RPP #6

My research is interested in tracing out the processes of peace between rising and ruling powers. Since this is a fairly specific set of cases (rising and ruling powers) constrained by a fairly rare condition (peace), zooming out and redefining the project is required for a statistical research design. Fundamentally, my research is still interested in war and peace, for which there is ample data. The Correlates of War Project offers valuable datasets on variables associated with conflict. Its database on National Material Production includes information on the material capacity of a country to wage war. Another one of its databases is Inter-States Wars, which provides data on the duration, outcomes, and instigations of war. Full citations to these datasets can be found in the bibliography.

The National Material Production dataset offers variables that can help understand the military capacity of a specific country. The variables most pertinent to my research are military expenditure and military personnel. Military expenditure, before 1914, is measured in thousands of current year British Pounds, while post-1914 is measured in thousands of current year U.S. Dollars. Military personnel is measured in thousands of persons. These variables offer some indication of the size and strength of a military, both of which are important factors during a war. The database offers statistics on most countries from 1816-2012, meaning there are thousands of cases. One limitation of this database is that military expenditure is not weighted proportionally to a state’s wealth or productivity. A smaller state will not be able to spend as much on its military as a larger state with access to more resources.

The most pertinent information from the Inter-State Wars database is the outcome of war variable and the instigation of war variable. The outcome of war variable is measured specific to the country on an ordinal scale. These outcomes include winning, losing, drawing, stalemates, and more. The database has 338 cases. One limitation of the database is the lack of nuance associated with a nominal variable of starting a war. The database assigns this blame to only one country using 1 as yes (X country did start the war) and 2 as no (X country did not start the war).

Potentially, I could design a research project that looks at how much of an impact (if any) National Material Production has on the outcomes of Inter-State Wars. For instance, does a difference in National Material Production between two states affect the outcome of their conflicts? Presumably, a bigger difference in National Material Production (measured by military expenditure and military personnel, which should give an estimate of the size and power of a country) would result in clearer victories. Furthermore, a smaller difference would indicate fewer material advantages between either side and result in conflicts with more ambiguous endings such as stalemates or draws.



Maoz, Zeev, Paul L. Johnson, Jasper Kaplan, Fiona Ogunkoya, and Aaron Shreve. “The Dyadic Militarized Interstate Disputes (MIDs) Dataset Version 3.0: Logic, Characteristics, and Comparisons to Alternative Datasets,” Journal of Conflict Resolution, (forthcoming 2019).

Singer, J. David. “Reconstructing the Correlates of War Dataset on Material Capabilities of States, 1816-1985,” International Interactions, no. 14 (1987), 115-32.

2 thoughts on “RPP #6

  1. Hi Jordan!

    This may be partially due to my lack of familiarity with your project, but I am just a little confused about what exactly the variables you identify are testing. I also had a fairly limited and specific scope in my project at first, but it seems like you did a good job at zooming out and finding an appropriate variable from reliable datasets. As you move forward I might just clarify what you anticipate your variables to indicate. Good luck!

  2. Jordan — the data sources and datasets that you identify here are suitable for your project as you’ve reconceptualized it for the large-n methodology that we are studying (but remember — you must provide footnote citations in your writing when you reference sources). As Hannah notes, though, there is some ambiguity in the variables here. It’d be good to have a more focused operationalization of your DV (remembering that you really need a DV operationalized as an *interval-ratio* indicator for this methodology). What would that DV/operationalization for this version of your project look like?

Leave a Reply

Your email address will not be published. Required fields are marked *