Comparative Analysis of Environmental Policy Network in Japan, USA, Germany and Korea

韩国和日本在过去三年间进行的全球环境政策网络调查的主要结果如下(我们还有另外两个关于美国和德国的正在进行的项目采用相同的研究框架)；(I)在选择在全球大气政策制定中最有影响力的行为者之前，我们要求大约10名外部评审员和所有相关行为者自己在该组织内进行评估。根据两组评价之间的相关性超过0.7的计算，认为遴选过程是适当的。(2)韩国和日本的行为者构成模式非常相似。以下是按份额排列的参与者构成：1.非政府组织、政府相关行为者；2.商业协会、政府智库和大企业；3.大众媒体、商业高峰协会、咨询委员会。(3)相当相关的行为者之间的同行评估显示，日本和韩国在感知的影响力方面略有不同。在韩国，大众媒体和政治家以及国际组织和非政府组织的影响相对比日本更大。(4)为了澄清信息网络的作用，我们考察了评价的分数与网络相关变量的关系，如“指数(被他人引用的次数)”，以及变量如何在这样的网络中显示出特殊的紧急性质，特别是“广泛性”。高度的相关性表明，到目前为止，这一领域的信息网络假设是稳健的。(V)关于信息交换和支持网络的宏观结构，根据分析得出的具体结果将很快公布。

The main findings of the Global Environmental Policy Network surveys conducted between Korea and Japan over the past 3 years are shown as follows (we have 2 other ongoing projects concerning the US and Germany which apply the same research framework) ;(i) Prior to the selection of most influential actors in the global atmospheric policy-making, we requested both around 10 outside referees and all the related actors themselves to evaluate within the group on the other. The selection process was seen to be appropriate based on the calculation that the correlation between the 2-group evaluations was more than 0.7.(ii) The patterns of the constitution of actors in Korea and Japan were quite similar. The following list is the constitution of those actors in order by share : 1. NGOs, Government-related actors, 2. Business associations, governmental think tanks, and big enterprises, 3. Mass media, Business peak associations, Advisory councils.(iii) The peer-evaluation among fairly relevant actors shows a slight difference in the perceived influence between Japan and Korea. In Korea, the influence of mass media and politicians as well as international organizations and NGOs are relatively stronger than in Japan.(iv) To clarify the role of the information network, we examined how the points of the evaluation correlate to the network-related variables such as the "indegree (the number of being referred to by others)" and how the variables show particular emergent property in such a network, notably "cantrality". The high degree of correlation indicates that the assumption of the information network in this area so far is robust.(v) As for macro-level structure of information exchange and support networks, the tangible result based on the analysis will be reported shortly.