve heard of what’s being discussed there before, and saying it’s related to mindset paradigms rather than just being a side effect of how the bureaucracy around some gifted programs works is a major stretch. As I understand it, since constant reevaluation is expensive and tiresome, a Gifted label tends to ride with the kid once applied. Now, recall the immediate but dwindling impact of “Head Start”-style interventions: competitive parents who taught their kid well at an early age can get them classed as gifted when they’re actually just better-educated. When this advantage dissolves later (just as you would expect if you’re a fixed mindsetter…) a reevaluation would put them back in with the standard students, but doesn’t happen because that would be very unpopular.

    • Tom Hoffman says:

      Thanks for posting this Janet. I’d add a few caveats.

      One meta-issue is that we’re never going to fix our schools with a grant-driven process. Schools need to be able to run themselves successfully from general funds. I know Janet did not invent our current school finance system, but people in her position tend to develop an extremely jaundiced view of our schools because their work is almost guaranteed to fail by its very structure — government or foundation driven grants. High performing countries have fully funded school systems.

      Performance issues are fairly specific to academic subject. We’ve always been bad at math, so it is a deep hole to dig out of. We also generally model it as a linear progression. Students are at a certain point in the path of math achievement. Setting aside your view of the growth model, this sets into play basic logistical problems. Once you get different groups of kids at different points in the line (and you will), it is difficult to figure out how to get them back together.

      But we have completely different problems in English. There is good evidence we do elementary reading as well or better than anyone else in the world, especially considering the complexity of English. We fall off in high school for still unknown reasons. We kind of stopped teaching writing for a while as a result of NCLB, which didn’t look at writing scores. We have Common Core ELA/Literacy standards, which have nothing in common with the math standards other than the name, and are half-baked and faddy in comparison to the math version.

      Science? Are we even trying? Look at the international survey data in the TIMMS testing report. We have a very high number of teachers reporting not having sufficient equipment to teach science. We talk about science, maybe hand out grants, but we’ve not begun to actually invest in it, even in simple ways like equipping all classrooms.

      History? We’re *really* not trying to teach history or social studies. We’re not even talking about it.

      Math as a subject is a relative outlier. If you use it as your basic model of how education or learning works, you’ll end up in a weird spot.

    • Desertopa says:

      Speaking as someone who actually works at teaching a lot of kids who’re at risk of failing in terms of the numbers on their report card, I think it’s worth pointing out that teachers’ experiences are generally going to reflect that some kids are really a lot harder to teach than others.

      Some of the kids I work with are just completely lost on some or all of their subjects. A lot of it is probably learned helplessness, but now their work is several steps advanced from the point where they got confused and gave up, so they’d have to go way back in terms of building their foundations before they could realistically learn that they have the capacity to get stuff done at the level they’re being asked to. Most of them have more or less given up on caring about academics if they ever did in the first place, and getting them to focus on learning even at levels they’re capable of is a constant struggle.

      Some of the students aren’t yet at the level of being completely lost, but are falling behind and in the process of sinking into learned helplessness. They don’t understand significant parts of their work, but aren’t necessarily so far behind that they couldn’t build the skills they need in a realistic time frame. But, any time spent working on the foundational skills they need is time taken away from completing the large amounts of work they’re loaded down with. They’d rather just have the answers handed to them than go through the trouble of grasping the process, because in the short term in the time it takes them really learn the process for completing one kind of task, they could plow through a few assignments, guessing on or skipping parts they don’t understand, and maybe get a grade that scrapes by at passing. They don’t have a lot of attachment to the long term ideal of being able to understand and complete all their work themselves and get grades significantly better than passing.

      Some of the kids have the fluency to grasp even the more advanced work they’re assigned, when they can be bothered to care about it. A lot of them spend a lot of their time distracting others from learning or getting anything done, and not all the kids who do this are bad students in their own right.

      Some of the kids are a comparative oasis. They care about their work, don’t resist spending time learning basic information and processes, and don’t distract others. If I wanted to create a selective class focused on intensive learning with higher expectations, it would definitely be tempting to fill it entirely with these students. They’re not the only ones capable of learning, but it’s much, much easier to get that return on an investment of time. Some of the other kids might be capable of learning at as high a level, but it’s much more effort to bring them along without making it harder to progress as a group. It’s a lot less trouble to teach a group of twenty of these oasis kids than three of the ones who’re constantly digging their heels in. So while it would be a shame if these kids were left to sink, it’s not surprising if educators are choosing to selectively target some kids over others.

      (For context, all of the students I work with are minorities, 100% black at one location, 100% hispanic at another, so I don’t think I’m subconsciously letting race filter my perception of their abilities.)

  37. “The more important you believe innate ability to be compared to effort, the more likely you are to” …

    This way of putting it ignores the distinction between thinking that effort has a low effect and thinking that ability has a high effect, either of which results in a high ratio of the effect of ability to that of effort. Thinking that effort has little effect is a reason not to put out more effort. Thinking that ability has a large effect isn’t.

    If you had an adequate measure of the total variability of whatever outcome mattered to you due to the two causes, the two beliefs would be linked, since the more variation you attributed to one cause the less would be left for the other. But that seems an implausible assumption, since there are lots of other things causing variation in outcome as well.

    • Paul Torek says:

      I am so glad you said this:

      This way of putting it ignores the distinction between thinking that effort has a low effect and thinking that ability has a high effect, either of which results in a high ratio of the effect of ability to that of effort.

      Yes it does ignore the distinction, and that’s a good thing. Because we are talking about emotions here: people are subject to cognitive dissonance, fragile egos, etc etc. If you fail at a bunch of math problems that Carol Dweck’s researchers just gave you (or so they tell you), and you think it’s about ability, that’s depressing. And as we just saw in the Chemical Imbalance thread, depression saps people’s motivation.

      Carol Dweck isn’t trying to make a contribution to Rational Choice Theory. She’s trying to describe the psychology of actual human beings. I find her account extremely plausible on its face.

  38. Levi Aul says:

    A short but of my own childhood experience that might illuminate some things: I had a really quite bad case of ADD. My parents chose not to medicate me for it, but still at least told me about it. I used this knowledge (mostly implicitly) to figure out that practicing things [at least things without tight reward loops built in, like video games or programming] was just a path that was closed to me. Anything I didn’t start with ability in, just couldn’t be my thing, because I had no willpower to work through the initial period where I suck compared to everyone else around me. No sports, no playing music, no Trivial Pursuit (or the closely-relatedHistory class.)

    But, since I knew this, I was always on the lookout for “the trick”: the thing that would bypass the need to spend time practicing in the first place. For trivia, for one, “the trick” is learning the pattern of the facts, so you don’t have to exert one unit of effort for each new fact. I think this mindset—it’s not really either fixed mindset or growth mindset—tends to make kids at least seem “gifted.”

    My own (extremely self-serving) hypothesis, though, is that this mindset’s appearance or lack thereof in early childhood development is actually responsible for much of the non-genetic component of (at least mathematical) IQ. All mathematicians, at least, are people who let this mindset flourish—so there’s some correlation between math IQ and this “find the trick” mindset. (It might just be the reverse, though: people with high IQs have the necessary horsepower to actually “find the trick”, so they end up with that mindset; those who don’t, but also can’t motivate themselves to practice, end up in the Dweck’s “bad fixed” mindset that results in cheating et al.)

    • Linked List says:

      This matches my school experience. However, it seems like in most non-academic tasks, there is no trick. There is no small set of underlying principles that will let you generate all the knowledge you need.

    • Those who learn early on that they can figure complicated things out are more likely to become experts in complicated fields . Children who are tested as dull tend to have intellectually dull professions and tend to not complete college (there are always exceptions, I know). Children are pretty good at figuring out what they are aren’t good at.

    • Shenpen says:

      All this modern western upbringing you people had… ADD was an unknown term in my childhood, the closest thing was “pay attention dammit or you’ll get beaten”.

      Weird thing is, sorta worked – fear is a strong motivator. But of course not without side-effects.

  39. Josh says:

    So, cranky comment: I think this is bloody obvious but tends to disappear in conversations about this: the relationship between experiences, beliefs, and behaviors is COMPLICATED. The idea that “growth mindset” picks out a meaningful dimension in characterizing a mind in the way “velocity” picks out a meaningful concept in characterizing a physical system to me seems absurd. There might be repeatable experimental results you can get by manipulating an individual’s context, but generalizing that to a cross-cultural profound insight about the human condition is a giant leap from that. So frankly I don’t care WHAT Dweck’s evidence is, because the hypothesis she’s pursuing doesn’t exist in the context of a real theoretical system (as opposed to a physicist doing experiments within the context of general relativity, for instance). I predict in 200 years no one will regard any of this as science….

  40. Nestor says:

    I think one aspect you’re ignoring is that not all effort is equal and not all the variance can be ascribed to “talent”. God knows I’ve experienced my fair share of helplessness and stalling, it’s a real thing, but nowadays I tend to think in terms of finding the right approach to a task. There’s this guru type guy you might have heard of, Tim Ferris, his whole schtick is “hacking learning” to do things more effectively.

    Certainly education is rife with mind games, status plays and social chaff, just yesterday I came accross some online games on seterra, and decided it was a shame I didn’t know all the countries in the world, sure I studied them at school, but it did not take because rote learning is shit. Nowadays I know enough about memory palaces and mnemonics to effortlessly associate every country with a tag that makes it easy to remember and bingo, I know all the countries in the world (Well, oceania is a bit annoying to be honest, dunno if I’ll bother). I am certainly not smarter than I was as a kid (Taken some blows to the head since then) but what I have now is a better learning method.

    Almost every sphere of human activity can be taught with an optimized method, or people can be left to hit their head against a wall until they break it.
    Perhaps Alice was told to do rote memorization, and someone gave Carol a book on memory palaces for her birthday.

  41. Albatross says:

    If a brutal dictator forced you to play chess one week from today, the loser is executed, and you could hire a chess coach, practice, etc. or you could spend a certain amount of hours or days bribing people to find out who your opponent is and their talent level what % of your time do you spend on effort vs. ability? How useful would it be to spend a day finding out you are evenly matched, or at a woeful disadvantage?

    What I like about Dweck is she seems to have lumped all thinking about ability into a time waste for the participants. Ability is real of course, but thinking about ability being 90%, 80%, or even 10% of the outcome is not useful to taking the SAT, passing the marshmallow test or getting into Harvard.

    Every second pondering the inferior or superior or equal qualities of the competition is wasted time. It gets us discouraged or gives us a false sense of security or is anxiety.

    Telling kids part of their SAT score comes down to genetics is unhelpful compared to telling them that taking the practice tests at the library boosts scores.

    Try not! Do! Or do not! There is no ‘try’.

    • but isn’t the truth better than willful ignorance? They, kids, will find out, anyway, and then there will be a shock when they learn that genes are more important then they were otherwise lead to believe. Knowing your intellectual limitations and talents can help you choose the most optimal course of life.

    • What if the dictator offered you the option of playing against an opponent, and getting a million dollars if you win, but killed if you lose (assume that your opponent wouldn’t face either reward or punishment).

      In this equally contrived scenario, isn’t knowing the skill level of an opponent much more important than training? (assuming you’re already a decent chess player).

    • Said Achmiz says:

      Deciding what to do is also important, not just how to do things, and knowing about ability is pretty darn important for that.

      If your options are either win the chess match, or DEATH, then yeah, whatever my ability is, I’ll try as hard as I can, because what the heck is the downside??

      Real life doesn’t work like that. In real life, the alternatives are: get into Harvard, or do any number of other things. If I’m just not gonna get into Harvard, even if I study, like, super hard, then maybe I should check out other colleges? Or take a year off and take a job? Heck, let’s go back to the chess analogy: if I like chess, and consider trying to compete, are you going to tell me to devote literally all of my energy, for the rest of my life to win the World Chess Championship? That would be insane!

      The notion that we shouldn’t think about, or care about, ability, when deciding how to live our lives, is absurd.

    • swanknasty says:

      Knowing your intellectual limitations and talents can help you choose the most optimal course of life.

      Because that’s what people want to say at the end: I lived optimally.

  42. Leif says:

    I think this article about the self-esteem movement, and criticism of it might be relevant. To summarize, it says the movement emphasizes self-esteem over all else, which makes people into lazy narcissists. In the first post, Scott says:

    And from what I’ve got, I find that until about 1980, every study including Dweck’s found that belief in ability was a protective factor. Suddenly this disappeared and was replaced with it being a toxic plague. What happened? I don’t know.

    “About 1980” seems to line up pretty well with when the self-esteem movement was becoming popular. So maybe belief in ability isn’t inherently harmful, but as the self-esteem movement became popular, harmful self-esteem movement beliefs started crowding out non-harmful forms of belief in ability.

    In other words, first the self esteem movement overreacted and decided that belief in (one’s own) ability was the only thing that mattered; then Dweck studied the world that movement created, overreacted, and decided that belief in ability was universally evil.

    (I apologize if this idea has been posted elsewhere. I looked, but I couldn’t find anything. And I also apologize if it’s very obviously wrong; I’m just a layman who’s read a few articles, and I might not have any idea what I’m talking about, but this seemed interesting enough to post.)

    • Douglas Knight says:

      Or the movement had no effect on the experimental subjects, but did have an effect on Dweck.

  43. Ian James says:

    Building on what Paul Torek said above, I don’t think people walk around with some sort of mental representation in their heads which takes the form of (Ability%) + (Growth%) = 1. Growth mindset is more of a “confidence that I feel while I’m expending effort” than a “representation of how effort contributes to success”. And not just any vague confidence: it’s a confidence that my actions really are making a difference in the long term.

    (SSC readers who have never had the experience of, say, looking at the textbook towards the beginning of a challenging course and thinking “what the hell can I possibly do to get my brain to internalize all this?” will have a hard time understanding what I mean here…)

    This would explain perfectly why encouraging kids to adopt growth mindset seems to work in the short term, while ability effects seem to dominate in the long term. Ability and success are correlated not simply as the “direct” result of ability, but because ability imbues the student with confidence (via the perceptions of others, and later, self-perception)–and unless the student is dumb enough to reject the Bloody Obvious Position, some of that confidence will manifest as confidence in the value of practice. Meanwhile, kids who aren’t getting those constant dividends from ability need equally constant intervention–not just encouragement to spend a certain number of hours working, but reassurance that the hours spent will make a difference.

    This conjecture pairs nicely, if depressingly, with your recent post showing that intensive and long-term interventions for at-risk youth translate to modest but real gains in life outcomes later on. Perhaps it would be possible for us to eliminate crime poverty, drug addiction, etc., if only there were enough high-quality mentors who could put in all those hours.

    This leads me to two observations… First, since all roads in my mind lead to basic income, I’d be remiss if I didn’t point out that a basic income would allow more people to spend time mentoring, teaching, etc. But secondly, even if basic income doesn’t come to pass anytime soon, these ideas make me more optimistic for the arrival of virtual reality, as it will widely circulate the presence of high-quality teachers and mentors, whereas, e.g., YouTube and Skype have so far only circulated their images. I’m assuming that the feeling of “presence” makes a big difference because it hits us deep in the reptile brain and allows more of an emotional connection with what we’re seeing. I’m also assuming, of course, that VR technology can deliver on this front–but early reports are very promising.

  44. Ed says:

    I haven’t read most of the comments so I might be repeating things that have been said 100 times already.

    It’s pretty clear why rationalsts would be biased against growth mindset: one of the core tenets of rationalism is that one should try to believe things that are true, and disbelieve things that are false. Behind much of religious belief, even though believers will tend to deny it, is the idea that even though the doctrines might be literally false, believing in them has various salutary social or psychological effects.

    It becomes a value vs. meta-value question, similar to questions of procedural versus substantive justice. Just as we can argue about whether it’s better to release criminals on the basis of improperly collected evidence, the question of when it could be worth it to believe something false for some pragmatic reason does not seem to have an obvious answer.

    It’s convenient then, to reinforce one’s principle by challenging the degree to which believing false things has real benefits. Religion probably isn’t as good for people as its proponents say it is. The difficulty with growth mindset, then, is that it seems like the best test case for the ‘false-belief-pragmatist’.

    In part this is because there are good studies. Another part of it is that the effort-primacy belief, while false, doesn’t seem pernicious beyond a general erosion of the principle of believing only those things that are true. It introduces a bias towards egalitarianism that many people (myself included, though this will probably be unpopular here) believe is warranted as a sort of ‘social justice prior’.

    A useful contrast is with what I might call ‘fairness mindset’, the idea that the world is basically fair and properly rewards good performance/character/whatever. I’d be curious to see studies of whether ‘fairness mindset’ has the same sort of positive effects on effort that ‘growth mindset’ does. If one’s effort isn’t going to matter because the game is rigged, why bother trying? But ‘fairness mindest’, while probably not any more false than ‘growth mindset’, encourages a pernicious blindness to unfair elements of the status quo, so nobody is clamoring that we encourage its adoption.

    • TeslaCoil says:

      The ‘fairness mindset’ normally goes by the name ‘Just World Fallacy’. As one would expect, it’s been studied before.

      Rubin’s article from 1975 is as good a
      starting point as any other. The conclusions are, of course, not a bit more trustworthy than Dweck’s.

    • Paul Torek says:

      Great comment, Ed. While I and a few others have briefly touched on the question, “do true beliefs always pay?”, nobody tackled it like you just have.

  45. Reading these posts, I think I realized why the K-12 schools are doing Growth Mindset training. The self-esteem posts made a light bulb go off for me. (I taught math and reading during the self-esteem craze and was told to stop the math and reading lessons for 9 weeks to teach self-esteem activities. I know that era well.)

    NCLB put the infrastructure in place so that we could say what kids should learn in each class/grade, and we can measure whether they have learned it. The next era is the Race to the Top era. This one is optional. You apply for the grant, and unlike any other grant in education until now, they tell you what you have to do rather than funding “innovative ideas.” You have to adopt Common Core, and you have to have a way to predict individual student academic growth, then measure teacher effectiveness using predicted vs. observed growth. Growth is big on educators minds.

    The idea of growth, and measuring it, is new in education. First we thought in terms of who is smartest, then moved to what percent are proficient. Move everyone forward from where they are, at a rate that is right for them, is a new idea.

    Educators need training on what this means, or how to think like this. (Educational training is usually about how to think, not how to do something.) Growth Mindset was already there, and could be used. They are treating this like the self-esteem craze. Dweck’s stuff was ready and Growth is the new thing in education. Her stuff was right place right time. Educators don’t necessarily believe her whole theory. They just need to adopt the idea of growth.

    After the state-wide race to the top grants, there was a round of smaller district rttt grants that called for finding way to meet each child where they are, and move them forward from there. We know from the data we can now get that measures proficiency, that some kids have already mastered all the objectives before they take a course. (How will the teacher show growth if she is to teach him stuff he already knows?) And some are not proficient when it is time to move on. Grant funds are being given to figure out how we can manage classrooms or schools where kids are met where they are and given lessons designed to move them forward from there. This is easier said than done when there is one teacher and 30 kids. But, now content can be delivered via the web, so it is possible. States with districts that have small rttt grants have to legislate credit by demonstrated content mastery rather than seat time.

    They are getting Dweck’s training because it is there. They need training to understand this new way of thinking. Something a little simpler that doesn’t make Fixed vs Growth so either/or would probably sell like hot cakes. They are using this because it was ready to go when they had to start understanding what Growth means.

  46. PGD says:

    And yes, it’s a sin to privilege your own experience and priors over the results of good studies, but sometimes it’s necessary. And it’s another sin to prefer the results of broad ecological studies to controlled experimental trials, but sometimes that’s necessary too.

    Totally disagree with this. Maybe it’s a sin to privilege your experience and priors over the results of physics and chemistry studies, but even the best social science studies are extremely assumption dependent, particularly when it comes to ecological / out of sample validity. Social science studies — including not just psychology and economics but important aspects of behavioral genetics — are just on an entirely different plane than the physical sciences. I’d almost say it’s a sin *not* to examine these studies very critically in light of your own experience and common sense. After all, we are all beneficiaries of massive amounts of observational data on how psychology and social systems work.

  47. Hilary says:

    Scott, I’m swinging by to drop two papers I just came across in your lap to see if you’d like to check them out. I admit I have a bias *toward* the growth mindset theory. Still, like you, I’m quite skeptical of big claims in education, because every other education intervention I can ever find tends to have inconsistent, marginal effects.

    Incidentally, I am currently reading about sign language interpreter education, and came across this, from Bontempo & Napier (2011):

    “Like self-efficacy, goal orientation is considered a relatively stable personality trait. In this paper we are specifically concerned with learning goal orientation (rather than performance goal orientation, a related but different construct), given its potential link to aptitude for interpreting. Learning goal orientation is sometimes described in the literature as “mastery orientation” or “action orientation,” and is a desire to develop oneself, to acquire new skills, to improve one’s competence or to master a new situation or context (Bell & Kozlowski 2002)…. Goal orientation is also positively linked to self-esteem (Button et al. 1996) and to self-efficacy — people with high levels of learning orientation seem to be “buffered” by the negative effects of failure (Bell & Kozlowski 2002). Indeed, people who are learning-goal oriented “are not threatened by failure; to them, failure represents an opportunity to extend one’s competence through enhanced effort” (Button et al. 1996: 31).

    That last sentence sounds an awful lot like growth mindset. And ‘learning- vs. performance- goal orientation’ seems to make a potentially relevant distinction between motivation to *perform* and motivation to *learn.* Maybe these are axes Dweck should consider measuring independently.

    (I am in the middle of writing a very different paper, on a time crunch, so I regretfully can’t go check out those citations myself, but thought Scott or others might be interested in doing so.)

    Full citations:
    Bell, Bradford S. & Kozlowski, Steve W. J. (2002). Goal orientation and ability: Interactive effects on self-efficacy, performance and knowledge. Journal of Applied Psychology 87, 497–505.

    Button, Scott. B., Mathieu, John E. & Zajac, Dennis M. (1996). Goal orientation in organisation- al research: A conceptual and empirical foundation. Organisational Behaviour and Human Decision Processes 67 (1), 26–48.

  48. pkinsky says:

    >On the other hand, I’ve also met people who say “I could succeed if only I put in some effort, but I have some mental block / depression / ADHD / low conscientiousness score that makes it impossible for me to work that hard, so better go eat worms”, then sabotage themselves at every opportunity.

    General request: Anyone have any tips for this issue?

    • Linked List says:

      Find something that’s intrisically, not just extrinsically, rewarding for you. That is, something that you find rewarding independently of whether you’re good at it or not.

      IME this is the only way to get rid of that procrastination-inducing anxiety.

      • Dude Man says:

        If none of the things I find intrinsically rewarding are extrinsically rewarding, how can I find something that meets both criteria?

  49. Ellen says:

    “I could succeed if only I put in some effort, but I have… low conscientiousness score that makes it impossible for me to work that hard, so better enjoy a fun-filled ambition-free life of leisure instead.”

    Scott, I have a serious question. Would you object to the above attitude? I don’t like growth mindset either, because most things in life have come easily to me, and I’ve never felt right about taking much credit. I’ve sometimes wished I could donate my natural academic or running ability to someone who would actually put some effort into “growing” it or doing something useful with it.

    Anyway, one of the first posts I read here was the Parable of the Talents, which talked about comparative advantages. Even if I knew what mine was, why should I pursue it? Your “All God gets to ask me…” conclusion works only if God exists. What if I’m in the top 1% on standardized tests and was my high school’s valedictorian, but am content at my super fun nannying job and love all the free time it gives me? I still feel fulfilled and super appreciated, and I do my small part to help the rest of the world through effective altruism.

    I’m a great nanny, but I don’t think nannying is my comparative advantage. Would you try to convince me I should be actively trying to discover what is and pursuing that instead? No, right? Since nobody is perfect and everything is commensurable? If I love my life and I don’t succumb to the (mostly good) societal pressure to do something ambitious, and I don’t have enough intrinsic altruism, is there really anything you can tell me?

    Anyway, I’m a new reader and want to tack on a huge “thank you” for all your writing! I enjoy your blog even more than my favorite novels; I’ve never found anything like it. I discovered it about a month ago, and since then I’ve read through tons of the archives, shared articles with at least a dozen people, got into Less Wrong, and have already read about 40% of the Rationality ebook. It has all been super relevant, since I very recently de-converted from Christianity. Before that, I was actually a bit more ambitious. Long before I had even heard of effective altruism, I had my own Christian version of it, in which I planned to pursue a high-income career and donate 90% to fund missionaries. Now that I don’t believe in hell, excessive ambition seems (relatively) pointless to me. :/

    • Irrelevant says:

      Scott Alexander, Priest to the Faithless.

    • Princess Stargirl says:

      You matter just as much as anyone else. If you manage to give yourself a fulfilling life that counts for quite alot. If you make your own life enjoyable and make the wider world a marginally nicer place you are doing quite well. If you ever considered donating 90% of your income I doubt you have it in you to be so selfish as to never help others at all. Do not forget how easy it is to accidentally harm others when you try to help. See for example how much pain people inflict on their own children for basically no gain (or they actively harm the children if they are lbgt etc).

      Imo just make yourself happy and be careful not to hurt other people, especially when you try to help them. If you pull this off you are doing great! Don’t worry about not doing more.

    • Scott Alexander says:

      I think there are two forms of ambition here. There’s the one where you want to be President. And there’s the one where you’re okay being a janitor, but you want to be a very good janitor who makes things especially clean. Growth mindset is useful for both.

      On the other hand, if you’re absolutely happy with your current life, then…neat. If the demon who was supposed to tempt you with ambition and desperate worries about your own inadequacy got lost on the way to your brain and ended up being eaten by wolves, by all means enjoy your good fortune.

      • Ellen says:

        Yep, it seems useful. With the kids I nanny, I’m consciously praising effort over ability now (literally now, as I check over perfect math homework…a real life consequence of your blog!). And if I end up valuing worldly success over fun, I’ll praise any hints of ambition far more than either practice or innate ability. You could have mentioned ambition in this context, but didn’t, probably because to you and the rest of the world it’s obvious that a desire for success helps people succeed.

        But yeah, what people praise is *huge*. I can easily trace my own lack of ambition to my upbringing. At my house, we bragged about acing tests without studying or having done assigned readings, getting homework for one class done in another, skipping college classes to hang out in the rec room, etc… My dad is smart and happy, working just enough to live comfortably and spending his free time playing, reading, and winning poker/fantasy sports tournaments. So naturally, I grew up convinced this was the optimal way to live. Then I started to get really curious about why everyone else seems to favor ambition so strongly. But if they’re mostly just possessed by inner ambition demons, I guess I’ll just be thankful that (1) those demons exist and (2) they missed me.

  50. TomA says:

    If I understand this post correctly, the “growth mindset” movement is an attempt to modify the behavior of adolescents via an emphasis on memetic reprogramming versus reliance on innate ability. This seems anti-evolutionary to me. Indoctrinating young people with a proclivity for memetic infection is the equivalent of suppressing someone’s mental immune system.

  51. Schmendrick says:

    Perhaps I’m just missing something here, but it seems to me that the SCP works something like this:

    1. Each person has a maximum amount of effort they can devote to a single task or multi-task project.

    2. There is a point p’ between 0% and 100% effort where expending further effort becomes displeasurable [sic?].

    3. Without motivation (either extrinsic or intrinsic), people will only expend p’ amount of effort on any given task or project.

    4. Believing “the amount of effort expended in the process of completing a task is an important predictor of the quality of the finished product” provides motivation to keep expending effort past p’.

    5. The more you believe (4), the more motivation you have to expend effort beyond p’, to a maximum of 100% possible effort. ie., believing that effort is 20% of the result is a weak motivator, while believing that effort is 80% of the result is a much stronger motivator. This manifests in practice by thinking: “aaaah, this is really hard, and how well I do at this is 80% determined by pre-existing factors, and I really want to go out bowling with my friends, so I’ll just stop here because working any harder won’t really make all that much of a difference.”

    6. Believing a cartoonishly exaggerated version of (4) – namely, “the amount of effort expended in the process of completing a task is the sole determinant of the quality of the finished product” – is the strongest possible motivation, and thus the most likely to produce 100% effort and a maximization of an individual’s probably-predetermined talents.

    • I don’t agree with your 6. That isn’t “the strongest possible motivation.”

      Suppose Bill believes that the only thing determining his income is how hard he works, and the range of possible incomes in his profession is from $20,000 to $30,000. Joe believes that talent also determines income and that, for someone with his level of talent, the range of possible incomes is from $20,000 to $40,000.

      The motivation depends on how sensitive you think outcome is to effort, not on how sensitive you think it is to other things.

      • Schmendrick says:

        That’s fair. All I was trying to say was that solely within the context of “try hard,” it could be useful to believe something that may not be factually accurate, but carries a useful incentive. Obviously in the real world, things get super complicated.

  52. mobile says:

    I didn’t make this connection until I saw Eric S. Raymond’s link to his essay onthe curse of the gifted, but I have identified growth-mindset theory of international economics.

  53. Brendan says:

    The whole thing seems harmless if children are not being pressured into doing things that they are neither good at nor enjoy. Particularly if we’re dealing like hobbies that are not socially necessary. [Individually speaking, not all hobbies put together]

    I also have a hard time imagining that it’s healthy to tell someone that [nearly?] all the variation in ability is due to effort. The lie might make them better at a particular thing, but who is actually better off by this?

  54. Joe says:

    I think this one can be solved by a thought experiment.

    Let us imagine that everybody completely believes in growth mindset and is putting in 100% at all times.

    What then seperates them?

    Innate ability.

    So it then occurs to me that those who believe in innate ability over effort and grit might have been in an environment where everybody involved was putting forth maximal effort. I’m thinking highly competitive siblings and the like.

    As to why someone might want to push growth theory, that’s easier to explain and more sinister. If some degree of your output is not actually due to your own efforts but actually due to something you got at random, then you cannot really claim 100% of the success. Equally, when it’s time to divvy up the spoils of any endeavour, claims of keeping it all based on merit are stillborn.

    Finally (and this is an adjunct of some people in the self help movements appalling lack of compassion at times) if everything is just effort, you get to handwave away other peoples problems as them simply not being, for want of a better term, “arsed”. Poor = lazy. No prospects = lazy. Discriminated against = lazy. Terrible economy = lazy. Ill = lazy. Mentally ill = lazy.

    Neo liberalism in microcosm.

    Final thought – how far does the growth theory rabbit hole go? Gay = lazy?

    • Liz says:

      And that right there is why I hate the “hard work is everything” mindset.

      As someone who’s always put a lot of effort into trying to do things correctly only to keep running into roadblocks that have jack all to do with lack of effort on my part, I am beyond sick of being told by all and sundry that my lack of success is due to my being stupid and lazy thanks to the “hard work is everything” mindset that hard work is the only possible factor involved. It’s utterly soul-sucking not just putting in 100% and having it not work, but then having nobody believe you put in more than 0% effort and then treat you badly as such.

      So why should I try at all? If I fail, I’ll get treated as if I put in 0% effort regardless of how much effort I actually put in, so I may as well go ahead and not even bother with any effort unless I know I’m virtually guaranteed to succeed. Why waste the effort for literally worse than no gain thanks to societal perception?

      • Joe says:

        I hear you.

        I’ve also noticed that the “hard work is everything” crowd will often (but not always) claim in other contexts that success is easy, you just have to do x,y and z!

        Perhaps it’s signalling of some sort, with attempts to gain superior social status depending on context for those people. Whatever the case I agree, it’s enormously annoying.

        It’s also soul destroying to believe in growth theory, design yourself a dream of some sort (better job etc) and then jump into the fray with the conviction that hard work will take you there, only for everything to drop to shit once reality has it’s say.

        What is it? 90% of new businesses fail in the first 2 years? Growth theory says those failures are all just being lazy!


        • Liz says:

          Yep, I’ve encountered that one a lot, too. I’ve dubbed it the “Rube Goldberg fallacy”, because it basically operates on the principle that if you do all these careful sets of actions which need to all perfectly work and luck out together (much like a Rube Goldberg machine) you’ll succeed, and so therefore it’s easy to succeed and you’re just lazy.

          I agree I think it’s a superiority complex. You don’t want to have to admit you got lucky on some level, so you just tell yourself it’s because you worked harder than everyone else.

          It’s not that hard work means nothing. Ability and opportunities won’t get you very far if you just sit back and do nothing with them. But likewise, all the hard work in the world can’t get you very far either if you don’t have the opportunities and ability to work with.

          Even circumstances can work against you. Like with the business example. I’ve personally worked for companies where they were running their businesses quite well and doing everything right, and then suddenly something happened that tanked the demand for their services.

          For instance, the full-service moving company that got slammed by the housing crisis, since people that can’t afford to sell or buy houses obviously can’t afford full-service moving over UHauls or not moving in the first place. And the downturn in the economy also tanked businesses needing full-service moving as a part of things like expansions.

          Or for an example more directly related to the question of ability: My gym classes in high school. I am a massive, massive clutz. My hand-eye coordination is terrible and my balance is wonky and my sense of direction is lacking. So I spent a LOT of time working really hard in gym class to the point of exhausting myself, only to fail repeatedly because of those physical limitations.

          And what did I get from my gym teachers? You guessed it. “You’re just not trying hard enough! Put some effort into it!” Even as I was actually sweating and tired and frazzled because I was putting a metric ton of effort in already.

          Finally one year I just went ahead and deliberately tanked my gym class grade. I showed up, dressed in my gym clothes, but refused to do absolutely anything unless it was really easy and non-competitive. (Which I think basically amounted to an aerobics module.) I decided to show them what “not trying” actually looked like.

          In high school I actually finally found a gym teacher I absolutely adored and worshipped, because he actually understood that there were people who existed that aren’t good at athletics. He recognized that even though some of us were failing, it was because we just didn’t have the ability and not because we weren’t trying. We even actually made a little progress for a change, because he was able to notice what stuff we were doing wrong that actually was fixable within our limitations and gently suggest to us how to do so. We didn’t become masters or even honestly basic competence, but we did at least fail a little less often.

          IMHO a progress mindset requires actual progress.

  55. Adam says:

    I have to say that these posts about growth mindset have been unexpectedly cathartic. I had completely swallowed the growth mindset propaganda, to the point where I had convinced myself that it didn’t matter how much innate ability I might have, I am going to try really really hard and never give up, dammit! I have had the mindset for years that I have a huge amount of work ahead of me, so that I can eventually achieve a bunch of goals and then make a bunch more goals for myself (with a bunch of new work). Now that I realize that there are other explanations for the studies that support growth mindset, I feel that I can relax and take a deep breath and have fun using my natural talents, plus make a strong effort sometimes when I feel like it.

  56. Jason K. says:

    I didn’t see this in the 1st 50 comments;

    What if we are getting cause and effect out of order? What if being really good at something tends to cause us to attribute to our own effort and not innate ability and visa-versa? In other words, we take personal credit when we are great but blame things out of our control when we suck. So success causes us to believe in our own effort, not the other way around. We can call this the RH (Robin Hanson) position.

    You kind of touch on it here, but it is backwards:

    “dumb people might preferentially believe ability doesn’t matter to make themselves feel better about not having it”

    This is illogical if my goal is to minimize personal accountability for failure. If I am dumb and believe that ability doesn’t matter, then my lack of personal effort is obviously the cause. As I have control over my personal effort, then I am to blame. However, if I believe ability is everything, then I have excuse that both allows me to not bother trying again and not take any blame for not trying. What can I do about lacking the ability to succeed?

    There is suggestive evidence for this in the differences in self assessment of wealthy vs poor individuals. Wealthy people tend to over state the degree which they are responsible for success (ex: ‘It’s your fault for being poor’), and poor people tend to understate the degree to which they are responsible (ex: ‘The world screwed me over’).

    So how much you believe in the value of effort is not the driver of success but it’s barometer, hence why the people that believe in the value of effort more are more successful. Growing up in Japan leads to learning Japanese, but learning Japanese does not cause you to grow up in Japan.

  57. Carl Shulman says:


    “The efficacy of academic-mind-set interventions has been demonstrated by small-scale, proof-of-concept interventions, generally delivered in person in one school at a time. Whether this approach could be a practical way to raise school achievement on a large scale remains unknown. We therefore delivered brief growth-mind-set and sense-of-purpose interventions through online modules to 1,594 students in 13 geographically diverse high schools. Both interventions were intended to help students persist when they experienced academic difficulty; thus, both were predicted to be most beneficial for poorly performing students. This was the case. Among students at risk of dropping out of high school (one third of the sample), each intervention raised students’ semester grade point averages in core academic courses and increased the rate at which students performed satisfactorily in core courses by 6.4 percentage points. We discuss implications for the pipeline from theory to practice and for education reform.”

  58. Pingback: What I learned this week: 16/2015 | suboptimum

  59. Pingback: A week of links | EVOLVING ECONOMICS