For my project, one dataset I would use is the Polity database. Its citation is as follows:
Gurr, Ted and Monty Marshall. Polity Data Series, Political Instability Taskforce. Updated June 2014.
The Polity dataset is a widely used source of data on the level of democracy in governments around the world. To produce a government’s final score (from 10 for a full democracy to -10 for an autocracy), the dataset uses six “component variables, which are as follows:
- Regulation of Chief Executive Recruitment: the level of explicit institutionalization in the selection of the head of government. More institutionalized is more democratic.
- Competitiveness of Executive Recruitment : the level mobility of figures up the executive hierarchy. Hand-picked selections are the least democratic and elections are the most
- Openness of Executive Recruitment: what part of the population is able to actually attain a position through this process. More opportunity is more democratic.
- Executive Constraints: how much institutional control there is over the executive. More constraint is more democratic.
- Regulation of Participation: the level of institutionalization of the political participation (e.g. voting) process. Clearer, more institutionalized participation is more democratic.
- The Competitiveness of Participation: how much opposition can realistically be pursued in the political process. More room for opposition is more democratic.
The dataset covers all states with populations above 500,000 in every year since 1800. All together, there are 17,061 cases in the set.
The limitations of the dataset can be found primarily in the process of transferring the data into a coherent understanding of democracy. Although the variables used are extensive in nature, they could probably never be extensive enough to capture the entire picture of a state’s level of democracy. And the collection of data also poses a challenge. Many of the variables may depend on records kept of elections, which may be unreliable or nonexistent in many countries.
Bobby — the Polity dataset is clearly relevant to your research, and you do a good job in discussing the coverage of the data as well as the limitations here. Please note, though, that the level of measurement for the particular indicator that you would use (potentially) for your DV is not discussed here as is required. Would you take the composite score, or would you take data from one of the elements? What is the level of measurement for the indicator that you would select?