03 Research Design - Large-n & Small-n Flashcards
What is the research goal / research question? Evaluate the fit between the research goal of the paper (“The Missionary Roots of Liberal Democracy”) and the method(s) used.
- The article investigates the role how conversionary Protestants heavily influenced the rise and spread of stable democracy around the globe, based on history and statistics, by leading to influence the relevant factors for stable democracies such as mass education, religious liberty, mass printing, colonial reforms etc. Especially the influence of protestant missionaries has largely impacted democratisation in former colonies.
- The author states that this article challenges the narrative that it is not secularism that influenced democratisation and modernisation but the process was rather impacted by religious factors
What is the purpose of using regression analysis in this paper? For example, to generate a rich understanding of the phenomenon, test theory-based hypotheses, generate hypotheses, triangulate with other methods, etc.
- To measure the influence of CPs on democratisation and to isolate the issue from other possible explanations or causes, the author uses following approaches:
o Observation of relation of Protestantism and democracy that are
different from Europe regarding history and class structures
o Show of the salient role of CPs in mass education, printing, civil
society and so on
o Statistical demonstration of how democracies in 142 non-European
societies were influenced by Protestant missionaries
▪ By using controls for alternative explanations
▪ Instrumental variable estimation - In the analysis, the author first lays out the historical analysis followed by the statistical research to determine the validity of the former
Landman, Todd (2008): “How to compare countries?”
(!) What does the author mean with the trade-off between “the level of
abstraction and the scope of the countries”?
If a rather “open” research is carried out, it is likelier that a larger sample size is used to generate a meaningful result, leading to a higher level of abstraction. But if the research is based on a specific issue that is already limited, then the use of a smaller sample size of relevant countries is of advantage.
(2) Comparing many countries:
(2.1) What kind of methods do the scholars use when they compare many countries?
- Quantitative: “statistical method”
- Qualitative: “comparative method”
(2.2) What are the advantages and disadvantages of comparing many countries?
As more countries are compared, a higher level of abstraction is required, larger sample size, quantitative analysis. Its advantage is the ability to use statistical controls, make meaningful explanations, including the ability to identify outlier countries. Its disadvantage is mainly the limited availability of data for certain countries or time periods. Furthermore, this type of statistical comparison may rather be unfitting to analyse complex causal or historical issues.
(3) Comparing few countries:
(3.1) What do the scholars usually do when the compare few countries?
- Their research is based on a small sample size, usually from two to 20 and compare these countries to find differences and commonalities between the sample size.
- There are two most-similar system design and most-different system design and this may also depend on the result one’s looking for such as e.g. comparing Germany to either to France (similar) or to Ghana (different)
(3.2) What are the advantages and disadvantages of comparing few countries?
- Saves time, fewer resources needed, more focus on a smaller sample
- Disadvantages: smaller scope leads to a smaller generalisation
(4) Single-country studies as comparison:
(4.1) What does it mean to use single-country studies as comparison?
Although the research is based on a single country, it may involve different time frames, different regions, different groups of people etc.
(4.2) What are the advantages and disadvantages of single-country studies?
smaller generalisation, but you can also find out a very specific and detailed case for a single country,
Questions on: Gschwend, Thomas and Frank Schimmelfennig (2007): “Introduction: Designing Research in Political Science – A Dialogue between Theory and Data”.
(1) These authors discuss many of the same issues as Landman (2008), but use a somewhat different vocabulary and talk about large-n vs. small-n studies.
(1.1) How do the authors define large-n and small-n studies?
- Large-n studies are “data-set observations” (starting from 50 cases)
- Small-n studies are rather case analyses (within or across) (1 to 9 cases)
- Medium-n studies (10 to 49 cases)
(1.2) What, according to the authors, are the main similarities and differences between large-n and small-n studies?
(1.3) What, according to the authors, are the advantages and disadvantages of large-n vs. small-n studies?
- Large-n studies seek to achieve and increase the validity of causal inferences by increasing the number of cases and data-set observations
- Small-n research prefers depth do breadth, which leads to more precise causal explanations