Call Us: US - +1 845 478 5244 | UK - +44 20 7193 7850 | AUS - +61 2 8005 4826

Income Mobility

Stochastic dominance results Our stochastic dominance analysis proceeds in three steps.5 The first two steps represent standard single-difference stochastic dominance tests. First, we test for SD within control and treatment over time. If outcomes in the intervention sites improved while those in the control sites worsened this would present evidence for a positive ALRMP II program effect. However, we find no difference between trends in intervention and control sublocations. Both have improved slightly, so on the basis of 5 Full results for the first two steps testing for stochastic dominance within treatment and control groups across time and across treatment and control sublocations at each point in time are available on request. 11 these tests we cannot conclude that ALRMP II has had no impact. Second, we compare intervention and control sublocations before and after ALRMP II. Control sublocations dominate in most cases and interventions sites never dominate. Again, this does not indicate any program effect. Third, we apply the difference-in-difference method outlined above to test for stochastic dominance between changes in intervention and changes in control sublocations