Growth mindset: What interventions might work and what probably won’t?

Whether discussed under the guise of ‘resilience’, ‘grit’ or ‘character’, there appears to be a great appetite for psychologically manipulating pupils’ personalities or their attributions about school. One concept which has particularly captured the imagination of teachers and school leaders is ‘growth mindset’: the idea that children who possess incremental theories of intellect (a growth mindset) appear to achieve better grades than those who possess an entity theory of intellect (a fixed mindset).

The claim that there are attributional differences between pupils which can affect their experience of school and their academic outcomes is well supported. You can read a bit more about some of the psychology behind the idea of a ‘growth mindset’ here: Growth Mindset: It’s not magic

However, accepting that these key attributional variables exist still leaves at least two important questions that school leaders and teachers should be asking before seeking to implement ‘growth mindset’ interventions in schools.

• Firstly, will changing a pupil’s attributions (their attitudes and beliefs) cause desirable changes in behaviour?
It’s possible that the causal arrow between ‘mindset’ and performance is not a straightforward one. It’s natural to assume that changing a person’s beliefs will alter their behaviour, but the evidence on this is much more complicated.

• Secondly, even where experimental psychological interventions are successful, will the implementation of such interventions in schools lead to the desired outcomes?
It’s possible that the elements of a psychological intervention which led it to be successful will be lost or negated when it is scaled at a school level.

Therefore, whilst I’m cautiously optimistic about the promise of research into the psychological and motivational factors which underlie improved experiences and outcomes for pupils, I remain sceptical that current attempts to implement psychological interventions in the UK education system will have the desired effects. At the moment, I suspect they will have very little effect – as so many previous attempts to tackle emotional components of education have had. I’m also concerned that poorly implemented psychological interventions may have a negative effect for some of our pupils.

Can we change behaviour through changing attitudes?

Changing behaviour is hard. Just about everyone I know is trying to change their behaviour in some way; trying to eat more healthily or take more exercise, cutting down on drinking or quitting smoking, being more environmentally friendly by recycling more or using their car less. However, simply because we hold certain beliefs and attitudes (should eat more vegetables, smoking leads to early death, it’s important to protect the environment) doesn’t necessarily mean we successfully change our behaviour.

For example, naïve attempts at encouraging smoking cessation through direct persuasion fails to change people’s smoking behaviour Nutbeam (2000).

“To take a concrete example, efforts to communicate to people the benefits of not smoking, in the absence of a wider set of measures to reinforce and sustain this healthy lifestyle choice, are doomed to failure. A more comprehensive approach is required which explicitly acknowledges social and environmental influences on lifestyle choices and addresses such influences alongside efforts to communicate with people.”

Worse than having no effect, sometimes poorly implemented public health interventions can have negative effects. For example, Glasgow et al (1999):

“Interventions delivered to large populations can also have unanticipated negative effects. Labeling someone with a potential illness may have profound social and psychological consequences.”

Another example of the complexity of the relationship between attitudes and behaviour is using fear to change health behaviour. It seems that attempts to change health behaviour through fear appeals can be very effective, but can also quickly backfire where individuals have low self-efficacy in their ability to avert that threat (Witte and Allen, 2000).

“when a threat is portrayed as and believed to be serious and relevant (e.g., “I’m susceptible to contracting a terrible disease”), individuals become scared. Their fear motivates them to take some sort of action—any action—that will reduce their fear. Perceived efficacy (composed of self-efficacy and response efficacy) determines whether people will become motivated to control the danger of the threat or control their fear about the threat.”

Whilst there’s clearly a relationship – a correlation – between attitudes and behaviour, the relationship is complex and interventions can potentially backfire. The reasons for this are complex and numerous, but Kollmuss and Agyeman (2002) summarise four reasons why attempting to change attitudes fails to lead to changes in behaviour.

“… quantitative research has shown that there is a discrepancy between attitude and behavior. Many researchers have tried to explain this gap.  Rajecki (1982) defined four causes:

Direct versus indirect experience: Direct experiences have a stronger influence on people’s behavior than indirect experiences. In other words, indirect experiences, such as learning about an environmental problem in school as opposed to directly experiencing it (e.g. seeing the dead fish in the river) will lead to weaker correlation between attitude and behavior.

Normative influences: Social norms, cultural traditions, and family customs influence and shape people’s attitudes, e.g. if the dominant culture propagates a lifestyle that is unsustainable, pro-environmental behavior is less likely to occur and the gap between attitude and action will widen.

Temporal discrepancy: Inconsistency in results occur when data collection for attitudes and data collection for the action lie far apart (e.g. after Chernobyl, an overwhelming majority of Swiss people were opposed to nuclear energy; yet a memorandum two years later that put a 10-year halt to building any new nuclear reactors in Switzerland was approved by only a very narrow margin). Temporal discrepancy refers to the fact that people’s attitudes change over time.

Attitude-behavior measurement: Often the measured attitudes are much broader in scope (e.g. Do you care about the environment?) than the measured actions (e.g. Do you recycle?). This leads to large discrepancies in results (Newhouse, 1991).”

Mindset interventions don’t work by trying to browbeat pupils into believing in the merits of hard work or that their ‘brain can grow’. Direct appeals and information alone don’t change behaviour very effectively at all. In fact, effective psychological interventions involve a subtle, well-aimed nudge, which initiates a more complex social process.

How do successful mindset interventions work?

Experimental studies have shown that very brief psychological interventions can lead to long-lasting changes in mindset and effects on pupil performance.

For example: Yeager, Paunesku, Walton and Dweck (2013) provide a review of some of this evidence. They relate students solving maths problems more successfully, months after being presented with a short mindset encouragement at the top of a computer screen compared to randomly allocated controls. Another experiment found that a single lesson on growth mindset over the internet reduced the failure rate by low-achieving pupils by 7%. A web-based mindset intervention also improved freshman completion rates by 3-4% (10% improvement amongst African-American students).

Some of these psychological interventions, lasting for only a few minutes, have been shown to not only improve school performance but also in other social contexts like encouraging voting participation. Any school leader or teacher thinking of implementing a growth mindset intervention would do well to read some of the work by Yeager and Walton – for example:

• Walton (2014) The new science of wise psychological interventions
• Yeager, Walton, and Cohen (2013) Addressing achievement gaps with psychological interventions
• Yeager and Walton (2011) Social-Psychological Interventions in Education : They’re Not Magic

Both Walton and Yeager identify some key components to successful mindset interventions: Psychological insight and precise targeting of a brief and stealthy intervention; and utilising recursive processes, essentially triggering a virtuous circle which supports the original intervention. These two components appear to be lacking in many school initiatives to exploit ‘growth mindset’ research, I contend.

‘Theory guided precision’

“In the spirit of Kurt Lewin (“There is nothing so practical as a good theory”), creators of “wise interventions” leverage specific psychological insights.”

Rather than being a generic appeal, successful psychological interventions tend to be highly specific – crafted to the precise psychological process being manipulated:

“A wise intervention begins with a specific, well-founded psychological theory. This theoretical precision allows researchers to create a precise tool, often instantiated in a brief exercise, to change a specific psychological process in a real-world setting. This psychological precision reflects the same values psychologists cultivate in laboratory research—keen insight into basic processes and methodological precision to isolate these processes. Wise interventions export this precision in theory and methodology to field settings.”

The intervention methods come from a solid understanding of the psychology of social influence and persuasion. For example, for growth mindset interventions:

“Rather than simply presenting an appeal to a student, each intervention enlisted students to actively generate the intervention itself. For instance, one delivery mechanism involves asking students to write letters to younger students advocating for the intervention message (e.g., “Tell a younger student why the brain can grow”). As research on the “saying-is-believing” effect shows, generating and advocating a persuasive message to a receptive audience is a powerful means of persuasion (Aronson, 1999).”

By targeting a psychological process in such a specific way, these interventions use ‘stealthy’ and brief delivery mechanisms that quickly change students’ beliefs. But, they’re not ‘magic’:

“They are not worksheets or phrases that will universally or automatically raise grades. Psychological interventions will help students only when they are delivered in ways that change how students think and feel in school, and when student performance suffers in part from psychological factors rather than entirely from other problems like poverty or neighbourhood trauma.”

This isn’t a non-specialist role, according to Yeager and Walton. They suggest that we need a new class of professional psychologist to scale the impact of social-psychological interventions in schools:

“Along similar lines, it may be useful to revisit past suggestions for creating a new class of professional—a “psychological engineer”—a person with the expertise needed to scale psychological interventions effectively. Such professionals would be trained in experimental methodology and psychological theory, although their primary work would be not to advance psychological theory but to understand and alter psychological dynamics in applied settings.”

Essentially, psychological interventions aren’t suited to generic attempts at amateur psychology. The people claiming to demonstrate some profoundly successful interventions suggest a level of expertise is involved; that to be successful, individuals designing and delivering an intervention require significant understanding of the psychological theories involved. In addition, they should not be seen as a panacea – they cannot, on their own, overcome significant problems caused by socio-economic deprivation.

‘Recursive processes’

Successful experiments in social-psychological interventions need to be stealthy and brief, according to Yeager and Walton:

“Often psychological interventions are brief — not extensive or repeated. Excessive repetition risks sending the message those students are seen as needing help or may undermine the credibility of a reassuring message (as in “thou doth protest too much”). In this way, delivering psychological interventions differs markedly from teaching academic content. Academic content is complex and taught layer on layer: The more math students are taught, the more math they learn. Changing students’ psychology, by contrast, can call for a light touch”

Thus, frequent repetition of ‘growth mindset’ messages through lessons or tutorials firstly doesn’t boost the effectiveness of the intervention, and secondly may actively undermine it. This is such a counter-intuitive point; I’m not surprised that it’s overlooked by teachers and school leaders. It’s also, I suspect, likely not in anyone’s commercial interests to make this point!

So, successful interventions are subtle and brief, so how do they have such apparently large effects on student performance? Walton explains:

“To understand them, it is essential that one consider how interventions change not a moment in time (“a snapshot”) but a process that unfolds over time (“a movie”; Kenthirarajah & Walton, 2013). In a relationship, every interaction builds on the previous interaction. By targeting psychological processes that contribute to recursive dynamics that compound with time, wise interventions can improve downstream consequence.”

To give an example, Walton and Cohen (2011) relate an experiment seeking to reinforce the sense of belonging for freshmen at college. They hypothesised that African-American students would particularly benefit from the 1 hour, social-psychological intervention to reduce the threat perceived by social adversity.

“The intervention gives students an alternative narrative for understanding negative experiences—namely that worries about belonging are normal in the transition to a new school but dissipate with time.” …

“How could this work? Imagine you are a freshman worried about whether you belong in college. Learning that such worries are common and improve with time may take the edge off negative experiences. … minority students in the intervention condition no longer saw daily slights as if they portended a global lack of belonging; this change in social construal mediated the 3-year improvement in grades. If everyday encounters feel less threatening, perhaps students can interact with others in more positive ways and build better relationships.”

Essentially, social-psychological interventions utilise the ‘self-fulfilling prophecy’:

“Wise interventions harness the power of self-fulfilling beliefs. … Believing that change is possible with effort — “When you learn a new kind of math problem, you grow your math brain!” — students may experience greater success, which discounts the sense they aren’t “gifted” at math and strengthens their self-efficacy.”

It is the experience of success which comes with effort which feeds into a student’s perception. The purpose of a ‘growth mindset’ intervention is a subtle ‘nudge’ which promotes the behaviour which is more likely to achieve that success. It is not motivational quotes, inspirational stories of ‘growth mindset’ heroes, students post-it notes on a growth mindset wall, growth mindset lesson objectives, roleplaying a TV show about overcoming a fixed mindset or other kinds of ‘rah-rah boosterism’:

“Bolstering a sense of belonging for poor-performing students requires establishing credible norms that worry about belonging are common and tend to fade with time — not rah-rah boosterism.” …

“Good teachers often know the importance of belonging, growth, and positive affirmation. But they may not know the best ways to bring these about. Well-intended practices can sometimes even do more harm than good.”

A successful psychological intervention involves a quick, well-targeted ‘nudge’; not repeatedly hitting students over the head with a sledgehammer!

Implications for ‘Growth mindset’ interventions in schools

Perhaps the key worry I have about the trend for implementing psychological interventions in schools is that it is easier to appear to be doing something than to actually do something well. Motivational posters and assemblies, or explicit lessons about the importance or benefits of holding a growth mindset may inadvertently have a negative influence on pupil attitudes and, even if they don’t, are unlikely to have the desired effect on pupils’ behaviour.

In the absence of an army of ‘psychological engineers’ to implement interventions in schools for us, what might work as interventions in schools to encourage growth mindset? Here are 5 suggestions that might help.

1) Focus on students achieving success, rather than tackling their motivation.

In ‘What makes great teaching’, Rob Coe and his team report that trying to tackle motivation alone has very little effect on student progress:

“Address issues of confidence and low aspirations before you try to teach content

Teachers who are confronted with the poor motivation and confidence of low attaining students may interpret this as the cause of their low attainment and assume that it is both necessary and possible to address their motivation before attempting to teach them new material. In fact, the evidence shows that attempts to enhance motivation in this way are unlikely to achieve that end. Even if they do, the impact on subsequent learning is close to zero (Gorard, See & Davies, 2012). In fact the poor motivation of low attainers is a logical response to repeated failure. Start getting them to succeed and their motivation and confidence should increase.”

It is vital to remember that it is the experience of success which leads to long-lasting change in attitudes to school. Even where attitudes are changed, it will have little long-term effect on behaviour unless the pupil enters that recursive, virtuous cycle of success. What makes a mindset intervention successful isn’t magical. It is a subtle nudge which encourages pupils to behave in ways which are more likely to achieve success. However, where pupils do not see that success, the efforts will be undermined or even have a negative influence (e.g. apparently confirming lack of intelligence).

2) Focus students upon the strategies they use

Whilst an absence of effort pretty much guarantees failure, ‘more effort’ on its own is not a guarantee of success. One positive development is that some schools are shifting the focus away from ‘praising effort’ to a more thoughtful approach involving developing students’ metacognition. In a recent blog by John Tomsett, he relates the development of learning tools – a range of strategies which pupils can select when they don’t meet success.

There are likely to be some common and maladaptive strategies which students employ. For example, students adopting avoidance strategies when they become anxious; e.g. behaving badly, truancy from a lesson or procrastinating when revising. However, my suspicion is that ‘learning tools’ will only be adaptive where they are domain specific – e.g. a range of particular strategies which students can use when they ‘get stuck’ with maths problems. The issue with a ‘generic skills’ approach to learning (e.g. developing dispositions, or learning to learn skills) is that they tend not to transfer between contexts.

3) Evaluate change in behaviour rather than attitudes

The use of surveys is a common way of trying to establish whether a pupil possesses a growth or a fixed mindset. We need to be very wary of these as measurements of impact. School mindset interventions which rely upon explicit mindset messages may temporarily alter student attitudes to their learning without actually changing their behaviour in the classroom or outside of school. Worse still, reliance upon ‘inspirational’ messages or explicit teaching of mindset may simply tell pupils the socially desirable response expected in surveys – giving the appearance of changing attitudes without genuinely changing the attitudes that pupils possess. This would render any attempt to measure ‘impact’ through – for instance – student surveys potentially meaningless.

Successful interventions will look at behavioural changes – rather than effects on attitudes. For example, one possible quantitative measure of impact might be to quietly measure specific students’ ‘time on task’ in lessons. Qualitatively, one might measure changes in effort through analysis of students’ work in books.

4) Focus on the normative influences within the school culture

Interventions need to consider the broader normative influences operating within a particular school context and for a particular child within that context. Norms need to de-emphasise the negative consequences of making mistakes and discourage social comparisons. There’s probably little point in ‘preaching’ a growth mindset within a broader school context which explicitly emphasises performance-orientated structures or goals. (Of course, it will be almost impossible to communicate this effectively to pupils if teachers in a school are subject to high-stakes accountability systems which do not embody the same values!)

For example, there’s the danger that even a successful attempt to alter pupils’ theory of intellect will be significantly undermined by the pupils’ experience of being relegated to a ‘bottom set’ or being given an artificial target of an ‘A’ that they repeatedly fail to reach.

5) Consider teachers’ implicit theories of intellect.

For an incremental theory of ability to become an unspoken social norm it would be useful to consider the attitudes and beliefs about learning which are held by teachers. For example, the ‘What makes great teaching’ report suggests that highly effective maths teachers tend to have some common beliefs about learning:

“How children learn:
• almost all pupils are able to become numerate
• pupils develop strategies and networks of ideas by being challenged to think, through explaining, listening and problem solving.

They used teaching approaches that:
• ensured that all pupils were being challenged and stretched, not just those who were more able
• built upon pupils’ own mental strategies for calculating, and helped them to become more efficient.”

Teachers unconsciously communicate their attitudes and beliefs about intellect and learning when they interact with pupils. To what extent do teachers hold ‘growth mindset’ beliefs about their pupils? Where teachers privately hold entity theories about ability in their subject, they are likely to communicate these to pupils (despite giving the ‘socially desirable’ response when asked in surveys).

Challenging the profession to learn more about the nature and nurture of intelligence and concepts like neuroplasticity may help them believe that their pupils can succeed with the right strategies and some effort. The purpose of this isn’t compliance with ‘acceptable beliefs’ but a genuine engagement with learning a bit more about the psychology of how children learn.

This won’t be easy for all teachers – we come from such a wide variety of disciplines – but actually, perhaps experiencing difficulty in learning some genuine neuro-cognitive psychology (rather than usual dumbed-down stuff teachers get) will also help model how effective learners behave to their pupils?

Advertisement