Is resilience even a thing?

/, psychology/Is resilience even a thing?

There is but an inch of difference between the cushioned chamber and the padded cell.

G. K. Chesterton

Resilience – being able to bounce back from setbacks and cope with challenges – seems an obviously good thing. If we can make ourselves, and our children, more resilient, then we definitely should. Trouble is, it doesn’t seem we can.

In 1907, William James – often dubbed the grandfather of modern psychology wrote the following in an article for the journal Science:

Compared with what we ought to be, we are only half awake. Our fires are damped, our drafts are checked. We are making use of only a small part of our possible mental resources. . .men the world over possess amounts of resource, which only exceptional individuals push to their extremes of use.

In her 2007 paper, Angela Duckworth addressed James’s challenge to work out how best to unleash humanity’s potential. In her own words, “Why do some individuals accomplish more than others of equal intelligence?”

It’s a good question. Lots of factors matter when predicting an individual’s success. For instance, according to one study, the correlation of socioeconomic status with academic achievement is r=0.4 and general knowledge comes in at a whopping r=0.81. Interestingly, this is identical to the effect of IQ on GCSE results. Ian Deary, one the world’s leading intelligence researchers, found the correlation between IQ tests taken five years previously and GCSE results taken at age 16 is also r=0.81.* But is there something else, something missing from these analyses?

Clearly, the finding that intelligence correlates with educational achievement does not predict the performance of an individual. It’s perfectly possible, indeed commonplace, for two individuals with the same IQ score to do very differently in a set of exams. There could be all sorts of prosaic reasons for these differences: health, injury, a never-ending list of possible personal crises. But that doesn’t tell us anything useful about how to support children to do better at school or in life. What we need is a thing. A thing we can measure and teach lessons on. Once we’ve identified a thing, we no longer have to wonder about the unpredictable, capricious nature of the universe.

Duckworth says that while there are all sorts of potential things from which to select, “some traits might be essential to success no matter the domain. We suggest that one personal quality is shared by the most prominent leaders in every field: grit.” Great, so what is grit?

Duckworth defines it as “perseverance and passion for long-term goals. Grit entails working strenuously toward challenges, maintaining effort and interest over years despite failure, adversity, and plateaus in progress.” Sounds an awful lot like resilience, doesn’t it? This is one of the problems when attempting to review the literature. Some psychologists are researching self-efficacy, others resilience, yet others conscientiousness. And now we have grit too. Are they distinct, psychological traits? And most importantly, can we measure how much they matter?

In the case of grit Duckworth has suggested it’s a factor underlying success for groups as disparate as Ivy League undergraduates, West Point military cadets, elite athletes and National Spelling Bee competitors. It’s exciting to believe that the achievement of difficult goals in life depends not upon talent alone, but ‘a sustained and focused application of talent over time’, but, in the case of the West Point cadets, grit didn’t turn out to account for all that much, increasing the probability of the cadets successfully completing their training from 97% to 99%.

Despite the enthusiasm, a recent meta-analysis has cast doubts about whether the concept stands up to scrutiny. The research looked at results of studies from 88 different samples involving over 66,000 individuals. They appeared to find that grit was only modestly correlated (0.18) with performance and strongly correlated to the personality trait of conscientiousness. Given these problems, can we have any confidence that interventions or initiatives based on the idea of developing grit will have an effect on student outcomes? Certainly, there’s no robust evidence to suggest so. Even Duckworth has admitted her findings of the independent impact of grit are what personality psychologists would place in the ‘small-to-medium’ range.

One of the common issues is that evaluations of programmes designed to develop character are often intensely subjective. Many of the glowing, positive claims come from evaluations carried out by those who were implementing the initiative in the first place! We need to remove confirmation bias, expectancy effects and plain wishful thinking to provide anything genuinely helpful for our students. Where more robust evaluation has been conducted, interventions to develop character traits have not always demonstrated long-term positive outcomes for students.

For example, a report in 2011 for the Department for Education evaluated the UK Resilience Programme. The programme, based on the Penn Resiliency Program (a curriculum developed by a team of psychologists at the University of Pennsylvania) attempted to improve children’s psychological well-being through resilience and the promotion of realistic thinking and coping skills. Cognitive behavioural therapy was an explicit part of the training. Unfortunately, the studies met with limited success: the one-year follow-up found a small average impact on pupils’ depression scores, school attendance, and English and maths grades However, even this modest effect had disappeared at the two-year follow-up, suggesting that, on average, pupils who had participated in resilience workshops were doing no better on these outcomes than pupils who had not.

For the majority of school-based character education interventions, we know even less about what their impact has been. It could be that schools have wasted teaching time and added to teachers’ workloads to no discernible effect. But this isn’t the worst possible outcome. Professor of education, Kathryn Ecclestone suggests that there may be a darker side to this well-intentioned focus on character. She suggests we’d be better off focusing on “a stimulating, enriched, challenging curriculum and extra-curricular activities” to help students develop the resilience we’d like them to have. A recent article by Paul Tough seems to share this view. His research led him to question whether qualities, like grit, curiosity, self-control, optimism, or conscientiousness actually exist as skills which can be taught. None of those educators he interviewed who appeared to be highly successful at engendering these traits gave pep-talks or motivational speeches to students.  In fact most made no mention of these ‘skills’ at all. What appeared to work was talking to the child and giving detailed feedback about the mistakes they had made and what they could have done differently.

And this leads me to the conclusion that resilience (or whatever you want to call it) doesn’t really exist as a teachable skill. I’m not saying that people don’t display a quality of resilience in their lives, just that we probably couldn’t teach it even if we knew what it was. This paper by André Tricot and John Sweller presents compelling evidence that there are no teachable generic skills. They say, “all educationally relevant knowledge acquired during instruction is, and only is, domain-specific.” That is, creativity, critical thinking, problem solving, resilience and whatever else you care to mention are evolutionary adaptations that we pick up from the environment as very young children, and, to be meaningful in school and later, they rely on hard-won, biologically secondary knowledge.

This interpretation is consistent with logic. A friend of mine recently told me that his 11-year-old son wanted to give up playing rugby, but that he was unwilling to let him quit as he felt that rugby would teach him to be resilient. I pointed out that the skillset required to be  resilient in rugby was unlikely to lead to resilience in, say, maths. In order to stick with tricky maths problems you need to know quite a lot of maths. The experience of holding fast in a scrimmage is unlikely to be of much use. I also pointed out that we all quit stuff all the time. We make choices about what we want to invest our time and energy in, and these are, usually, based on what we’re good at. Just because I gave up being a Cub Scout after a few weeks didn’t predict that I would drop out of university or be unable to hold down a job. I just didn’t like Cubs.

And that’s my conclusion. People are no more resilient than they are lazy. No one is lazy at doing what they most enjoy and no one – or vanishing few people – resiliently stick at stuff they think is pointless and stupid. As F. Scott Fitzgerald put it, “Vitality shows not only in the ability to persist, but in the ability to start over.” We make calculations about how much effort to invest based on our feelings, the approval of those around us, the carrots and sticks waved at us and a multitude of other imponderables.

Of course, we don’t want children to give up when things get tough. We all probably need to be a bit better at sucking it up and getting on with chores.  I love, Housman’s advice: “Shoulder the sky, my lad, and drink your ale.” But still, I tend to think we need to offer more than ale. My suggestion is simply this: if you want someone to be resilient, first ask, at what? Then help them improve their performance at whatever it is: getting better at something is the best way to improve motivation.

* If you need a reminder about how correlation works, take a look at this post.

2018-01-26T22:25:42+00:00May 3rd, 2017|Featured, psychology|


  1. MarySue May 3, 2017 at 3:56 pm - Reply

    I feel I have to disagree, some students are naturally able to cope with that uncomfortable, stuck, “I can’t do it, I got it wrong” feeling, and others find it incredibly difficult. But if, every time a child feels that horrible, gut wrenching, “I just can’t” feeling, they get a consistent “but this is what learning feels like – remember when you couldn’t…. and we insisted you do it over and over until you could, and now you really enjoy…” then they do learn that this is a truth – that this is what it feels like to learn stuff and that this is a necessary precursor to enjoying any activity.
    I have an autistic daughter, for whom not being able to “get it right” is enormously difficult. Every activity she does I have had to go through this same process – from riding a bike to learning the piano to body boarding on holiday – stage 1 – “No, I hate it, I just don’t like doing it” – stage 2 – “well your going to do it over and over X times because mummy says you have to – stage 3 “Gee, this is brilliant, look at me doing this”.
    At 13 she is finally getting it, and I can see and talk to her about the feeling of discomfort, and she knows that while she hates it she will only achieve what she wants if she goes through it to the other side.
    Kids can “learn” this, can “improve their skill” in it – but I do agree that it can’t be done outside of learning a specific skill – it does require that every time they are going through the process of learning that this process is pointed out, made explicit, and compared to other processes of learning in other areas where they have succeeded after continuing through that stuck phase. And it takes a long time of having that success after stuckness pointed out and reinforced consistently.

    • David Didau May 3, 2017 at 4:25 pm - Reply

      Yes, children are naturally resilient. If this wasn’t an evolved adaptation we probably wouldn’t have survived as a species. And, like any human trait there will be a normal distribution of ability.

      Sadly, there’s zero evidence that parenting makes any difference to how children turn out; the reason children are similar to parents is due to heritability.

      Of course, kids *can* learn resilience as I laid – you seem to agree with this.

      • maeysue May 3, 2017 at 7:34 pm - Reply

        Really??? Zero evidence??? I’m getting old and outdated these days, but back in the day I did a genetics based degree and I was under the impression evidence pointed to it being about 50/50 nature/nurture? If parenting makes no difference then surely teaching, which a child spends far less time exposed to, makes no difference either? And why is there such an obvious difference between pp and non pp, and students in the care system do so badly? Is that due to poor genetics??? I’m not entirely certain but zero difference seems unlikely

        But yes, I think you are arguing a different side of the same coin – I totally agree that these inherent abilities need to be encouraged through teaching of activities, there’s no way to teach them as such independent of learning a skill or fact

        • David Didau May 4, 2017 at 9:28 am - Reply

          Nurture’s the wrong word. It’s 50% heritability, 50% environment. Environmental effects are broken into shared and non-shared environment. Non-shared environment accounts for pretty much all of that effect. The reason teachers can and do have a more significant impact than parents is explained here

          • marysue May 4, 2017 at 1:45 pm

            Quite right of course – I was going for common language rather than scientific accuracy – thanks for picking me up on that.

            I’m still looking for evidence either way as to effects on long term outcomes for shared vs non-shared – I maintain that zero effect for shared environment on how kids turn out can’t be true. I’m totally with you (and evidence does clearly support) that personality traits are not effected at all (my son will always be easy going and chilled, and my daughter will always be very high stress) – this, however, isn’t an outcome.

            i would be very surprised indeed to find that the evidence suggests that if, as a parent, I go out to the pub every night and never read to my kids or mention how they do at school other than to laugh at them if teachers tell me they are doing well – you are telling me this makes ZERO difference to if I spend time playing with them, reading to them and encouraging them to believe that education is a worthwhile pursuit? It seems to me you’re saying the same child in the former environment would do better at a great school than the child in the latter environment in a totally rubbish school?

            I’m totally with you on the grammar schools – I continue to read your blog above all others because I appreciate what you say, how you say it, and why you say it! You’re one of few that really make me think and change the way I see things – totally prepared to accept you’re right, but to make a statement that sweeping, and that far off what seems logical, you’re going to have to back it up with some pretty firm evidence for me 🙂

          • Pique Boo May 4, 2017 at 9:27 pm

            I’m still looking for decent descriptions of the division between shared and nonshared. I think a more helpful way of thinking about those categories is shared experience vs. unique experience. But that’s too simplistic because nonshared is everything that is not genes and shared, so it also includes experimental errors, noise, chance, kitchen sink and quite a few holy grails of education.

            Shared experience is not all aspects of parenting twins and can includes aspects of ‘school & peers’. If Thing1 and Thing2 go to the same school in the same class with the same peers, then what parts of their school day are shared or nonshared?

            Unique experience is not all aspects of schools & peers and can include aspects of parenting. Getting your twins to read to you in turn (as opposed to in unison at the same time) might be unique experience for each, and one that could escalate into a significant future difference. Perhaps Thing2 got a more difficult word on their page one day and that was the when they decided that they were rubbish at reading because Thing1 didn’t struggle earlier when they read their pages.

            Average genetic influences on [insert trait] is useful, but not obviously for any given individual and reducing parenting to ‘basic food and shelter’ seems like a leap too far right now. I could also spend quite a while rattling off bits of parenting that have set my 13yo daughter on this or that fork in the path to adult. Just one of those was ‘musician or mathematician’ and although they would both have roughly the same IQ and lower score on the extraversion scale, to most people they look like quite different futures.

          • Marysue May 5, 2017 at 10:20 pm

            Book looks most interesting, will have to give it a go. First thought is that as a child of graduates going to a very duff state school, I chose my peers as other graduate children and we all went to uni and did well, where the children of those in trades hung out together and went on to apprenticeships – the largest effect I saw on choice of peer group was parental culture – this I’ve seen repeated over 15 years teaching. What makes a child choose to be part of an in group or an out group? Surely your peer group is fairly strongly predicted by your parenting culture – although this can be used to argue for heritability!!! I’ll read it and get back! Thanks for making me think outside my comfort zone – always most appreciated!

        • praxisenglish May 4, 2017 at 9:12 pm - Reply

          How can there be zero evidence? Zero? Surely it makes logical sense that parenting has an impact.

          • David Didau May 4, 2017 at 9:31 pm

            Well, it has indirect influences but what seems logical is the effect of heritability

      • marysue May 3, 2017 at 8:56 pm - Reply

        Could you give any studies to back up the zero evidence comment? Quick google search very much disagrees! See – seemed fairly through. What they struggle to find evidence for is that interventions to attempt to improve parenting have any educational impact.

        • Michael Pye May 5, 2017 at 6:00 pm - Reply

          Marysue he can’t give any studies to back up the zero evidence, that not how it works.

          The sentence David used was “Sadly, there’s zero evidence that parenting makes any difference to how children turn out;” this is saying that he doesn’t know of any research that supports the idea. (There may be a distinction between quality of evidence here as there are often poorly designed studies supporting all-sorts of craziness). You discovered this with your line “what they struggle to find evidence for is that interventions to attempt to improve parenting have any educational impact.”

          I to find it hard to accept parenting does not have a profound effect but this is an important distinction. By the way zero effect means zero measurable effect which is different. You are in danger of the backfire effect by focusing on slight details rather then the key arguments and by flipping the question to demand refutation of a counter argument.


          Parenting interventions don’t improve outcomes
          How can parenting not effect outcomes?

          Notice how the second is begging the question.

          (When I re-read Davids post I noticed he did not specifically say that parenting is not important- remember no one writes their arguments perfectly you need to consider them in their strongest form before reviewing them ).

          We all react to to opposing assertions as attacks on our core ideas but if we can instead imagine them as principles and consider what a world would look like if that was the case. Does this make sense? Does it match what I see? What other effects would we expect? If this is true what would it mean?What kind of unintended consequences would arise? What would it require for me to change my mind on this?

          For context I am not yet convinced by Davids arguments so don’t feel I am asking you to agree with him.

          P.s Is your name a cool Star-trek/sci-fy reference?

          • David Didau May 5, 2017 at 7:13 pm

            1. They’re not my argument’s I too needed to be convinced. This is the case made by Judith Rich Harris in The Nurture Assumption. Here’s a pdf of the book:
            2. As discussed, parenting does indirectly influence peer culture; when family values don’t run counter to peer group values they continue to exert residual influence.
            3. Parenting *is* important. It makes a great deal of different to a) how children behave in the family home, b) how happy children’s childhoods are, and c) parents’ relationships with their children in adulthood.

          • Michael Pye May 5, 2017 at 9:49 pm

            I have the link bookmarked for Judiths Book. Sorry I thought you were largely convinced.

          • David Didau May 6, 2017 at 12:42 am

            I am. She convinced me.

          • Marysue May 5, 2017 at 10:33 pm

            Thanks for the feedback Michael, much appreciated. Like my daughter, I too am aspergers – I gave up all debating many years ago, long before social media etc., as I found my lack of comprehending ways of putting my point across irritated people, and for the sake of learning social skills gave up! I figure it’s time to relearn an old skill I quite enjoyed, but help in getting messages across gratefully received!

            I read David’s blog because he seems happy to accept debate and be swayed by others, as am I – never really convinced myself if any ideas I have are good/bad, right/wrong! But I like to know where the ideas come from – the book looks great.

            I felt I was pointing out a fairly important issue in his idea – I think resilience CAN be taught, because I THINK I’ve taught it to my daughter – I think the kids that do well in school seem to do so because they have learned resilience at home and successfully transferred it into a new setting. I’d like to think as a teacher I can enhance this massively important skill in my students. Sadly I don’t feel like I’ve ever had any huge success at this so I do kind of want to be convinced by David, as that would let me off the hook nicely!

            I’m interested to find out what sci fi reference you’re thinking of! Mum was Zimbabwean, like Americans they often give girls double barrelled names!

          • Pique Boo May 6, 2017 at 11:55 am

            “She convinced me.”

            JRH is convincing. A persuasive writer worth reading. I think the essential problem is that she throws in a lot of stuff, hits many targets you want shot down in flames and you overlook some parts that don’t quite hang together. There are some devils in the detail, for instance:

            How do you measure a personality trait?

            One approach is self-report, but there is a compelling argument that personality is what other people observe, not what you think you are, or perhaps want to be and try to fake etc. This premise has led to research papers where they used multiple peer reports of the subject’s personality to decrease personality measurement error. One of those from the end of last century had Big Five personality trait heritability at roughly 80%. Another similar purpose paper from 2010 had heritability at roughly 85%, with ‘Openess’ the highest at 92%.

            None of this is my territory so I’m open to correction, but if those results are reliable then I think the idea of parents, peers, teachers or Fluffy the pet cat significantly influencing personality is running a bit short of space given that the residue still needs to accommodate more than environment. I suppose it is still feasible courtesy of the likely distributions of heritability behind the headline figures, but it seems much less feasible. On average.

  2. Michael Pye May 3, 2017 at 6:28 pm - Reply

    That article by Paul Taugh is interesting. He referenced C. Kirabo Jackson but offered little detail about how certain teachers developed these traits indirectly. It claimed these were different teachers then those that developed students test scores. Are you familiar with Jackson work as it is seems central to a lot of your ideas, potentially supporting and undermining them. You could do a whole post just on that Atlantic article as it has a lot of interesting ideas that are often antagonistic.

  3. Maxine Davies May 4, 2017 at 8:12 pm - Reply

    If environment is not as crucial as we think, why then are factors like poverty, parental education, trauma & attachment so important for and to learning and development? Teachers work with the results of these things and they can have an utterly detrimental impact on learning. The works of Alan Schore spring to mind. As does the long term research undertaken on children in Minnesota that showed how important attachment was to educational success.

    • David Didau May 4, 2017 at 9:30 pm - Reply

      Who says environment isn’t important? Certainly not me.

      • Michael pye May 6, 2017 at 2:11 pm - Reply

        Google marysue possibly with star trek or scy-fy. It was originally a reference to a short bit of fan fiction in which the lead character was obviously a proxy for the author and was unbelievably over important to the plot. It is now used to suggest a character is overpowered and therefore unrealistic and uninteresting. There are frequent arguments other weather a character qualifies.

        • marysue? May 6, 2017 at 5:57 pm - Reply

          Ha, oh lord that’s hilarious! I can’t believe that as a nerdy trekkie kid growing up in the 70s this was never bought to my attention – especial as I was that irritation know it all teenager who loved to be better at everything than everybody – one reason I learned to keep my mouth firmly shut as an adult!!! Going to be giggling about that for ages – slightly embarrassing though – so do I go on with the name as badge of pride or give it up in embarrassment though!!!

          • Michael Pye May 7, 2017 at 9:34 am

            Keep it.

  4. Alison Honeybone May 6, 2017 at 3:13 pm - Reply

    I think the point about resilience varying between situations has so much relevance to education. I know this is anecdotal, but I work with young people in an FE college and they frequently show huge resilience in the paid jobs which they are juggling around their studies. They can work ridiculous hours in jobs in the service economy and then they come into college and get into trouble because they are not showing much resilience about studying Shakespeare for their fourth re-sit of GCSE English. I firmly believe that they should be doing the Shakespeare, but I can’t say that their lack of commitment to his works shows a general lack of resilience which is fundamental to their character. Not when they’ve done 25 hours in a service industry job that week already. They don’t need to be more resilient in general – they need somehow to become more resilient around the hard, unpaid slog of challenging texts. Resilience developed in the paid workplace doesn’t always transfer to academic work…and I can’t say I’m really surprised.

    • Michael Pye May 6, 2017 at 9:22 pm - Reply

      Alison your post made me think of research on willpower. I remember reading about how willpower is in-fact a finite resource. Use of it earlier in the day increases the difficulty later on, though in general those who practice it gain greater amounts over time similar to exercise. It struck me that resilience and willpower may be useful proxies for each other. Can anyone think of any differences between the two concepts or are they effectively the same?

      • David Didau May 7, 2017 at 8:35 am - Reply

        The research on ego depletion is highly contested

        • Michael Pye May 7, 2017 at 9:35 am - Reply

          Cheers. Do you think willpower and resilience are basically the same or different? If not what are the key differences?

          • David Didau May 7, 2017 at 9:58 am

            It’s a good question. Psychologists regularly talk past each other because there’s more mileage in inventing your own construct than in researching someone else’s. It’s probably most useful to talk about conscientiousness as it’s more stable and has better predictive power.

  5. Michael Pye May 8, 2017 at 10:35 am - Reply

    How would you describe conscientiousness? I noticed you using this word in earlier blogs and was a bit confused until i realised I had misunderstood its meaning.

    Google’s top answer was

    Conscientiousness is the personality trait of being careful, or vigilant. Conscientiousness implies a desire to do a task well. Conscientious people are efficient and organized as opposed to easy-going and disorderly.

    I am guessing this is pretty accurate in the context you are using it.

    • David Didau May 8, 2017 at 1:28 pm - Reply

      Yes. Conscientiousness is one of the ‘big 5’ personality traits. The others are Openness, Extraversion/introversion, Agreeableness and Neuroticism. Together they form the acronym OCEAN. I’m a bit sceptical about how stable these constructs and whether they truly represent people’s personalities, But conscientiousness is far more stable and better researched than any of its variants.

  6. Joe Eden May 10, 2017 at 7:19 pm - Reply

    What do you think of Walter Mischel’s research on self-control? He is reasonably optimistic that self-control can be built-up and learned as a skill, and his longitudinal research suggested it correlates with lots of positive outcomes across a range of domains within the lives of those in the study. If he’s right, that’s a generic skill that can be taught.

    • Rob hitch August 26, 2017 at 3:52 pm - Reply

      Really enjoyed this blog.

      I think resilience is achieved by incremental success, and I think can apply to connected fields or domains, (or at least fields we perceive as connected) not universally.

      I’m pretty resilient in most situations, but give me anything I class as DIY and I give up quickly!

      Do you think schools are fighting a losing battle tried to find ways to foster resilience? Is it just a matter of getting better at generating and recognising that incremental achievement?

      Get Outlook for iOS

  7. […] The antipathy of very many otherwise sensible people to the concept of intelligence is really quite remarkable. This aversion seems only to be increased by bringing up the subject of IQ tests. The idea that IQ tests are only useful for showing how good some people are at taking IQ tests is a deeply ignorant view based upon a breathtaking piece of intellectual dishonesty. It’s difficult to believe that people like Professor Guy Claxton who really should know better, are willing to repeat utter garbage like, Angela Duckworth’s construct ‘grit’ is a better predictor of academic success than IQ. This is entirely false. The correlation between ‘grit’ and academic acheivement is only 0.18 whereas an IQ test taken at the age of 11 produces a massive correlation (0.81) with GCSE results. (See here) […]

Constructive feedback is always appreciated

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

%d bloggers like this: