淡江大學機構典藏:Item 987654321/34281
English  |  正體中文  |  简体中文  |  全文笔数/总笔数 : 62805/95882 (66%)
造访人次 : 3933704      在线人数 : 461
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜寻范围 查询小技巧：
  • 您可在西文检索词汇前后加上"双引号"，以获取较精准的检索结果
  • 若欲以作者姓名搜寻，建议至进阶搜寻限定作者字段，可获得较完整数据
    •  进阶搜寻


    jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/34281


    题名: 學術論文被引用次數之分析研究：以Science Citation Index Expanded及Scopus為例
    其它题名: An analytical study of citedness score on scholarly literatures: based on science citation index expanded and scopus
    作者: 石秋霞;Shih, Chiu-hsia
    贡献者: 淡江大學資訊與圖書館學系碩士班
    黃鴻珠;Huang, Hong-chu
    关键词: 學術論文;被引用次數;科學引文索引;SCIE;Scopus;citedness score;citation counts;scholarly literatures;SCIE;Scopus
    日期: 2005
    上传时间: 2010-01-11 05:06:59 (UTC+8)
    摘要: 論文被引用次數常被用為評估學術研究成果的質，對學術界的重要性不可言喻，為求公允，引用次數的運算宜力求正確，然而，由於提供查詢論文被引用次數之引文索引、索引摘要或網路搜尋引擎等引用資訊系統，系統發展背景不同，其製作程序與資料處理原則相異，使得各篇論文被引用次數在不同系統之檢索結果因而產生差異。

    本研究除藉由文獻探討外，選擇淡江大學工學院獲得校內研究獎助之學術論文372篇與中國大陸、臺灣地區所出版的251種工程期刊為研究樣本，以SCIE與Scopus兩個不同時代背景所發展的系統作為研究對象，分別透過實際檢索論文與期刊的被引用情形，比較與分析SCIE與Scopus系統計算論文被引用次數存在的現象與問題，以歸納影響引用資訊系統計算論文被引用次數之因素，期作為國內學術評鑑體制與未來改善現有系統或建置新系統之參考，並提出改進的建議。

    由於影響系統計算論文被引用次數之因素太多，包括：引用資訊系統收錄資料的範圍、系統著錄各篇論文之參考文獻的完整性、書目記錄的資料品質以及文獻連結機制等四方面，造成不同系統所檢索的論文被引用次數相異。本研究據實證研究所得到的結果與發現，提出七項結論：一、學術論文被引用次數不單取決於引用資訊系統收錄的期刊或其他資料類型的種數，各刊期完整或選擇性收錄與系統收錄資料的類型等原則亦會影響引用次數的運算。二、從核心期刊觀察學術論文的被引用情況，確實符合經濟效益。三、參考文獻之著錄項目與格式影響被引用次數的計量。四、引用資訊系統著錄參考文獻時，對文獻單位的認定原則不同，影響被引用次數之統計。五、引用資訊系統對書目記錄的查核與修正原則，亦是影響論文被引用次數產生差異之因素。六、原始文獻的引用錯誤，易影響引用資訊系統計算論文被引用次數之正確性，尤其是利用書目資料比對進行串連文獻機制之系統。七、引用資訊系統所產生的製作疏失影響其計算論文被引用次數之程度較原始文獻引用錯誤之情況高。

    根據研究結果，本論文提出五項建議，分別為：一、引用資訊系統應重視會議論文與其他資料類型的引用，建議增加除期刊論文以外的資料類型之收錄。二、加強引用資訊系統中書目記錄的資料品質控制，以提昇系統計算論文被引用結果之正確性。三、引用資訊系統應著錄參考文獻的篇名，可彌補刊名與作者以縮寫名稱著錄的不足，俾便於論文被引用結果之判別。四、引用資訊系統對於參考文獻之認定應有一致的原則。五、引用資訊系統宜採用記錄辨識號碼取代書目資料比對方式，作為串連引用文獻與被引用文獻的連結機制。
    Citedness score is a widely accepted metric for the quality of scholarly literatures, therefore, the forming calculation and corresponsive consequence of citation counts becomes of importance for academic researchers, especially for the research assessment. However, the distinction of citedness score exists in heterogeneous information systems resulted from considerations for system construction and development, such as purpose, procedure and policy of data processing, and that would bring various meanings and impacts for citedness score.

    This paper aims to examine the distinctive factors and results on calculation of citedness score of scholarly literatures between SCIE and Scopus by adoption of comparison as research methodology. In this empirical study, we selected 372 research papers from research grant publications of the College of Engineering at Tamkang University, and 251 engineering journals published in Taiwan and the Mainland China as research subject to test the citation counts from SCIE and Scopus simultaneously. As a consequence, one may generalize that four factors heavily impacts on the different citedness scores between SCIE and Scopus and factors are: coverage of collected data, integrity of cited references, quality of citation record and citation pointer between citing and cited references.
    Based on our results in this study, one may draw conclusions in the following:
    1.In addition to the amount of scholarly literatures determined by citation information system producers, citation counts are also varied resulted from the selection policy, especially concerning about comprehensiveness of journal coverage and type of scholarly literatures, such as conference proceedings and so on.
    2.It has proven to be economical effectiveness for the use of core journals, according to the citation counts both of SCIE and Scopus.
    3.Record fields and style of cited references collected by information systems has a far–reaching influence on calculation of citation counts.
    4.The distinctive recognition on granularity of references for scholarly literatures between SCIE and Scopus also deeply impact the calculation of citation counts.
    5.The policy of quality assurance on data verification and correction is another influential factor resulting in various calculations of citation counts.
    6.Errors stemmed from misquotation of original scholarly literatures get the accuracy of citation counts into disputable.
    7.The errors from production of information system are much worse than those stemmed from misquotation of original scholarly literatures, such as omission of cited references, record linkage errors and so on.

    Based on the conclusions above, five suggestions for improving citedness score of information systems are raised as follows:
    1.To expand the data coverage of journals into the other scholarly literatures, such as conference papers.
    2.To enhance quality assurance on citations.
    3.To record the fullness of paper title and author rather than in an abbreviation ways, in order to facilitate the identification of cited and citing references.
    4.To build up a consistent policy for inclusion of cited references.
    5.To adopt record identifier mechanism for reference pointer function, instead of reference-matching one.
    显示于类别:[資訊與圖書館學系暨研究所] 學位論文

    文件中的档案:

    档案 大小格式浏览次数
    index.html0KbHTML353检视/开启

    在機構典藏中所有的数据项都受到原著作权保护.

    TAIR相关文章

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - 回馈