Editing Democratic peace theory (section)

===Statistical significance===
One study has argued that there have been as many wars between democracies as one would expect between any other couple of states. Its authors conclude that the argument for democratic peace "rests in an ambiguity", since empirical evidence not confirm neither deny democratic pacifism, and strongly relies upon what degree of democracy makes a government democratic; according to them "because perfect democracy is infeasible, one can always sidestep counter-evidence by raising the bar of democracy".{{sfn|Schwartz|Skinner|2002}}

Others state that, although there may be some evidence for democratic peace, the data sample or the time span may be too small to assess any definitive conclusions.{{sfn|Spiro|1994}}{{sfn|Gowa|1999}}{{sfn|Small|Singer|1976}} For example, Gowa finds evidence for democratic peace to be insignificant before 1939, because of the too small number of democracies, and offers an alternate [[#Realist explanations|realist explanation]] for the following period.{{sfn|Gowa|1999}} Gowa's use of statistics has been criticized, with several other studies and reviews finding different or opposing results.{{sfn|Gelpi|Griesdorf|2001}}{{sfn|Ray|2003}} However, this can be seen as the longest-lasting criticism to the theory; as noted earlier, also some supporters agree that the statistical sample for assessing its validity is limited or scarce, at least if only full-scale wars are considered.{{Citation needed|date=January 2023}}

According to one study, which uses a rather restrictive definition of democracy and war, there were no wars between jointly democratic couples of states in the period from 1816 to 1992. Assuming a purely random distribution of wars between states, regardless of their democratic character, the predicted number of conflicts between democracies would be around ten. So, Ray argues that the evidence is statistically significant, but that it is still conceivable that, in the future, even a small number of inter-democratic wars would cancel out such evidence.{{sfn|Ray|2003}}{{efn|The 1999 [[Kargil War]], subsequent to the period considered, does satisfy the objective requirements for democracy and war set in Ray's study.}}