About the project
What is the aim of the project?
White British boys from low-income backgrounds underachieve throughout primary and secondary school in England. As a result, at GCSE, these pupils are outperformed by girls and by male pupils from other ethnic groups. The SOMEONE LIKE ME project aims to test whether role models can help tackle this issue.
What are the research questions?
Role models are frequently sent into schools or featured in education resources to ‘raise aspirations’. This project aims to provide evidence on this approach by asking:
Can role models encourage pupils to do better at school?
Are male role models more effective for boys than girls?
Are university students or apprentices more inspirational role models for boys?
How will you answer the research questions?
We are developing video case studies of university students and apprentices talking about their experience at school and current studies - you can watch an example below.
The video case studies will be tested via a randomised controlled trial (RCT). In participating schools, all tutor groups from Year 8 upwards will take part but some classes will watch the videos and some won't. Then all pupils (whether they watched the videos or not) will complete some surveys and tests. By comparing pupil's answers, we can work out if the videos have any effect on how they feel about education.
Why do you use maths tests? Why not look at GCSE/KS5 qualifications?
As well as surveys, we will measure pupil performance on low-stakes maths tests which they will take as part of the project. We use these tests because:
If we focused on GCSEs/KS5 qualifications we could only involve certain year groups in the project.
It is much harder to shift performance in high-stakes exams and it is unlikely that our light-touch approach will have a dramatic effect on pupil results at this level. However, if we see improved performance in low-stakes tests, it will suggest that the use of role models could be scaled-up to help build student motivation and performance throughout school.
We use maths tests because the scores have been shown to be more dependent on effort than other subjects; therefore, improving pupils' attitudes to school should translate into an improvement in test performance. All the tests contain 10 multiple-choice questions and should not be arduous for pupils to complete - they will be 5 minutes long and not need calculators.
Why do you randomise?
A randomised controlled trial (RCT), where only some of the classes are randomly selected to watch the videos, is the 'gold standard' of evaluation and will help provide the most robust evidence on whether the videos actually have any effect on pupils.
The whole project has been given ethical approval by the University College London Research Ethics Committee (application number 11263/001).