後冷戰時期影響國際恐怖主義發生因素之探討─應用事件史分析法之研究

A Study on the Factors Affecting International Terrorism Occurrence in the Post Cold War Era─An Application of Event History Analysis

李佳怡(Chia-yi Lee)

指導教授 ： 黃旻華

國立臺灣大學/社會科學院/政治學研究所/碩士(2007年)

https://doi.org/10.6342/NTU.2007.10481

摘要

本文的研究目的在於利用事件史分析法找出後冷戰時期影響國家發生國際恐怖主義的主要因素，研究的時間範圍是1992年至2002年。利用ITERATE資料庫中這段時間的資料，本文建立了假設國家間有差異性的Cox共享脆弱性模型，估計各個因素對國際恐怖主義事件發生風險率的影響程度，並進一步將事件畫分成不同種類的恐怖活動，運用同樣的模型來檢視之，除了各項因素的影響力外，也探討國際恐怖事件在時間上的變化。
結果發現影響國家發生國際恐怖主義的因素主要有三個，分別是民主程度、中東地區與否、失業率，三者都對恐怖主義的風險率有正向影響力。此外，若將所有國際恐怖事件依攻擊手段和活動方式的不同，又可分成人質事件、爆炸事件、武裝攻擊事件、劫機事件、暗殺事件和威脅事件六大類，分別受不同因素所影響，其中影響人質事件最主要的因素是民主程度、失業率和種族齊一性；影響爆炸事件的因素則是都市化程度、國家硬體能力和陸地面積；而武裝攻擊事件則易發生在都市化程度較低和雖民主但尚不穩定的國家；至於和劫機事件的風險率有負相關的因素是國家硬體能力，正相關的則是人口和陸地面積；暗殺事件易發生在GDP成長率低、失業率高的國家；威脅事件則受民主程度和中東地區的正向影響。
此外，除了人質事件外，其餘五項事件多半集中在冷戰後三、四年，之後幾年的事件數則相對較少。而實際檢視資料後則發現，前幾年的事件數攀高是來自於特定恐怖團體所為，或是來自冷戰餘波的政治影響，而隨著冷戰的氣氛逐步消解、和平氛圍漸長，以及部分恐怖團體和政府和談成功後，恐怖事件在後幾年則顯著較少且較隨機。

關鍵字

國際恐怖主義 ； 恐怖事件 ； 事件史分析法 ； 風險率 ； Cox共享脆弱性模型

並列摘要

The purpose of this paper is applying event history analysis to finding out the factors which affect international terrorism occurrence in the post cold war era. The observed time is from 1992 to 2002. Using ITERATE data set during this period, this paper builds Cox shared frailty models, which assume that there are differences between countries, and estimates the effects of factors on the hazard rates of international terrorism occurrence. Furthermore, terrorism events are classified into six types and Cox shared frailty models are used to estimate the effects of different factors on them. Besides, the variations of the numbers of international terrorism events over time are considered in this paper.
The result shows that there are three factors affecting international terrorism occurrence, including democratic degrees, middle-east countries, and unemployment rates. These three factors have positive effects on the hazard rates of international terrorism occurrence. Moreover, according to incident types, terrorism events are classified into hostage events, bombing events, armed attack events, hijacking events, assassination events, and threat events, and the factors affecting these six events are different. There are three factors affecting hostage events, including democratic degrees, unemployment rates and ethnic homogeneities. Bombing events are affected by urbanization degrees, national material capabilities and land areas. Armed attack events occur in countries with low urbanization and unstable democracy more probably. National material capabilities have negative effects on hijacking events, but land areas and populations have positive effects on them. Assassination events occur in countries with low GDP growth rates or high unemployment rates more possibly. Threat events are affected by democratic degrees and middle-east countries positively.
Finally, most events occurred during the first three or four years after the end of cold war except for hostage events, and the numbers of events declined afterwards. According to the data, the fact that events occurred mostly during early years in the post cold war era comes from specific terrorist organizations or the political aftermath of cold war, and because the atmosphere of cold war went down, peaceful atmosphere increased, and some terrorist groups negotiated with countries successfully, the numbers declined in the late years.

並列關鍵字

international terrorism ； terrorist events ； event history analysis ； Cox shared frailty model ； hazard rate

參考文獻

江啟臣，2005，〈APEC架構下反恐議題發展之研究〉，《政治科學論叢》，第25期，頁29-66。

Ho, Szu-yin. 2005. “The Financial Front of War on Terrorism.” EurAmerica, Vol. 35, No. 4 (Dec., 2005), pp. 673-687.

張中勇，2002，〈國際恐怖主義的發展與演變〉，《戰略與國際研究》，第4卷，第1期，頁1-58。

廖育德，2005，《台灣上市上櫃電子業廠商首次投資中國大陸行為之動態研究──事件史分析法之應用》，國立成功大學都市計劃學系碩博士班93學年度碩士論文。

Bremer, Stuart A.. 1980. “National Capabilities and War Proneness.” in J. David Singer ed., The Correlates of War, pp. 57-82. New York: Free Press.