The nail in Growth Mindset’s coffin?

//The nail in Growth Mindset’s coffin?

As I’m sure everybody already knows, mind sets are beliefs about the nature of characteristics like intelligence. The theory is that students with growth mindsets believe their ability can be changed with effort and therefore do better academically than their peers who have fixed mindsets. Given the appeal of this theory, it’s small wonder that schools around the world have rushed to intervene with their students in order to mould their mindsets.

In January last year I wrote one of my most popular blog posts to date, the controversially titled, Is Growth Mindset Bollocks? In it I detailed the reasons for doubting the efficacy of what has become probably the most over-hyped, go-to intervention in the last ten years. In discussing a high profile failure to replicate, and Carol Dweck response that, “Not anyone can do a replication,” I concluded: “It could be that they’re just doing it wrong and have a ‘false growth mindset’, or it could be that such appealingly simplistic ideas about making profound changes to children’s academic attainment are bollocks?”

So, what’s changed? Well, a new, preregistered*, study of the effects of Growth Mindset on students’ Grade Point Averages in the US has been released. The study has a huge sample of over 12,500 students in 65 different American schools with the result that the “intervention reduced by 3% the rate at which adolescents in the U.S. were off-track for graduation at the end of the year”.

This is good news, right? Growth Mindset interventions actually work! As ever, we should proceed with caution. The study appears to show that giving students two 25 minute sessions on how the brain forms synaptic connections when we struggle has a small, but real effect on students’ outcomes for a very low cost. The authors also note that some students benefitted far more than others, and those who seemed to benefit most were lower achieving students and students in schools with “supportive behavioral norms”. What this might suggest is that students who have previously underachieved improve when told that if they took more responsibility and worked harder they might do better, and that good behaviour makes a positive difference to any intervention. Neither of which are all that surprising.

This is important because, as Timothy Bates – one of the researcher Dweck accused of going about replicating her studies in a “willy nilly” way – points out, whether the results where down to believing that basic ability is malleable, or that working harder improves results. The first option – the growth Mindset hypothesis asks us to believe in magical thinking where as the second is about how conscientiousness we are. He says of his research, “We find beliefs about the malleability of basic ability are irrelevant: it’s all about work”.

In other news, this paper, reports on two meta-analyses into the circumstances in which growth mindset interventions are effective. From the abstract:

In our first meta-analysis (k = 273, N = 365,915), we examined the strength of the relationship between mind-set and academic achievement and potential moderating factors. In our second meta-analysis (k = 43, N = 57,155), we examined the effectiveness of mind-set interventions on academic achievement and potential moderating factors. Overall effects were weak for both meta-analyses. However, some results supported specific tenets of the theory, namely, that students with low socioeconomic status or who are academically at risk might benefit from mind-set interventions.

The paper is behind a paywall, but Stuart Ritchie shared this figure showing the effects of over 40 studies into the effects of mindset interventions. He summarises the main findings of the study thus:

1) Correlation of growth mindset with achievement is small, r = .1
2) Effect of growth mindset interventions on achievement is tiny, d = .08.

What can we salvage from all of this?
Well, one conclusion is that maybe it’s worth giving “students with low socioeconomic status or who are academically at risk” 50 minutes worth of animated bobbins about synapses and brain cells. Certainly, there’s not too much of an opportunity cost. But we should all reflect on the fact that, as things stand, there’s no reason to believe spending more time on this sort of intervention is likely to be worthwhile. If nothing else, maybe we should take away the well-worn truths that well-behaved students in orderly, supportive environments, and students who understand the relationship between effort and outcomes are things we should continue to strive for.

What ought to be obvious to anyone who’s spent anytime reflecting on their own habits and behaviour is that we all try hard at things we believe we are good at, and we all quit things we think we suck at. This is human nature. If we’re serious about changing students’ beliefs about their ability we ought to commit far more time to ensuring they can be successful at the subjects we teach.

Of course this won’t be the last word on growth mindset interventions. Like so many other ‘good ideas’ they’ll lurch, zombie-like, around the educational environment for years to come. Many will carry on with what they think to be right, no matter what the evidence turns up. As for all true believers, faith trumps reality. For those who are concerned but not yet ready to let go the belief in the power of the growth mindset, ask yourself, what sort of evidence would convince you to change your mind?

* For those who don’t know, a preregistered study is one where the research team register their plans before the trial so that they cannot manipulate the data later. It’s also worth noting that this is a preprint – a version of a paper that precedes publication in a peer-reviewed journal. This might suggest it hasn’t yet been fully validated through the peer review process. The authors say, “The findings and manuscript will almost certainly change before publication. The investigative team welcomes suggestions for clarifying or improving the research or the documentation of the findings.”

2019-01-25T13:50:03+00:00March 6th, 2018|psychology|


  1. Anne Watson March 6, 2018 at 10:48 am - Reply

    My own view is that what needs to change first is the mindset of teachers and school systems who believe that, and act as if, certain groups of students, or students who show certain behaviours, are going to be unable to do mathematics simply because so far they have not done well at it. Only then is it worth working on their students’ self-belief, aspirations and perceptions of the subject because then they can do it with conviction, however they do it So anything (including mindset studies) that shifts the log jam is useful when used by committed teachers who really want to change students’ mindsets and convince students to change – it’s the old Hawthorn effect used for positive purpose. But bad science is bad science (such as ‘making mistakes makes your brain grow’) and there needs to be a more critical look – as you have provided – at the data before certain methods to become THE way to do so. What is it that makes people more willing to grab every passing whim than to understand at the learning of concepts? The mindset movement has achieved, to some extent, what I say in my first sentence.

  2. ANDREW ORDOVER March 6, 2018 at 8:24 pm - Reply

    If anyone is teaching or promoting growth mindset as magical thinking, then they’re going about it wrong, for sure. For me, the crucial difference between the old “self-esteem movement” work and growth mindset is precisely that it’s NOT just about belief; it’s about the actions we take based on our beliefs. If we believe that effort influences outcome more than innate ability, we’ll work harder. If we know that error and failure will be treated in the classroom as opportunities for exploration and growth, rather than opportunities to be shamed and punished, then we’ll stick our necks out and take more risks in areas where we feel shaky and uncertain. I agree that a lot of this OUGHT to be more common-sense than rocket-science, but whatever you want to call it, it needs to be emphasized and supported. Too much of what we do in schools, whether we mean to or not, is oriented towards the fixed mindset and the idea that some kids can do it and some kids just can’t.

    • David Didau March 6, 2018 at 10:54 pm - Reply

      “If we believe that effort influences outcome more than innate ability, we’ll work harder.” Really? What if you work really hard and still perform poorly? Logic dictates we’d be better off expending effort on something we’re more likely to succeed at. The truth is, hard work is not enough. Yu need the mental representations of success to become successful. We need to stop messing around with growth mindset and teach kids the foundational knowledge to gain mastery of what we want them to learn.

      • ANDREW ORDOVER March 6, 2018 at 11:03 pm - Reply

        Foundational knowledge: absolutely and 100 percent. But I don’t see that as an either/or with messaging re mindset.

        • David Didau March 6, 2018 at 11:40 pm - Reply

          But “messaging re mindset” seems to be pretty pointless. Why not teach them instead?

          • ANDREW ORDOVER March 7, 2018 at 12:01 am

            Again, i just don’t see it as necessarily “instead.” Teach them. Expect a lot of them. Demand rigor of them. And don’t accept things like “I’m not good at math. ” Give them more than one opportunity to get things right. Don’t let them think that what was, yesterday, is what must be, tomorrow. Ultimately, good teaching is good teaching. The labels are only helpful if they pinpoint things we’ve been missing.

          • David Didau March 7, 2018 at 12:51 pm

            Yes, I agree with all of this. What is it we’re disagreeing about again?

  3. Catherine Scott March 6, 2018 at 9:22 pm - Reply

    Christ on a crusket did you really just say mindsets isn’t worth squat because it only helps the kids who need it the most?
    As a tool for self understanding it also really helps self handicapping high achievers. I teach at the tertiary level and I’ve had a large number of students come up after the mindsets lectures to say that they finally understand their own difficulties. If it had been around when I could have benefited from some self awareness on the topic it would have helped me too.
    If you are up for a chat email me at

    • David Didau March 6, 2018 at 11:44 pm - Reply

      No, I didn’t say that 🙂
      I said, mindset interventions have a tiny effect on acheivement but ever so slightly more so for “those who need it most”. What this suggests is that “those who need it most” would be far better off with something more than self-indulgent guff.
      Here’s what the authors of the study say:
      Some researchers have claimed that mind-set interventions can “lead to large gains in student achievement” and have “striking effects on educational achievement” (Yeager & Walton, 2011, pp. 267 and 268, respectively). Overall, our results do not support these claims. Mind-set interventions on academic achievement were non- significant for adolescents, typical students, and students facing situational challenges (transitioning to a new school, experiencing stereotype threat). How- ever, our results support claims that academically high- risk students and economically disadvantaged students may benefit from growth-mind-set interventions (see Paunesku et al., 2015; Raizada & Kishiyama, 2010), although these results should be interpreted with caution because
      (a) few effect sizes contributed to these results,
      (b) high-risk students did not differ significantly from non-high-risk students, and
      (c) relatively small sample sizes contributed to the low-SES group.
      The results do not support the claim that mind-set interventions benefit both high- and low-achieving students (e.g., see Mindset Scholars Network4). Mind-set interventions are relatively low cost and take little time, so there may be a net benefit for students’ academic achievement. However, there may be a detriment relative to fixed-mind-set conditions when students are confident in their abilities (Mendoza-Denton et al., 2008). Regardless, those seeking more than modest effects or effects for all students are unlikely to find them. To this end, policies and resources targeting all students might not be prudent.

      The message that hard work = success has been around for 100s of years. Did you really not know that effort was strongly correlated with acheivement when you were at school? Does we really need pseudoscience to tell children that being conscientious makes a difference?

      • snowqueen March 7, 2018 at 6:00 am - Reply

        Actually, no I didn’t. I thought I was ‘good at English’ and ‘bad at Maths’ and so gave up on the latter. I didn’t think effort would help. Later in life, I found that despite less natural aptitude, I COULD improve my maths. As a commenter above notes, the teachers all happily colluded with my perception that I was a lost cause. It’s all very well saying teachers should ‘just teach’ but it ignores the fact that teacher education has a cultural context and practices are linked to discourses of the pupil.
        Whether it’s old or new news, Dweck’s work strengthens a helpful notion and also provides useful communication strategies to support teachers in feeling confident in working with students. Also some of the gains for students are emotional – reduction of anxiety in high achievers, for example – which aren’t necessarily demonstrated through outcome/attainment studies.
        In my experience people often don’t bother to read Dwec’s work fully and think the significance is in ‘hard work pays’. It’s not, it’s in providing an alternative narrative of individual capacity.

        • David Didau March 7, 2018 at 12:50 pm - Reply

          Being bad at maths (or anything) is primarily caused by a lack of knowledge. You’re right to say that teachers must not conclude that poor performance in a subject reveals anything innate but ought instead to point them towards instruction designed to fill these knowledge gaps. Growth mindset adds nothing except woo woo distraction from the real issue.
          If you suspect I’ve not read Dweck’s fully please feel free to search through the many articles I’ve written about growth mindset on this blog and suggest where my knowledge gaps might be 🙂

      • Michael Pye March 8, 2018 at 2:38 pm - Reply

        Careful, Carols theory is not pseudoscience. That can only occur when we have strong evidence it doesn’t work but we use it anyway. Might be better to say that it doesn’t seem to add any more value then the belief that hard work makes success more likely. We could also advocate that a direct comparison between these approaches as a valid way of exploring if the theory really adds anything other then a trivially obvious truth.

  4. Simon Butler March 7, 2018 at 3:42 pm - Reply

    I have two children both brought up in the same environment with two teachers as parents. Both have learnt about growth Mindset at school and home. They both know that if you practice something you are likely to get better at it. One eats books, the other spits them out! One has a growth mindset for academic learning one has a growth mindset for the X-Box. In the end the desire to learn anything has to come from within the individual. No amount of coaxing, cajoling and occasionally chastising has thus far convinced the latter to have a growth mindset for academic work. Hopefully though he night become a game designer!

  5. Catherine Scott March 8, 2018 at 12:19 am - Reply

    There is no doubt that how we think about and explain performance is hugely important, otherwise we wouldn’t find the consistent cultural patterns we do. In Australia Kids from cultures, basically Asian ones, that emphasise effort perform better on everything from VCE/HSC to NAPLAN and PISA compared to those from cultures that attribute success to inbuilt ability. James Flynn (of the Flynn effect) has estimated that the extra effort that Asian students apply is equivalent to around a seven point IQ advantage.
    Messing around inside a fixed mindset culture like ours is swimming against the grain. Hard to get traction. At the same time we are trying to promote growth mindset we are setting up G&t programs, labelling some students as winners, but the majority as ‘losers’. I honestly believe that our slippage in PISA and failure to improve on NAPLAN are at least partially explained by this.
    I’m also always impressed by the number of people who contact me after I’ve taught mindset to say that at last their experience makes sense. These are people who were labelled as smart and then spent their childhood and youth self handicapping to avoid possibly discovering that they ‘aren’t smart’.
    Mindset dovetails nicely with all the metacognitive and non cognitive stuff, the importance of having good knowledge of learning strategies, good study habits, persistence etc, which are better predictors of performance than IQ.

    • David Didau March 8, 2018 at 7:35 pm - Reply

      The correlation with IQ and academic performance is r=.81. I don’t know if anything else that comes close to this.

  6. goddinho March 8, 2018 at 3:04 am - Reply

    So if I’m reading this correctly- having a growth mindset won’t make any difference if your not high in trait conscientiousness. This seems perfectly reasonable.

    • David Didau March 8, 2018 at 7:35 pm - Reply

      I think it’s more that all these traits are context dependent.

  7. Michael Pye March 8, 2018 at 2:44 pm - Reply

    No. David is suggesting that the effective component of a growth mindset is conscientiousness. If true focusing on conscientiousness would be a better idea. The new ideas overlap but are not the same so the distinction matters. For example you can lack a growth mindset but can be very reliable and professional.

  8. Michael Pye March 8, 2018 at 2:46 pm - Reply

    two ideas

  9. Mario March 8, 2018 at 10:17 pm - Reply

    A tiny growth mindset intervention (two online survey/interventions of 25 minutes) has “a tiny effect on acheivement” would be a more accurate description.
    Dweck’s Self-theories is just one of many lines of research on motivation. Are there any other theories that you’ve looked at that you feel are worth pursuing?

  10. Glen Gilchrist March 9, 2018 at 10:04 am - Reply

    Love (as always)
    Like smacking a puppy and criticising Harry Potter, you have the knack for igniting exactly the debate that we should be engaged in. (Not suggesting that you engage in those activities)
    It doesn’t matter what my opinion / anecdote is (so I wont share them – as they are just that opinions) and we all agree that making children more resilient / motivated is self evidently a good thing – what matters is a critical debate about when an idea becomes “mantra” or “dogma” — especially when a lot of money is made on the back of it.
    We (the education profession) should base our opinions on evidence – either primary research and/or our own smaller scale action research; not subscribe to the “it must work as it makes sense – and it worked for me once” – if it then passes those tests, it works for you – good. But other things might equally work and prove more cost effective. What is worrisome is when things become “trendy” and it seems unacceptable to ask probing questions.
    Given the range of conflicting evidence out there over GM, I think this meta study does a good job of synthesising the bigger picture.
    Schools spend public money on initiatives such as Growth Mindset; it’s not a universal panacea and the outcomes might just not be what is anticipated / sold to them – all the cognitive biases come into play when we discuss this issue.
    Cheers David.

    • Sadada March 28, 2018 at 11:56 pm - Reply

      Been in the game long enough to have seen Brother Love’s Travelling Salvation Show come to town many times with new and improved snake oil to swallow… I agree evidence is needed…

  11. […] The nail in Growth Mindset’s coffin?, by David Didau […]

  12. Luc Kumps March 16, 2018 at 9:37 am - Reply

    A few thoughts…
    First, remember that sometimes “A little is a lot” ( “there are processes by which individually tiny influences cumulate to produce meaningful outcomes”. Mindset interventions, like reading comprehension interventions, might produce meaningful outcomes in the long term.
    Second, see for Dweck’s thoughts on focusing on effort: “A lot of parents or teachers say praise the effort, not the outcome. I say [that’s] wrong: Praise the effort that led to the outcome or learning progress; tie the praise to it. It’s not just effort, but strategy … so support the student in finding another strategy. Effective teachers who actually have classrooms full of children with a growth mindset are always supporting children’s learning strategies and showing how strategies created that success.
    Students need to know that if they’re stuck, they don’t need just effort. You don’t want them redoubling their efforts with the same ineffective strategies. You want them to know when to ask for help and when to use resources that are available.”
    Third, let’s not forget that mindset theory (and research) also offers insights in what NOT to do. It’s not only about “praising Process instead of Product”, it’s also about NOT praising the “third P”, Person. This was one of the major outcomes of a few decades of research on mindsets. For example, as documented in Dweck’s “Self-Theories” (1999), there’s a major problem with “smart girls” who are being praised both for being “smart” and “good” in elementary school, develop a fixed mindset and show the largest drop during the transition to secondary education. Problems like these aren’t documented in recent research.

    • David Didau March 16, 2018 at 5:07 pm - Reply

      I’ve no problem with attribution theory. As far as I’m aware the findings and predictions are both plausible and well validated. What I struggle with is the magical thinking that suggests that praising (the right kind of) effort increases a growth mindset view of intelligence, whereas praising intelligence promotes a fixed mindset view of intelligence, and further, that mindset predicts students’ grades, cognitive ability, and enhances learning over time. These predictions don’t seem to replicate.

  13. Luc Kumps July 10, 2018 at 12:20 pm - Reply

Constructive feedback is always appreciated

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: