Skip to content

Instantly share code, notes, and snippets.

@metaforte
Last active September 27, 2017 15:53
Show Gist options
  • Save metaforte/daa7567fa32092a81787b89ac9a302d1 to your computer and use it in GitHub Desktop.
Save metaforte/daa7567fa32092a81787b89ac9a302d1 to your computer and use it in GitHub Desktop.
JPAL 102X How to design randomized evaluations

The is comprised of five sections.

First, designing randomized evaluations,

  • where we tackle the question, what is evaluation?
  • What is the purpose of evaluation in general?
  • And what is the theory behind our evaluation question?

Second, selecting a sample.

  • How do we figure out who will participate in the study?
  • How do we find them?
  • And how large should our study be?

Third, measurement

  • Determining what data we will collect,and how.

Fourth, data collection and management.

  • We'll discuss here how to conduct high-quality data and the logistics of data collection.
  • Will we use paper surveys or digital?
  • How do we train our survey team?
  • How will they travel to households they intend to survey?
  • How do we manage the inflow of data?
  • And how do we ensure the security of our data?

And fifth, and finally, research integrity, transparency, and reproducibility.

  • Here we will talk about what one must do to make sure our research meets the highest level of ethical standards, how the scientific community expects research to be conducted, and practical tips for overall project management.

This course will be comprised of lecture sequences, case studies, exercises, and quizzes, and an exam at the end.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment