3. How to randomize I? (by Dean Karlan)


First, a practical introduction on the randomization methods is given: (1)basic lottery; (2)phase-in; (3)rotation; (4)encouragement. With respect to different contexts, necessities and questions the evaluators want to answer to, the researchers can choose to use one or combine different methodologies. Moreover, a correct randomization will help to measure the natural spill-over effects and to limit the research/unnatural ones.

Second, stress is given to the importance of working with collaborative actors, “who want to know the answers”. That is why the organization(s) implementing the program has to be actively involved in the research process by the evaluators.

Third, a part of the lecture is dedicated to ethic issues, too. It is often argued that it’s wrong to use people as guinea pigs and if it works, then it’s wrong not to treat everyone. Dean’s answers are very brief and, I believe, convincing questions themselves: why are prescription drugs different?; is it useful to know which is the best solutions among several to address a particular problem?

Finally a PLAN draft divided in 5 phases is provided for evaluators: (I)identify the problem to be addressed (market failure) and the proposed solution (its logic); (II)identify the key players; (III)identify the key operational questions to include in the study in order to build a win-win opportunity for the commissioning institution and the evaluators; (IV)design the randomization strategy; (V)define a data collection plan.

Advertisements

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