Chapter 8 Variable derivation of SARMD

In this chapter, we present the variable derivation of the most recent version of the harmonized variables of each dataset in the SARMD collection. Figure 8.1 shows the Stata code used to create each harmonized variable per country. As some countries have more than two surveys over time, it is necessary to scroll left and right in order to compare the variable derivation across years. Alternatively, you may select a different set of years in the filter at the bottom of the figure.

This chapter has been written with two purposes in mind. First, we ease the burden of the SARMD user by letting them know specifically how each variable in SARMD was derived. In this way, the user will not only understand the logic behind the harmonization but is also able to modify the source code as they please. Second, and more importantly for the SARTSD, we want the SARMD users to be active users in the improvement of the harmonization. That is, we would like you to contribute to the harmonization if you find out that something has not been done correctly. This is why we have also made avaialable all the harmonization do-files in the worldbank/SARMD Github repository. As long as you have a Github account (ideally with your @worldbank.org email account), you can contribute in two ways:

  1. Once you are in the repository’s main page, you can “create and issue” by following these instructions. Basically, you only need to click on the “Issues” tab, make that your issue has not been mentioned before, and, if not, create a new issue. From there, Github will notify you via email of any change made to the repository to address your comments. Then, you can reply to the message directly from your email account.

  2. If you know how to solve the problem that you found, you could modify the source code by forking the repository, make the modification, and create a pull request. Though it may seem a little tedious for many, learning Git is not as difficult as many think and way more usuful than people actually know.

SARMD variable derivation

Figure 8.1: SARMD variable derivation