Adult numeracy randomised controlled trials: Embedding maths in health and social care

The Embedded Maths intervention was developed as part of the Department for Education (DfE)’s Adult Numeracy Trials, aimed at building the evidence base for what works to improve numeracy skills in adults. It supported maths learning by expanding on the Level 1 maths concepts in the curriculum of Health and Social Care (H&SC) Level 2 vocational qualifications, aligned to tasks undertaken in H&SC jobs. The intervention re-worked the Level 2 H&SC curriculum[1] to incorporate 18 guided learning hours (GLH) of H&SC-relevant Level 1 maths content, covering approximately 60% of the Functional Skills Qualification (FSQ) Level 1 maths curriculum while also delivering in full to the H&SC Level 2 curriculum. This meant that learners on the H&SC courses had the opportunity to enhance and reinforce their maths skills in naturally occurring, relevant work-related contexts.

The original intention was to evaluate this intervention though a randomised controlled trial (RCT) of approximately 60 providers. However, various challenges in the recruitment and retention of adult education providers meant neither this, or a smaller, pilot trial could be delivered. Of 27 providers that originally signed up, 12 registered for training, 6 partially or fully trained tutors and 2 providers proceeded to deliver the intervention. Rather than stop the trial at this stage, the Department agreed to take the evaluation forward as a small-scale, non-randomised descriptive implementation and process evaluation (IPE). The rationale was twofold. First, it meant information could be captured about the intervention and tutors’ and learners’ response to it. Second, it would provide understanding of the practicalities and challenges of running RCTs in the adult education sector, which was one of the Adult Numeracy RCT programme’s aims. The evaluation was therefore re-designed with 2 objectives: a) to maximise learning about attitudes towards the intervention and the RCT itself from the 2 participating providers; and b) to understand why other providers dropped out between sign-up and delivery, and gauge whether the problem was primarily the intervention or other factors. The aim was to inform future attempts to run trials using embedded maths in vocational courses.

The key findings from the research are:

• The Embedded Maths intervention has potential to improve non-maths tutors’ confidence in teaching maths. In particular:
  • The 2 H&SC tutors who successfully delivered the intervention reported improved confidence in covering the additional maths content due to its real-life application to H&SC roles. They were also positive about the effect of the Embedded Maths intervention on their own practice.
  • These 2 tutors attended all pre-programme training and both were positive about the content, saying they found it flexible and helpful.
• These 2 tutors made only minor adaptions to the intervention when delivering it in the classroom, focussing on improving accessibility. In future, these adaptions can be added to the intervention design.
• The intervention has the potential to improve learner capability and confidence in using maths in the H&SC workplace. Learner attendance at the Embedded Maths sessions was high and learners reported improvements in confidence with using maths at work.
• Some H&SC learners were already confident in maths and/or had a Level 2 maths qualification, which suggested the value of the vocationally-relevant approach for a range of H&SC adult learners.
• It was difficult to establish the specific contribution of the Embedded Maths intervention to H&SC learners’ perceived outcomes as many learners were also engaged in other maths courses with their provider. However, these courses did not focus on maths skills in H&SC jobs. This means the increase in learners’ confidence using maths in the workplace is promising in terms of the effect of Embedded Maths. A future 3-arm trial could explore the extent to which completing a FSQ Level 1 maths qualification alongside the Embedded Maths course affects learner outcomes, compared to completing Embedded Maths on its own and to business as usual (BAU) H&SC courses.
• While learners taking other maths courses alongside the Embedded Maths H&SC course could not differentiate particularly between the maths content in them both, they indicated that overall the courses worked well together.
• Some H&SC learners’ maths skills were higher than expected of the Adult Numeracy Trials’ target population; in one participating provider, learners were more highly qualified in maths due to qualifications gained in home countries. Because the intervention could not apply maths-specific eligibility criteria for practical and ethical issues[1], some learners may already have had the maths skills being taught. This could be minimised by adding maths-ability eligibility criteria for the H&SC course or targeting specific classes of vocational learners. However, the feasibility of recruiting sufficient learners with these extra criteria would need to be tested.
• Provider drop out was driven by a combination of: low H&SC tutor willingness or confidence in their ability to engage with and teach the H&SC-related maths content; the volume and structure of the tutor training which some found off-putting; and organisational confusion around the demands of taking part in a RCT on adult numeracy. The necessary delivery timescales also meant starting the first phase of recruitment in parallel with the finalisation of training and curriculum design. As a result, questions about what allocation to the treatment group would entail could not be answered in detail early in the recruitment period. Additionally, and outside of issues related to the trial and intervention design, some providers were simply unable to recruit enough learners to the H&SC course to deliver it.

While some of the reasons for provider drop out could be addressed in future research, the issue of low confidence (sometimes presenting as maths anxiety) in H&SC tutors when asked to teach H&SC-related maths is a more difficult challenge. It is a noteworthy point for future research into embedded approaches to consider. Earlier engagement with H&SC tutors ahead of delivering the Embedded Maths content might help to address tutor withdrawal in future research. Offering voluntary enrolment of tutors and a more flexible training schedule may also help. With some adjustment, this intervention would appear to have value and be suitable for further exploration.

[1]H&SC embedded maths content ran through the course, so could not be restricted to only those leaners with weak numeracy skills. Leaners signed up to the H&SC course based on criteria that did not include maths qualifications; adding these in post-hoc would have excluded some learners from provision that they had already been accepted onto.

[1] Although several awarding bodies offer this qualification, the intervention was applicable to all of them.