SUITABILITY OF CONFLICT DATA FOR POWER LAW FOR TURKEY AND A GROUP OF COUNTRIES

Today, many social problems involve constant violence towards an individual or a group. These problems are; domestic violence, sexual violence against women, civil attacks, violent conflicts, violent protests, acts of terrorism, and national security. In particular, governments and policy makers have developed strategies through the measurement of these violent events, resulting in the most harmful consequences of many social events. Today, the power law distribution is used to measure the severity of these violent incidents. In general, the power law defines that the frequency of an event is decreasing faster than the increasing size of the event. Technically, a power law refers to the mathematical relationship between two quantities that change as the power of one s other. In this study, the poweRlaw package of R, which simplifies power laws and other long tail distributions, is used. Using the conflict data obtained from GDELT between 2014 and 2017, Iraq, Syria, Turkey, Palestine, Ukraine, Afghanistan, Israel, and Russia are tested whether their conflict data is suitable for power law. The results of the analysis of the conflicts in Turkey and Afghanistan are not suitable for power law distribution; the conflicts in Iraq, Syria, Palestine, Ukraine, Israel and Russia are found to be suitable for distribution of power law. It is concluded that the distribution of power law does not depend on the scale, that is, the data obtained from various places has the same distributional characteristics.