different data used as control variables or dependant variables. Analyses that use the Correlates of War data on international conflict end at 1992, the year of the last available update. Similarly, the Penn World Tables are the source for the wealth data and they begin in 1950, eliminating two years from the beginning of the analysis when these data are used. Every effort is made to maximize the period of each analysis and consequently the temporal domain will sometimes vary between tables.
There is a more general temporal limitation for this entire study. A crucial aspect of the theoretical foundation for many of the hypotheses tested is that free presses are able to exchange information. The information has to flow quickly enough to effect the policy-making process in both countries. It also has to move in sufficient quantities to allow the free press in each country to influence content of the news media in the other. The study is limited to post--World War II international politics to ensure that no problems are created by presses that simply could not communicate with each other. Any effort to push the analysis back prior to World War II will have to scrutinize every case to ensure that the free presses in the nations involved were able to communicate effectively.