The Work Programme is delivered by 18 private, public and voluntary sector organisations, working under contract to DWP. These organisations are known as prime providers, or "primes", and operate within a geographical Contract Package Area (CPA). Each CPA has either two or three primes and individuals entering the Work Programme are randomly assigned to one of these. Comparing the outcomes of individuals assigned to each prime within a CPA provides robust estimates of relative effectiveness. Overall, there are 40 contracts across Britain, allowing 26 within-CPA comparisons. An interim report confirmed both that randomisation had worked well and that there was statistically significant variation in outcomes between primes. This report extends the interim analysis to relate the variation in outcomes to differences in the services offered by primes. By doing so, it aims to understand what makes some primes more effective than others. The results provide evidence of certain aspects of service being more associated with better outcomes than others and that much of this is concentrated among particular subgroups. Several relationships have been revealed which could provide some clues as to what is driving the variation in provider performance. However, the analytical approach provides evidence of an association only and does not enable us to establish whether these relationships are causal in nature. In addition, much of the variation in performance is not associated in a statistically significant way with the provider characteristics that we have been able to measure. This may suggest that the key characteristics have not been taken into account. |