探索臺灣大學人文社會高等研究院學者研究產出之總合軌跡

Abstract

人文學與社會科學學者研究產出的評鑑是近年來廣受爭議的課題，雖然引用文獻分析因其客觀的特質而常成為學術評鑑的方法，但是經常被認為僅是一種可能的方法，不可過於偏重。隨著網路的發展，社群媒體工具提供了不同於傳統的計量指標與途徑，雖可用以評估學術的影響力，但是少有研究同時運用並比較前述二種方法，而總合軌跡正是說明傳統正式學術傳播影響與非正式學術傳播影響之匯總呈現。 本研究以臺灣大學人文社會高等研究院學者為代表，探索人文學與社會科學學者研究產出之總合軌跡，進行學術被引與網路使用的比較分析。在蒐集與建立學者著作清單之後，選擇THCI、ACI、WOS與Scopus四個引文索引資料庫做為研究工具，以探討人文學與社會科學學者研究產出在國內與國外的被引情形；同時亦嘗試透過社群媒體工具Mendeley，來分析學者研究產出在網路上的使用情形，並且比較學術被引與網路使用兩者的異同。 研究結果顯示，學者著作的語言以中文為主，數量是英文的3倍，資料類型以期刊論文最多，年代分布主要集中在2001-2010年之間。在國內引文索引資料庫中，學者的中文著作遠比英文著作更常被引；國內引文索引資料庫中的被引筆數，也遠多於國外引文索引資料庫，顯示學者研究產出的影響力仍多侷限於國內。另一方面，學者著作在Mendeley的使用筆數與WOS、Scopus的被引筆數相近；國外引文索引資料庫的被引著作數量與Mendeley的學者著作數量高度相關，顯示Mendeley與國外引文索引資料庫一樣能反映著作在國外的學術影響力。 本研究建議對人文學與社會科學學者研究產出的評估，若採引用文獻分析，應以國內引文索引資料庫為主，國際性引文索引資料庫為輔。並由於altmetrics採行不同於傳統引文分析的方法，因此建議也使用如Mendeley等社群媒體工具，應用網路使用分析來評估人文學與社會科學學者研究產出之影響力。

The evaluation of research output for scholars of humanities and social science has been a controversial issue in recent years. Citation analysis has been heavily used as the method of evaluating research output because of the objective characteristics. However, it is also often considered just one of the possible approaches. With the development of the Internet, social media (as opposed to traditional metrics) provide alternative ways of assessing the research impact, but few studies have simultaneously adopted these two approaches. Therefore, the composite trace approach demonstrates both traditional scholarly impact and alternative metrics. In this study, we investigate the research outputs by comparing how the researchers’ publications were cited in academic journals and how they were used on the Web, as examplified by researchers at Institute for Advanced Studies in Humanities and Social Sciences, National Taiwan University. After compiling a list of researchers' publications, we used the citation databases of THCI, ACI, WOS and Scopus to analyze the pattern of citation distributions domestically and internationally. We also chose the social media, Mendeley, to examine the Web use of the researchers' publications. Finally, we compared the results of citation analysis and Web usage. The research results show that most publications are written in Chinese, which is three times more than English. The major type of publications is journal article. Most journal articles are published between 2001 and 2010. The results also reveal that research outcomes have greater impact domestically than internationally because of the fact that Chinese publications are more likely to be cited and there are more cited references in local citation databases than international citation databases. On the other hand, there is little difference between the number of the publications collected by Mendeley users and the number of cited references in WOS or Scopus. In addition, the number of the cited references in international citation databases is highly correlated with the number of counts in Mendeley. Therefore, the findings suggest that Mendeley, an alternative to citation database, could be able to desmonstrate the international impact of academic research. It is reocommended that if citation analysis is chosen for the assessment of humanities and social science scholars, the local citation databases should be used as the primary data sources and international citation databases could be used as auxiliary data sources. Since altmetrics provides different aspect from the traditional citation analysis, data from social media such as Mendeley are also recommended.