We address the problem that occurs when inferences about counterfactuals—predictions, ‘‘what-if’ ’ questions, and causal effects—are attempted far from the available data. The danger of these extreme counterfactuals is that substantive conclusions drawn from statistical models that fit the data well turn out to be based largely on speculation hidden in convenient modeling assumptions that few would be willing to defend. Yet existing statistical strategies provide few reliable means of identifying extreme counterfactuals. We offer a proof that inferences farther from the data allow more model dependence and then develop easyto-apply methods to evaluate how model dependent our answers would be to specified counterfactuals. These methods require neither sensitivity testing over specified classes of models nor evaluating any specific modeling assumptions. If an analysis fails the simple tests we offer, then we know that substantive results are sensitive to at least some modeling choices that are not based on empirical evidence. Free software that accompanies this article implements all the methods developed. 1

Inferences about counterfactuals are essential for prediction, answering ‘‘what if ’ ’ questions, and estimating causal effects. However, when the counterfactuals posed are too far from the data at hand, conclusions drawn from well-specified statistical analyses become based on speculation and convenient but indefensible model assumptions rather than empirical evidence. Unfortunately, standard statistical approaches assume the veracity of the model rather than revealing the degree of model-dependence, so this problem can be hard to detect. We develop easy-to-apply methods to evaluate counterfactuals that do not require sensitivity testing over specified classes of models. If an analysis fails the tests we offer, then we know that substantive results are sensitive to at least some modeling choices that are not based on empirical evidence. We use these methods to evaluate the extensive scholarly literatures on the effects of changes in the degree of democracy in a country (on any dependent variable) and separate analyses of the effects of UN peacebuilding efforts. We find evidence that many scholars are inadvertently drawing conclusions based more on modeling hypotheses than on evidence in the data. For some research questions, history contains insufficient information to be our guide. Free software that accompanies this paper implements all our suggestions. Social science is about making inferencesFusing facts we know to learn about facts we do not know. Some inferential targets (the facts we do not know) are factual, which means that they exist even if we do not know them. In early 2003, Saddam Hussein was obviously either alive or dead, but the world did not know which it was

The paper outlines and compares two models of how globalization is likely to affect the risk of civil war -- a liberal model and structuralist model. Overall, we find considerably more support for the liberal model than for the structuralist, anti-globalist model. Trade does appear to have a capacity for increasing internal peace -- not directly, but via trade's beneficial effects on growth and increased political stability. Overall, we find economic openness to be associated with higher growth. Our results give no support to the idea that globalization reduces growth, not even for poor countries. We found some evidence that trade increases income inequality. However, in contrast the robust link established between income inequality and violent crime, we do not find any relationship between inequality and civil war. In sum, the beneficial effect of trade and foreign investment outweighs whatever violence may be generated by increased inequality. We find that economic openness is associated with greater stability of political systems. This effect is particularly strong for democracies, but also positive for inconsistent regimes and autocracies. Finally, in our analysis of the factors increasing the likelihood of civil wars, we find no direct impact of economic openness. However, countries with a high income per capita and a stable political system have considerably lower risk of civil war than those without. Hence, since we find economic openness to increase average income and political stability, we do find an indirect conflict-reducing effect of globalization.

Simulation & Information

In this article, we show that de Marchi, Gelpi, and Grynaviski’s substantive analyses are fully consistent with our prior theoretical conjecture about international conflict. We note that they also agree with our main methodological point that out-of-sample forecasting performance should be a primary standard used to evaluate international conflict studies. However, we demonstrate that all other methodological conclusions drawn by de Marchi, Gelpi, and Gryanaviski are false. For example, by using the same evaluative criterion for both models, it is easy to see that their claim that properly specified logit models outperform neural network models is incorrect. Finally, we show that flexible neural network models are able to identify important empirical relationships between democracy and conflict that the logit model excludes a priori; this should not be surprising since the logit model is merely a limiting special case of the neural network model. We thank Scott de Marchi, Christopher Gelpi, and Jeffrey Grynaviski (2004; hereafter dGG) for their careful attention to our work (Beck, King, and Zeng 2000; hereafter BKZ) and for raising some important methodological issues that we agree

This study extends the early-warnings ap roachp resented by O'Brien (2002), which examined macrostructural factors to forecast the intensity level of country-specific instability. This analysis adop ts O'Brien's pattern class ification algorithm and sp litsamp le validation design to establish baseline forecasting r esults. It additionally ap lies maximum-likelihood ordered logistic estimation as an easy-to-use alternative top attern classification that allows for twop rimary revisions and extens ions to this line of research. First, this study revises the selection of variables to better facilitate cross-country comp arisons of the effects of macrostructural factors. Second, it draws up on sp line regression methods top rovide new emp irical evidence for establi shed theories that relate social homogeneity to regional instability. The resulting model ma intains reasonable forecasting accuracy while quantifying the relationship s betwee n macrostructural factors and regional instability. Thep ractical imp lication is to foc us analysts not only on highp riority geograp hic regions, but also on those factors that might escalate or mitigate the likelihood of conflict.

Trade, foreign investment, and other forms of economic interdependence have grown throughout the post-World War II period, along with a stronger global political consciousness and increased regional cooperation. After the end of the Cold War, not only have these phenomena accelerated, but the lack of any opposing

Abstract: This paper offers a framework for analysing the effects of armed conflicts on households and the ways in which households in turn respond to and cope with the conflicts. It distinguishes between direct and indirect effects, and shows that the indirect effects are channelled through (i) markets, (ii) political institutions, and (iii) social networks. Drawing upon the recent empirical literature, the paper portrays the processes running along these various channels and offers policy suggestions to be adopted at both national and international levels.