Problem
The Determinants of Democracy
Scholars of political development have long been interested in studying the determinants of democratic regimes. What causes one society to adopt democratic institutions and others to adopt alternative forms of government? Some argue that ethnic heterogeneity is a key predictor. The argument is that when citizens possess different languages, religious values and cultural beliefs, they are less likely to adopt democratic institutions. This is because democracy requires citizens to work with one another to govern effectively. In short, ethnic fractionalization is expected to decrease the probability of instituting democratic institutions because it heightens the costs associated with deliberation and collective action.
In this problem set, you will examine the relationship between ethnic fractionalization and the likelihood that a country is a democracy.
To answer the following questions, use the world.csv dataset, which is available on the course website. The dataset contains many more variables than you will need. The variables that you will be using are listed in Table 1. A full codebook for the dataset can be found on the course website.
As you work through the questions, pay attention to the direction of the relationship (e.g. an increase or decrease). Include the direction of the relationship in your answers (when appropriate). In addition, please copy and paste your R code at the end of your problem set.
Table 1: Variable Definitions
|
Variable
|
Definition
|
|
democ_regime
|
= 1 if the country is a democracy
= 0 if the country is not a democracy
|
|
frac_eth3
|
= 1 if the country has a low level of ethnic fractionalization
= 2 if the country has a medium level of ethnic fractionalization
= 3 if the country has a high level of ethnic fractionalization
Ethnic fractionalization is a measure of ethnic heterogeneity. The ethnic fractionalization scale is calculated using the amount of linguistic and racial heterogeneity in country. This variable collapses the scale into three categories (low, medium and high). A high level of fractionalization means the country has a high level of linguistic and racial heterogeneity.
|
|
gdp_cap2
|
= 1 if the country has a low per capita GDP
= 2 if the country has a high per capita GDP
This variable uses data from the United Nations Development Programme.
|
This variable uses data from the United Nations Development Programme.
1. State the testable hypothesis described in the first paragraph of the problems set. (1 point)
2. Use R to create a cross-tabulation table that allows you to test the above hypothesis. Be sure the table follows the conventional rules described in Pollock. Include both frequencies and column percentages. Include your cross-tabulation table in your write-up.
3. Based on the cross-tabulation, table what is the result of the hypothesis test. In your answer, cite information from the table.
4. Think about the role that a country's per capita GDP might play in the relationship between ethnic fractionalization and democracy. Do you think per capita GDP might influence this relationship? If so, explain the way in which it might influence the relationship. If not, explain why not.
5. Create a controlled comparison table that allows you to determine the effect of ethnic fractionalization on a country's type of government, controlling for per capita GDP.
6. According to Table 3, what is the controlled effect of going from a country that has a low level of ethnic fractionalization to a country that has a high level of ethnic fractionalization on government type? Show your work and then state your final answer in sentences (Hint: the answer has two parts.)
7. What is the partial effect of ethnic fractionalization on government type, controlling for per capita GDP? Show your work. (Hint: this is a weighted average.)
8. Take a look at the controlled comparison table and your answer to Question 6. Is there evidence of an additive, interactive or spurious relationship? Explain.
9. Let's flip the question around. What is the partial effect of per capita GDP on government type, controlling for low and high levels of ethnic fractionalization? Use the controlled comparison table to calculate this effect. Show your work, and include a sentence to summarize the result. (Hint: first calculate the two controlled effects and then calculate the weighted average of the controlled effects.)
10. In R, create a bar plot to display the information in the controlled comparison table graphically. Be sure your plot is clear and informative (e.g. proper labels on the axes, helpful legend, etc.)
Attachment:- codebook-for-world-data.rar