地理学报 ›› 2012, Vol. 67 ›› Issue (8): 1109-1124.doi: 10.11821/xb201208009

• 区域经济发展 • 上一篇    下一篇

基于报纸媒体内容分析和信息熵的广交会综合影响力时空分布

戴光全1,2, 谭健萍1,2   

  1. 1. 华南理工大学经济与贸易学院, 广州 510006;
    2. 华南理工大学现代服务业研究院, 广州 510006
  • 收稿日期:2012-01-06 修回日期:2012-03-21 出版日期:2012-08-20 发布日期:2012-08-20
  • 基金资助:
    国家自然科学基金项目(40971090/D010202); 广东省教育部产学研结合项目(2009B090200062)

The Spatial-Temporal Distribution of Integrated Impact Index for the Canton Fair: A Study Based on Content Analysis Method (CAM) & the Information Entropy of Chinese Complete Newspaper Database (CCND)

DAI Guangquan1,2, TAN Jianping1,2   

  1. 1. School of Economics and Commerce, South China University of Technology, Guangzhou 510006, China;
    2. Academy of Modern Service Industry, South China University of Technology, Guangzhou 510006, China
  • Received:2012-01-06 Revised:2012-03-21 Online:2012-08-20 Published:2012-08-20
  • Supported by:
    National Natural Science Foundation of China, No.40971090/D010202; The Cooperation Project in Industry, Education and Research of the Ministry of Education in Guangdong Province, No.2009B090200062

摘要: 从报纸媒体的角度研究广交会作为重大事件的综合影响力的时空分布特点。以中国知网(http://www.cnki.net/) 所属“中国重要报纸全文数据库”(CCND) 2000-05-05 到2011-05-30 时间段内“广交会”为主题的4333 条新闻报道为研究文本, 通过传播学中的内容分析法建构分析类目, 统计类目包括新闻的时间、所属的届数与期数、报社所在城市、新闻版面、新闻字数等, 共由3 个编码员进行统计, 并对类目编码进行信度检测。在对新闻内容进行计量统计的基础上, 引进信息熵度量信息量, 利用信息熵和报道篇数构建广交会综合影响力指标, 进一步构建广交会空间综合影响力指数方程, 分析广交会综合影响力在全国的分布情况, 并检验综合影响力指标和经济指标的相关性。结果发现:① 从时间分布来看, 以新闻报道数量来衡量, 在事件(广交会) 进行时其综合影响力存在时间衰减现象, 各展期综合影响力的年际变化则呈现为周期性的年度重复现象。② 从空间分布来看, 广交会综合影响力空间分布与城市参展商数量、经济情况和距离有着密切的联系:受影响大的城市主要集中在沿海经济发达的省份, 受影响小的城市主要分布在西部和北部, 整体上综合影响力大小和城市分布的密度由东到西递减;综合影响力总体上存在显著的空间距离衰减, 但对于北京和上海这两个特殊城市而言, 空间距离已经不是广交会综合影响力的主要因素, 而在广交会举办地的“当地”(广东省内) 较短的距离内, 距离衰减并不显著, 当空间距离超出一定范围, 则存在明显的距离衰减现象。

关键词: 中国进出口商品交易会(广交会), 重大事件, 报纸媒体, 时空分布, 内容分析法, 综合影响力指标, 信息熵

Abstract: This paper studies spatial-temporal features of its integrated impact index (III) by using the Content Analysis Method (CAM), the concept of the information entropy measurement (Hartley) and statistical analysis through SPSS. The fundamental data are searched from the Chinese Complete Newspaper Database (CCND) by the theme words "Canton Fair" (CF, 广交会), referring to the records of its news reports in newspapers between May 5, 2000 and May 30, 2011. The number of records in total is 4333.
The findings are as follows. (1) Temporal distribution: The number (N) of the news increased year by year at first and reached the peak in 2006, and then began to decrease. The news reports concentrated in April and October every year. During the fair, the number of news reports of every CF's session decreased from the first phase to the third one. And the interannual change shows an annual periodic duplication by the number of news reports during the three phases. (2) Spatial distribution: The number (N) of news reports about CF and its Hartley of a given region/city are introduced into the CF's III on this region/city. The results show that the newspaper publishing companies are mainly located in Beijing, Shanghai and Guangzhou, and that the local/regional ones are concentrated in the east coastal provinces: Guangdong, Zhejiang, Jiangsu and Fujian. The city's III decreases from east to west. This result is similar to the exhibitors' spatial distribution of the 109th Canton Fair. (3) There are some relationships between economic development and distance with the III's spatial distribution reflected by correlation analysis. But Beijing and Shanghai do not obey the distance decay rule. For these two cities, distance is not an important factor affecting the spatial distribution of their IIIs. It is also found that the distance decay rule does not work at a short range within its host province, whilst it works very well beyond certain distance within all the cities except Beijing and Shanghai.
This study employs the perspective of media, as the media management is by all means a critical consideration of the event organizers. But nowadays with the acceleration of the information transition the media management may be unable to control the spatial distribution of impacts. This viewpoint needs more researches in the future. In addition, the intrinsic factors, such as operation, and extrinsic factors, such as economic structure and information technology development, have significant influence on the impacts of the events. The relationship between these factors is worthy of further study.

Key words: China Import and Export Fair (Canton Fair), mega-event, newspaper media, spatial-temporal distribution, Content Analysis Method (CAM), integrated impact index (III), information entropy (Hartley)