SILC Innovations Research Brief 3: Group Performance in Fee-for-Service Savings Groups
To assess the model and inform future SILC rollouts on this fee-for-service, savings- group delivery channel, CRS carried out a broad research study using a Randomized Control Trial (RCT) design. The research was set up to make a fundamental comparison between two delivery channels: the fee-for-service PSP model and the more conventional project-paid FA model. To rigorously compare the two, an experimental design established statistically comparable cohorts of agents serving members in comparable environments over approximately a one-year interval (see the additional research background section on page 8). In total, the study tracked 333 randomized agents across two cohorts (separated by about one year). The agents were assigned either fee-for-service PSP status or stipend- paid FA status for the research interval, which followed a 12-month training phase in which all agents were paid a stipend. Management Information System data was collected from all agents on a quarterly basis and included a multitude of data points. This brief draws on the data specifi cally pertaining to group performance, focusing on the comparison between groups served by randomized PSPs and groups served by randomized FAs. A central question was at the core: can we detect systematic diff erences in group performance between PSP-supported and FA-supported groups?
To make these comparisons, we employed the data set for groups created and served in the RCT period by the randomized agents (n = 1,996 groups). The data that went into the set was drawn from the quarterly observation following the end of the one-year randomization period in each region of the study.2 The metrics tracked and reported here can be divided into two group-performance categories: membership and fi nances. On membership, we see that PSPs were supporting signifi cantly3 larger groups on average (consistent with fi ndings elsewhere in this RCT). PSPs also led on membership growth within cycle, with the country breakdown indicating that the results are being driven clearly by the Kenyan subpopulation. On dropout rates and percentages of female members, we see mostly parity across the two delivery channels, though in certain subpopulations the FA-supported groups hold a moderately signifi cant edge.