Introduction
Over the past decade, there have been dozens of quantitative studies dealing with the onset of civil wars. Through the quantitative studies on civil wars, significant determinants have been proven statistically. This has been possible through the separate studies on the duration and onset of civil war as the determinants for these two processes are different. However, the separate studies have often resulted in the difficulty to find the any statistically significant association between the onset and duration of the war. This due to the selection effect, which has continually been inherent in the duration models, which make it hard to differentiate between statistically, and substantive significant relationships. Practically duration and the onset of war should be studied together to obtain the correct reports on the civil wars at any point in time for any country.
Literature Review
Civil war prevalence is the likelihood of observing two separate events at any given time t: where t is onset time, t -1 is the time that peace has prevailed and the continuation of war at any given time is given by t. Taking this into consideration it gives a correlation on the prevalence of civil war in many countries. Some of the theories of war models are also explained such as how economics acts as a big motivator of civil wars. Greed and grievance have been the big motivators in the duration and onset of most of the civil wars.
The analysis used is based on the theoretical synthesis in relation to the influential model on the onset of civil war and the model on the duration of civil war. The article does not aim at advancing the theoretical knowledge but the available theories are used for the empirical analysis of the civil war prevalence from 1960 to 1999 in the populace of countries. The data used in the journal is organized in panels of 5 years. This data has been drawn Collier and Hoeffler (2000) and improvements made on it. This data has then been used to investigate the impact of ethnic divisions and democracy on the prevalence of civil war. In the methodology no new estimator is developed but the easily usable estimators are used which are more appropriate than the estimators which are pooled probit in the analysis of the civil war prevalence. The random effects probits model (RE probit) is used in the comparison with other alternative estimators. Some of the explanatory variables are further explored using the concepts of problem of endogeneity. With the assumption of endogeneity, it reduces the significance of democracy and changes the results in the core model of war prevalence.
Works Cited
Elbadawi, I., & Sambanis, N. (2002). How much war will we see? Explaining the prevalence of civil war. Journal of Conflict Resolution, 46, 307-334.
