From theory…


I guess introducing my “mission” here is a good starting point. I have been assigned by Teach A Man To Fish to an evaluation role of its projects in the rural community of La Bastilla, in Nicaragua. The project ‘Financially Self-Sufficient School for Rural Entrepreneurs’ and “Small Business Programs” are respectively carried out for high school students and primary school pupils’ parents. In few words, the former is a technical high school curriculum aimed to make the educational institution financially self-sufficient in 5 years. The idea is to help the agricultural school to commercialise their products in order to reach the economic sustainability in 2014. The latter, is a voluntary program of parents involved in an annual micro credit scheme that destines 10% of the profits to the primary school. In the next months I will collect the data and carry out my impact evaluation!Will I? My idea is to mix qualitative methodology (focus group) and quantitative techniques (Difference in Difference and PSM) methodology in the technical high school and to take a census of the families in the primary school, using quasi-experimental methods to match treatment and control. Next step, collecting the data. Now, I need your ‘yes you can’…

Advertisements

3 responses to “From theory…

  1. yes you can, Antonio!
    watch out the spillover effects!

  2. Yes you can ! btw what is PSM?

    • PSM stands for Propensity Score Matching, a quasi-experimental approach used to measure treatment effect. That is to say, in order to estimate the impact of an intervention, we need to compare a treatment – which receives the intervention – with a control – which doesn’t. Ideally, treatment and control group should have the same features before the intervention, i.e. the effect of the intervention is the estimated difference in the outcome of interest. However, unless there was not a randomization preceding the assignment of the treatment, some issues of not random selection in the treatment can bias the estimate if we don’t take into account them. Thus, PSM allows to match a posteriori (after the intervention is assigned) the treatment and the control group, through observable characteristics of the individuals. An interesting and intuitive reading on this topic is ‘Assessing the Economic Impact of HIV/AIDS on Nigerian Households: A Propensity Score Matching Approach’ (Canning et al. 2006). Here, the compelling issue is that for estimating the impact of HIV/AIDS (the treatment), it is not enough comparing an affected household with the average unaffected Nigerian household. Otherwise, it is necessary to select a control that has similar (not statistically different) probability of being affected by HIV/AIDS. Only then the impact effect can be assumed being unbiased. For example, it is arguable that the probability for a household with average age 60 to be affected by a sex disease is different from the probability for a household with average age 25. The PSM matches an affected household aged 25 with an unaffected one aged 25, too. Age is only one among plenty of observable characteristics we can use to match treatment and control.

      I hope you found it useful, looking forward to reading a new post on your blog…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s