Impact Measurement Based On Repeated Randomized Control Trials: The Case Of A Training Program To Encourage Social Entrepreneurship

Research Summary Designing effective entrepreneurship training programs is still a challenge despite the investments in training made by governments and private institutions, and its importance for economic growth. We report a case of impact measurement of a social entrepreneurship program based on repeated randomized controlled trials (RCTs), discuss challenges of conducting repeated RCTs, and implications for policy evaluation. Impact measures from the first edition of the program showed no detectable treatment effects. The second edition was adjusted by reducing leadership training and increasing traditional entrepreneurial skills training, and had strong treatment effects on entrepreneurial activities, the creation of a new venture during the program, and subsequent start-up activity. Employing sequential field experiments can improve entrepreneurship training programs despite the challenges of executing RCTs in the field.Managerial Summary Governments, institutions, and businesses increasingly invest in innovative entrepreneurship training programs that tackle societal problems. However, we know very little about the effectiveness of such trainings. To measure the causal impact of a social entrepreneurship training program we use an experiment where we assign one group of applicants randomly to training. A comparable group is selected randomly to not obtain the training. There were no detectable differences between the two groups the first time the program was run. After substantial changes were made, where analytical skills training was boosted at the expense of a reduction in social leadership skills training and social entrepreneurial identity development, and individualized coaching intensified, we find a large impact on entrepreneurial activities, both during the program and 3 years later.