About once a month I spend the entire day in a windowless room at the Early Intervention Foundation's HQ discussing the finer details of programme evaluations with my evidence team colleagues and external experts. This is the penultimate stage in our programme assessment process, which leads to interventions receiving an evidence rating and appearing on our online guidebook.

It is not unusual for us to spend time debating the finer points of trial design, differential attrition, structured equation modelling and propensity score matching. We know this type of discussion isn't for everyone and we don't do it to satisfy our inner geek. We do it because we believe that providing high-quality independent assessment of the quality of evidence for early intervention programmes can help to improve outcomes for children far beyond our little windowless room.