We finished the first week of the International Workshop on Surveys and Impact Evaluation on Friday 20th January with the presentation of four examples of impact evaluations of programs in the region.
1) Productive Sector
Rodolfo Stucchi from the IDB presented the results of an impact evaluation of the Chilean Supplier Development Program, aimed at improving and stabilizing the commercial linkages between small and medium-sized suppliers and their large firm customers.
Using data that is representative of firm population during the period 1998-2008, it was possible to evaluate the impact of the program on both suppliers and clients (large firms).
Applying propensity score matching and difference-in-differences, the findings show that both groups of firms benefited. Particularly, providers experienced an increase in sales, employment, and sustainability. Large firms registered an increase in sales and in their export probabilities.
Moreover, the impact on providers was faster than on large firms.
During the presentation, Ryan explained how we can simultaneously reach an ideal situation in both the design of the program and of the impact evaluation.
This is exactly what happened in the case of “un techo”, where it was possible to better target the program and at the same time randomize the intervention in order to measure in a more rigorous way its impacts on human development indicators.
Francisco Gallego from the Catholic University of Chile presented an experimental evaluation on the role of information on how parents choose schools for their children in Chile. The study uses two treatments: 1) Information leaflets on the performance of schools 2) Videos displaying three experiences in similar contexts.
These are the evaluation questions that have already found answers:
Can simple information about learning performance change school selection process for low income parents? And, is there any additional effect if parents are informed about returns to quality of education?
Here are the answers:
- The information leaflets have a statistically and economically significant effect on the score, price and type of chosen school.
- The probability of choosing a school increases with the score of standardized tests, while it decreases with values and discipline.
- There is no effect on already registered students.
- The use of videos does not show any additional effect.
We will be waiting for an answer to a third question: Is there an impact on the learning results of those children who changed school?
4) Early Childwood Development
David Evans from World Bank presented a study on the impact of early childhood development both on mothers and children in Brazil.
The main purpose of this evaluation was to measure the impact of attending a public nursery (0-3 years old) on the learning, emotional and physical development of the child, as well as the welfare of the family, measured through changes in income and stress levels.
Given the excess in demand, it was impossible to offer a place at the nursery for every child. Hence, together with the municipality, a lottery was designed and implemented.
The experimental evaluation, eight months after the lottery, displays the following results on family welfare (it is too soon to show any impact on the children). It was possible to observe that the families who won the lottery experienced an increase in their income and the mothers, in spite of having to wake up earlier to take their children to the nursery, are less stressed.