Behaviourally-Informed Colleges: Translating research into practice

In the dynamic college environment, everyone can feel on their own treadmill.  But whether you're a teacher striving for top grades for your physics students, a pastoral leader addressing absenteeism, or a senior leader organising the academic calendar, one thing we all share is that our core business is in some form of behaviour change. 

We all aspire to work smarter, not harder, and to ensure the best outcomes for our students. However, this goal is complex—students and teachers alike can fall into the traps of being unmotivated, confused, emotional, and lacking self-control, especially at the most stressful times of year. Understanding and addressing the underlying causes of our behaviours is crucial yet challenging.

Insights from Academia

Teenagers are night owls. This is not just anecdotal: research has found they go to bed late and sleep in compared to adults. Despite this, most colleges start their day before the working world does. Why? There may be historical, logistical, and cultural reasons for this schedule, but would it not be better if we instead asked what timetable would improve attendance and alertness, considering students' behavioural tendencies? This is an example of the kind of questions schools and colleges adopting a behaviourally informed approach would be asking.

Making Colleges Behaviourally- and Learner-Informed

To make colleges more behaviourally- and learner-informed, we can leverage insights from over forty years of academic research. Adolescents make complex educational decisions while their brains are still developing, and we are all beset by myriad cognitive biases. These biases often lead to sub-optimal decisions; we prefer immediate rewards over future benefits, avoid risks and fear losses, neglect non-salient information, struggle with multi-step problems and extensive choices, and follow maladaptive social norms that impede learning.

Although substantial progress has been made in identifying these biases and establishing behavioural science as a field, a gap exists in translating this research into practical applications in colleges and classrooms. The development and rigorous analysis of theories should be the challenging part, while implementation should be straightforward. Why, then, have we neglected this critical aspect of evidence-informed education?

Challenges in Translating Research

Dissemination: Significant strides have been made to bridge the gap between academic research and educational practice, largely through the efforts of organisations like the Education Endowment Foundationand Evidence-Based Education. These bodies work to synthesise and translate educational research into practical toolkits and frameworks. However, this task is complex. For instance, a lack of theoretical consensus on the definition of metacognition makes translating research on metacognition into classroom practice challenging for colleges. Simplifying research can lose nuance, while retaining technical terms can confuse. Ultimately, there are no perfect solutions, only trade-offs.

Context, context, context: Education research often aims for external validity—findings that can be applied across various contexts. However, achieving this is notoriously difficult. Laboratories and controlled experiments ensure high internal validity, but classrooms are dynamic environments with many variables. The complexity of classroom interactions makes it challenging to replicate the exact conditions of academic studies, limiting the applicability of some findings. What works in one college may not work in another due to student demographics, teaching styles, and resource availability, while research on younger children (or on young adults like university students) may not always be applicable to 16-18s.

Academic Incentives: Academics are often incentivised to publish findings in journals rather than to ensure their practical application. Researchers seek recognition from their peers, but few invest the extensive hours needed to translate their research for practitioners. Consequently, once published, research can be overlooked or forgotten, confined to academic silos rather than actively implemented in classrooms. Educators require clear, actionable insights rather than discussions bogged down by effect sizes, significance levels and p-values. We need more teachers involved in research and more researchers returning to teach, bridging the gap between theory and practice.

Implementation Challenges

Context Matters: Every school or college operates as a unique entity, with distinct characteristics and challenges. "Off-the-shelf" research solutions may not be suitable for all contexts and often require customisation. As a former head of maths in a school, I’m reminded of when I implemented a new behaviour system for missed homework “off the shelf” from my previous school. It failed because, unlike my previous school, most students in the new school couldn't stay after hours due to bus schedules. Understanding the specific behaviours you aim to change and for whom is critical.

Evaluation Should Not Mean More Data: A behaviourally informed college must be willing to test ideas and evaluate their effectiveness without generating unnecessary data. Data collection should be meaningful. At a previous school, an Ofsted inspection praised our use of retrieval quizzes but questioned whether their effectiveness had been evaluated. In response, the senior leadership team periodically required arbitrary data entry to "show progress," which was ineffective and counterproductive.

The Kitchen Sink Approach: Behavioural interventions often have small effect sizes, and their power in studies run at a national level by governments or universities comes from large sample sizes, which are hard to achieve at the institutional level. This can lead to colleges applying multiple techniques simultaneously, such as new behaviour monitoring systems, increased sanctions, parent contact protocols, and restorative conversations. This "kitchen sink" approach can muddle the effectiveness of individual interventions. Research suggests that using multiple interventions at once can dampen their effects or cause them to counteract each other, especially if teachers become overstretched.

Long Timelines of Research Projects: Educational research projects often take years, which can clash with the need for immediate, adaptable classroom solutions. For example, an EEF study on Carol Dweck's growth mindset found little additional progress, perhaps partly because by the time the research concluded, many teachers were already familiar with the concept, limiting the study's impact.

Addiction to Adding: Behavioural science emphasises removing frictions and unnecessary barriers as much as increasing incentives and applying new techniques. However, colleges often respond to problems by adding more interventions and processes to teachers' and middle leaders' workloads, which is typically ineffective. There's an addiction to adding rather than subtracting, which needs to be addressed for more efficient problem-solving.

Recommendations

1. Use Behavioural Science Insights

• Make it Easy: Implement defaults for desired behaviours, such as which question the student should be working on. Start independent study sessions in class to improve completion rates. Simplify schedules and use a single virtual platform such as Google Classroom where possible.
• Make it Attractive: Highlight important information in communications, reduce wordiness, and provide high-value tasks. Recognise high performance publicly. Students must know what the behaviour is you want them to exhibit. Make it clear and attractive to do so.
• Make it Social: Create networks for students to observe and support each other,  optimise classroom seating to encourage collaboration, demonstrate when the majority of students are complying with norms. Show them a social model of what good work looks like.
• Make it Timely: Set up reminder systems and establish clear routines for students. 

2. Context is Key: Be wary of applying “off the shelf” ideas. Disseminate findings but with the critical lens of your own situation. Ask: will this work in our context? Then provide clear guidance on adapting these insights to different classroom environments across the institution.

3. Subtraction First: As Dylan Wiliam and John Hattie advocate, de-implementation and removing unnecessary frictions or "sludge" from systems should be a primary strategy for behaviourally informed organisations. Begin by identifying and eliminating barriers rather than adding new interventions.

4. Adapt and Test: Adopt a critical and evaluative approach. Incorporate student feedback alongside quantitative analysis. Avoid burdening teachers with meaningless data collection. Instead, select a technique, test its effectiveness, assess the results, and then refine your approach. Focus on precise, targeted interventions rather than a broad "kitchen sink" strategy.

George Kinkead is a PhD student at King's College London studying the application of behavioural science to improve outcomes for GCSE mathematics resit students​. He is a former maths teacher and deputy head of sixth form.

References

Foliano, F., Rolfe, H., Buzzeo, J., Runge, J. and Wilkinson, D. (2019). Changing Mindsets: Effectiveness trial. [online] Available at: https://d2tic4wvo1iusb.cloudfront.net/production/documents/projects/Changing_Mindsets.pdf?v=1718102318 [Accessed 11 Jun. 2024].

Gascoine, L., Tracey, L., Fairhurst, C. et al. (2022). ReflectED Evaluation Report. London: Education Endowment Foundation.

Hamilton, A., Hattie, J., & Wiliam, D. (2023). Making Room for Impact. Corwin Press.

Mažar, N. and Soman, D. (2022). Behavioral science in the wild. Toronto ; Buffalo ; London: University Of Toronto Press.

Service, O., Hallsworth, M., Halpern, D., Algate, F., Gallagher, R., Nguyen, S., Ruda, S., Sanders, M., Pelenur, M., Gyani, A., Harper, H. and Kirkman, E. (2014). EAST Four simple ways to apply behavioural insights. [online] Available at: https://www.bi.team/wp-content/uploads/2015/07/BIT-Publication-EAST_FA_WEB.pdf

Soman, D. and Yeung, C. (2020). The Behaviourally Informed Organization. University of Toronto Press.