After-school clubs have become an integral part of learning in the past couple of decades. Successive governments have had an expectation that schools will provide extended provision, and part of their role is to enhance learning in the classroom and to build and develop other personal and study skills.
Code Club UK supports a nationwide network of volunteer and teacher-led after-school programming clubs. It was founded in 2012 and, in 2015, joined forces with the Raspberry Pi Foundation.
Code Clubs are run in schools and libraries, for children aged nine to 13 years. They run for around an hour a week during term-time and have about 15 participants who learn to programme by making games, animations, applications and websites. Code Club UK’s projects and materials support the teaching of Scratch, HTML/CSS and Python. The aim of Code Club is to develop children’s programming skills and to inspire them to consider programming and other digital careers in the future.
Five years ago there were only a handful of Code Clubs in the UK. Now, there are nearly 6,000, attended by more than 82,000 children. There are also more than 10,000 Code Clubs running around the world.
Between June 2015 and March 2017, Code Club UK worked with the National Foundation for Educational Research (NFER) to undertake an evaluation of Code Clubs. The evaluation took the form of a randomised control trial (RCT), with an associated process evaluation. It explored the impact that Code Clubs make on children’s computational thinking and programming skills, as well as their attitudes towards computers and coding. In some cases, Code Clubs were set up purely to participate in the trial. NFER’s report, Randomised Controlled Trial and Process Evaluation of Code Clubs, was published recently.
The trial involved 21 schools. Pupils who expressed an interest in taking part in the clubs completed a series of baseline and end-point assessments. The Bebras Computational Thinking Assessment was chosen as the primary outcome measure, with secondary measures including a coding quiz and pupil attitude survey.
Following the baseline assessments, pupils were randomly assigned to intervention and control groups. Pupils in the intervention group attended Code Club for a year and schools were asked to cover Scratch, HTML/CSS and Python. Control group children were assured of a place in Code Club the following year.
The findings from the analysis of pupils’ baseline and end-point scores on the Bebras measure showed that attending Club Code for a year did not have an impact on children’s computational thinking over and above positive changes that occurred anyway. Researchers concluded that this was probably because computational thinking was being developed as part of the normal computing curriculum and that Code Clubs were consolidating learning and skills rather than further developing them. Researchers also suggested that a year of participation might not be a long enough period to see significant changes in computational thinking.
However, the analysis showed that attending Code Club did significantly improve pupils’ coding skills in Scratch, HTML/CSS and Python, and this happened even when a control group of pupils were learning Scratch as part of the computing curriculum in school.
Intervention pupils were using all of the programming languages more. This was particularly the case with Scratch, for which the proportion of intervention children who reported using Scratch every week rose by 27 percentage points between the baseline and end-point.
The study also found that pupils attending Club Code were using computers more and, in addition, a higher proportion at end-point reported that they were good or very good at making things with code compared to control group pupils and the baseline.
Pupils’ interest in learning about coding, and learning about coding languages in the future, was high for both groups at both baseline and end-point, with two-thirds or more of pupils in both groups reporting that they were interested or very interested in these areas at end-point.
In addition, just under half of pupils in both groups reported that they were interested in a job that involved coding at end-point. However, Code Club was not shown to have an impact on pupils’ already high levels of interest in these three areas.
Teachers reported that early impact on pupils’ understanding of concepts and development of confidence in coding could be seen at the end of the first term of Code Club delivery. However, two or three terms were needed to see progress in terms of pupils working independently and displaying the resilience to work out problems for themselves.
Echoing pupils’ own reports, teachers reported a range of positive outcomes for pupils, including the development of confidence and skills in coding, IT and using computers. In some cases, teachers were drawing on the skills of Code Club pupils to support other pupils in the classroom who were struggling with coding.
Teachers reported that pupils were developing important skills for the future: “It’s a great thing to be exposing children to and in the future they will be glad that they did it. All the world is based on IT now and they need to do it, and it’s very positive for their future.”
In addition, teachers emphasised that Code Club gives some pupils the opportunity to succeed in coding when they are not doing so well in other school subjects: “It’s a fantastic resource and it enables pupils to succeed when they haven’t elsewhere. It’s a very good skill for pupils to develop.”
Most club leaders reported that there were not any differences in engagement and impact for boys and girls. Others reported greater impact for boys, or girls. Some club leaders reported that boys were more enthused, committed and resilient and achieved more than girls. Other teachers reported that girls might lack confidence initially but that they performed better than their male peers over time.
One respondent said: “A couple of the girls lacked confidence at the beginning of the sessions but their confidence and self-belief increased greatly as the weeks went by – they ‘dug deep’ and solved many of their own errors. In contrast, a couple of the boys were over-confident but when the going got tough with the HTML projects they gave up easily and constantly asked for help, wanting their issues solved straight away rather than just be pointed in the right direction.”
Club leaders also pointed out that they had benefited themselves from running the clubs. Benefits included increased confidence in coding and greater familiarity, expertise and skills in all of the programming languages, which they could disseminate to other staff.
Most clubs ran smoothly during the academic year and club leaders reported that access to Code Club UK’s projects and teacher notes had been key to the successful running of their club. As this club leader commented: “Great resources: pupils really engaged with the material. They loved adapting and creating their own games.”
Where difficulties were experienced, they related to insufficient time, technical problems, a lack of knowledge of the languages and pupil drop-out. Half of the Code Clubs did not cover all of the programming languages during the year, with a number focusing on Scratch which pupils found more enjoying and accessible.
Philip Colligan, chief executive of the Raspberry Pi Foundation, said: “The trial has given us lots to work on. What do we mean by computational thinking and how do we best support volunteers and teachers to teach those concepts in an engaging way? How do we get better at managing the transition from visual to text-based programming languages? How can we most effectively identify and spread practice between Code Clubs? These are questions that are important not just for Code Club but for the whole field of computer science education.”
He added: “With the evidence of positive impact and the insights generated by this research, we are in a much stronger position to build the skills and confidence that young people need to thrive in an increasingly digital world.”
- Suzanne Straw is deputy head of Centre for Evaluation and Consultancy at the National Foundation for Educational Research.
Further information
- Randomised Controlled Trial and Process Evaluation of Code Clubs, NFER, May 2017: www.nfer.ac.uk/publications/CODE01
- Code Club UK: www.codeclub.org.uk
- Raspberry Pi Foundation: www.raspberrypi.org
NFER Research Insights
This article was published as part of Headteacher Update’s NFER Research Insights series. A free pdf of the latest Research Insights best practice and advisory articles can be downloaded from the supplements page of this website: www.headteacher-update.com/supplements/