POIR 611

Professor Graham

Spring 2014

Homework 1: Data Management (1)

For this homework, you must post a working .do file to the course dropbox folder. Write the answers to questions 1-3 as comments in the your .do file immediately following the analysis that they require.

Save the raw data you download from the World Bank in the “611 Homework” dropbox folder, with your initials as part of the file name. Without this data, I will not be able to run the .do file, and you will not get credit for submitting a working .do file.

We have not really covered the theoretical material you need to understand to do Part 2, but I think most of you have enough of a social science background to give me reasonably intelligent answers. This is something I want you to muddle through and talk to each other about. Particularly questions 1 and 2 and focused on the relationship between concept and measure – how well do different measures map to the theoretical constructs with which they are associated? We will talk about your answers to all this in class. For question 3, you may need to do some independent research on log scales if you don’t currently have intuition regarding how they work (which I imagine most of you don’t).

This homework is not supposed to be easy, so if you struggle, that is as it should be.

Part I:

Step 1: Download the following information from the world development indicators: GDP per capita, GDP, population, infant mortality. Download this information for all countries and all years.

Step 2: Reshape this data and append gleditsch-ward numbers. There is a template .do file for each of these steps in the course dropbox folder. You will have to combine these tasks into a single .do file. There is also an instructions sheet. Kat Vasku, one of the undergraduate students in the class, wrote the instruction sheet, so you can go to her for help as well if she has time.

I recommend beginning with the “1 Ben’s template.do” and amending that and pasting and typing in other pieces of code as you go.

Part 2:

1. How rich are countries in 2010? Give me two different answers. In one short paragraph, give the substantive meaning of the numbers you have calculated.Which of the two is more useful? (Note: There are many different ways this can be calculated – be sure you understand the implications of the decisions you make).

2. How rich were countries in 1980? In 1-2 sentences, give the substantive meaning of the number you have calculated.

3. Take a look at how population and infant mortality are distributed. If you were using either of these as an independent variable, would it make sense to log it first? Why or why not?

Part 3:

Download and merge-prep a dataset relevant to your area of research. Upload this .do file and the raw data file to the homework 1 folder.