Although it is important to establish causal relationships in social policy evaluation, the effects are difficult to observe due to sample selection. To evaluate the performance of estimators designed to handle sample selection bias, we analyze data from a Norwegian rehabilitation project with a...

Explanations of technological diffusion focus on information being spread by epidemic processes, on profit heterogeneity, or oligopolistic competition. The model in this paper integrates the epidemic and the heterogeneity-based approaches. Predictions are tested on Norwegian data concerning the...