Understanding ‘what works’ to change attitudes and behaviours

by Ben Wilson

Published: 09 Nov 2017

"Advertising is based on one thing: happiness. And do you know what happiness is? Happiness is the smell of a new car. It's freedom from fear. It's a billboard on the side of a road that screams with reassurance that whatever you're doing is OK. You are OK."

This quote from Mad Men’s Don Draper reminds us that the notion that human decisions are often based on emotion or biases rather than reason is certainly not a new one. Marketers in public bodies have used this insight to drive many of the most influential and important campaigns to change and improve society.

From Lord Kitchener’s ‘Your Country Needs You’ call for recruits in the First World War to campaigns to discourage smoking, tackle drink-driving and promote road safety, Britain is the modern country it is in part because of the influence of behavioural marketing campaigns.

Now, behavioural science including ‘nudge’ techniques are part of the mainstream of economics, organisational strategy and policy-making, as is a ‘what works?’ approach to evidence-led policy.

Whether it is auto-enrolment into pensions or the presumption of consent on organ donation, whole organisational strategies and models are being changed as more is understood about the ‘limitations of reason’. 'Nudge’ pioneer Professor Richard Thaler was awarded this year’s Nobel Prize for Economics for his work on behavioural economics.

Our #PowertotheBump and Working Forward campaigns that have used tested techniques to tackle pregnancy and maternity discrimination and promoting flexible working are nationally recognised. More than 150 organisations have signed up to Working Forward covering 1.3 million employees. We’re already seeing an increase in reporting and changes of workplace practices within organisations as a result.

Like many public bodies or regulators, the Equality and Human Rights Commission relies on a strong evidence base in order to know where and how to act. This week, we are publishing a programme of research as well as our first randomised controlled trials, looking at what influences discriminatory attitudes and behaviours and what we know about ‘what works’ to tackle them.

Randomised control trials

We teamed up with the Behavioural Insights Team to conduct two trials into pregnancy discrimination and flexible working, led by the brilliant Dr Tiina Likki.

The first trial found that ‘social norming’ – businesses wanting to keep up with trends and follow other businesses – was the most effective way of prompting similar behaviour when it comes to promoting flexible working.

The second raised questions about the effectiveness of one-off interventions designed to change behaviour. Our randomised controlled trial found that managers who completed an online exercise to encourage perspective taking and increase empathy towards women and new mothers, actually made them less empathetic than those line managers who did not complete the exercise. This is surprising given that this type of diversity training is frequently used in organisations across the world.

More research may be needed here as this could raise important questions about whether this sort of one-off, quick-fix intervention actually works – and whether interventions in the workplace are being robustly evaluated. We need to look at whether designing bias out of systems, for example through name-blind recruitment, could be more effective than interventions attempting to get individual people to change their behaviour. Perhaps an approach which looks at changing both structures and attitudes is what will ultimately result in real progress in the workplace.

What we know about 'what works'

We know from previous research that the evidence on ‘what works’ to tackle prejudice and discrimination in Britain is not strong enough. To address this we have widened the evidence available to policy-makers by evaluating four interventions:

  • One Globe Kids mobile resource (app and website) from Globe Smart Kids, evaluated in partnership with the University of Kent
  • Show Racism the Red Card educational work in secondary schools, evaluated in partnership with the University of Kent
  • Kumon Y’all community initiative, evaluated in partnership with Cambridge Policy Consultants
  • Race on the Agenda training sessions, evaluated in partnership the University of Greenwich and the Runnymede Trust

We have also developed a guide to evaluating anti-prejudice interventions in a manageable and meaningful way, so that all organisations can reveal the impact of their own projects.

Each intervention we tested was proven to be effective to some degree in tackling prejudice and unlawful behaviour. The work, however, did underline the challenge of establishing ‘what works’ across a varied and complex set of issues. We will be sharing our findings along with our new guide to evaluation. As a result of this research, we have four main recommendations:

  • legislation, policies, and other interventions that aim to reduce humiliation, harassment, violence or abuse based on who people are need to be robustly evaluated
  • given there is no one-size-fits-all solution, policymakers need to take a nuanced and targeted approach to tackle prejudice and discrimination in different contexts and for different groups, while identifying where there are opportunities to make use of best practice in other settings
  • policymakers need to be mindful that interventions can have unexpected outcomes and unintended consequences and therefore they need to be evaluated and adapted on an ongoing basis
  • change is likely to come at a slow pace, so policymakers should encourage longer-term investment and planning to establish the impact of projects, including evaluating activities after the intervention itself has finished.

In the context of tight public finances, it is important that we have good evidence about the problems we want to solve and understand how to drive change to create a fairer Britain. Knowing ‘what works’ for who, and why, matters.

We must recognise, however, that this is only one of the levers we must use – and that’s why we’re also putting greater emphasis on the robust use of our enforcement powers. As Andrew Rawnsley put it recently:

"Some problems are just too big to be fixed by adjusting the ‘choice architecture’. Not a gentle hand on the elbow, but a muscular kick up the arse."

Find out more about the Commission's work on prejudice and unlawful behaviour.