Das Forschungsfeld der Bildungsforschung befindet sich seit der ersten PISA-Studie in einer starken Expansionsphase. Bisherige Analysen dieses Feldes beschränken sich allerdings auf allgemeine Übersichten zu verschiedenen Aspekten, etwa zum Personal, zur Finanzierung, zu Forschungsprojekten oder zu Publikationen. Das Ziel der Dissertation ist es dagegen, speziell die kognitiven Strukturen der Bildungsforschung in Deutschland, und hier insbesondere die in diesem Feld behandelten Forschungsthemen sowie seine Wissensbasis, mittels der bibliometrischen Methoden der Kowort- und der Referenzanalyse zu untersuchen. Die Datengrundlage bildeten zum einen die Schlagwörter zu 23389 Publikationen des Zeitraums 2000 bis 2009, die durch eine schlagwortbasierte Suche in der FIS Bildung Literaturdatenbank ermittelt wurden. Zum anderen die in den Beiträgen der dritten Auflage des Handbuchs Bildungsforschung zitierten 3921 Referenzen. Die Ergebnisse der Referenzanalysen zeigen eine nach wie vor sehr große Bedeutung der Publikationstypen Monographie und Sammelband und einen relativ starken nationalen Fokus im Zitationsverhalten, aber auch große Unterschiede in den Zitationsmustern zwischen den einzelnen Beiträgen. Zudem zeigt sich, dass die Schulleistungsstudien und die Bildungsberichte die einflussreichsten Publikationen der aktuellen Bildungsforschung darstellen und die Pädagogik, die Soziologie, die Psychologie sowie die Wirtschaftswissenschaften die wichtigsten Bezugsdisziplinen. In der Analyse der thematischen Schwerpunkte bestätigt sich die zentrale Rolle des Themenbereichs Schülerleistung und ist ein Bedeutungszuwachs der Forschung zum Thema Lernen zu erkennen. Aus den Analysen lässt sich folgern, dass Handbücher eine informative, aber noch wenig genutzte, Datenquelle der Wissenschaftsforschung darstellen. Ferner sind mit der Methode der Kowortanalyse auch differenziertere Untersuchungen von Forschungsfeldern, die nicht gut in Zitationsdatenbanken abgedeckt sind, möglich. / Since the first PISA study the field of educational research is going through a phase of strong expansion. Previous analyses of the field are largely confined to general overviews of various aspects like scientific staff, funding, research projects or publications. The aim of the thesis, in contrast, is to study the cognitive structures of educational research in Germany, in particular the research topics and the knowledge base of the field, using the bibliometric methods of co-word analysis and reference analysis. Two data sources were used: the keywords of 23389 publications from the period 2000 to 2009. These publications were identified by a keyword-based search in the German Education Index. And the 3921 references cited in the articles of the third edition of the Handbuch Bildungsforschung. The results of the reference analysis show a high relevance of the document types monograph and edited book and a relatively strong national focus but also considerable differences in the citation patterns of the individual articles. Furthermore, student assessment studies and education reports are the most influential publications of current educational research and pedagogy, socioloy, psychology and economics the main reference disciplines. The analysis of the thematic priorities reaffirms the central role of the topic student achievement and shows a growing importance of the issue of learning. It can be concluded that handbooks are an informative but still largely unexplored data source in science studies. Moreover, the method of co-word analysis facilitates more detailed investigations of research fields that are not well covered in citation indexes.

人文與社會科學引用資料類型反映了人文社會科學研究者的資料使用行為，進而突顯其出版作品之資料類型的重要性。學術研究之發展會隨著時間的進展而發生變化，因此，本研究以2000年與2009年為主要研究範圍，以二個年代加以比較各學科引用資料類型之差異。探討各學科引用文獻之資料類型、各學科引用圖書與期刊之變化及其是否因期刊性質與年代不同而有差異。所以，本研究亦將探討各學科期刊引用文獻老化現象，並進一步以期刊性質與年代加以比較，期能幫助研究人員與專家學者對於人文社會科學領域使用資料之行為更加了解，並協助圖書館館藏發展與管理的工作。 研究結果如下：1.人文與社會科學引用參考文獻分析(1)人文學以圖書為主要引用資源，社會科學以期刊為主；(2)評述型期刊的引用文獻數量較高；(3)人文學科的研究型期刊與評述型期刊多引用圖書資源，社會科學中多引用期刊資源；(4)不同年代的人文學評述型期刊與研究型期刊主要引用資料為圖書、期刊；社會科學不同年代的研究型期刊主要引用圖書及期刊資源，評述型期刊引用較多的是圖書與電子資源。2.人文與社會科學引用文獻老化現象呈現出(1)被引用圖書的文獻老化時間較長，其中又以人文學的圖書文獻老化所需時間比較多；(2)人文學的引用圖書資源的出版年份久遠，近期與過去出版的文獻皆有引用，文獻老化程度較不顯著；(3)人文學各學科的被引用期刊半衰期可達13年至19年；另一方面，社會科學各學科被引用期刊半衰期可達7年至19年間；(4)研究型期刊之文獻老化現象中，人文學科的引用年代偏向年代較早的文獻，傾向引用6年至22年的參考文獻作研究，而社會科學者經常引用出版6至11年的文獻；(5)評述型期刊之人文學科的引用年代傾向於引用12年間的參考文獻作研究，而社會學科學者經常引用出版8至10年的文獻；(6)研究表現出研究型和評述型期刊的類型並不完全影響該學科的文獻老化程度。 本研究結果可應用於規畫重要的期刊文獻類型，有助於圖書館或相關研究單位評估人文與社會科學的相關館藏是否足以支援研究，並且藉由與其他領域的老化速度之比較，作為為採購資料之時效性以及期刊裝訂時的必要考量。同時可將研究分析應用於人文社會科學學者的學術資訊需求與特性，藉以提供研究人員完善的資訊服務，由文獻老化速度可以知道學科的發展情形，可作為人文與社會科學學者館藏研究規劃之參考。 / The Humanities and Social Sciences reference categories reflect the use of information of the researchers and highlight the importance of the publication categories. The development of academic research changes over time. Therefore, taking 2000 and 2009 as the major research areas, the study compares the difference between each discipline referring categories. The study also researches the references patterns of different disciplines, the change of referring monographs and journals in different disciplines, and whether they change the pattern due to the transformation and aging of these reference books and journals. Consequently, this study will also probe into the Literature Obsolescence in various references categories of different disciplines, and furthermore to compare the nature and years of journals, hoping to help researchers and experts in humanities and social sciences to understand the information research behavior better and to assist the Library collection development and management. The results are as follows: 1.The analysis of Citation and Reference in Humanities and Social Sciences (1)the humanities mostly refer to monograph, and the social sciences mostly journal; (2)Review journal is in a higher number of citations; (3) research and review journals of Humanities cites more monograph resources, and social science journals resources; (4)Reference materials of review and research Journals of Humanities in different generations are mainly monographs and periodicals; research journals if social sciences in different years more cites in monographs and periodicals, while review periodicals prefer to reference more monograph and electronic resources. 2. Humanities and Social Sciences Citation Literature Obsolescence reveals that (1)the cited monographs take longer for obsolescence, among which monographs in Humanities take more time; (2) the cited monographs in humanities has published from long time ago. Cited recent and past literature, citation literature of aging are less significant. (3) The half-life of the cited journal of each discipline in the humanities is up to 13-19 years. On the other hand, the half-life of Journal reference in social sciences disciplines are up to 7-19 years; (4) About the Literature Obsolescence of research journals, the humanities tend to the refer literatures in earlier years, from past 6 to 22 years, while the Social Sciences often cited published literature published from 6 to 11 years ago; (5) Review journals in the humanities tend to quote the reference materials of past 12 years for research, and social science scholars frequently cited literatures published from 8 to 10 years ago; (6) The study demonstrated the research and the review types of periodicals are not completely affect the academic literature aging. The results can be applied to planning an important type of journal, contributing to the libraries or related research units to assess if the humanities and social sciences collection sufficient to support. Furthermore, by the comparison of other areas in speed of aging, this study also can be a necessary measurement for the timeliness of procurement and for the periodicals binding. Meanwhile, the analysis can be applied to the behavior and characteristics of academic information requirement of humanities and social sciences scholars, in order to provide researchers a complete information services. From the speed of Literature Obsolescence, scholars can acknowledge the development of a subject. It can be used as the reference in humanities and social sciences collection planning.

本研究透過書目計量法分析臺灣博碩士論文加值系統中引用光復後臺灣省省級機關(構)檔案情況。本研究目的有兩項:1.從不同典藏來源方式探討不同學位、校院系所、出版年份、研究內容時期之我國博碩士論文引用光復後臺灣省級機關檔案的分布情形；2.從不同的檔案全宗探討不同學位、校院系所、出版年份、研究內容時期之我國博碩士論文，引用光復後臺灣省級機關檔案的分布情形。 本研究所獲得的研究結論有六項:1.半數光復後臺灣省級機關檔案被博碩士論文引用類型是專書(49.4%)、專書與數位檔案(20.5%)；2.半數的博碩士論文(52.3%)引用臺灣省政府檔案史料彙編；3.臺灣師範大學(15.4%)、政治大學(10.3%)、臺灣大學(10.3%)、中央大學(10.3%)、臺南大學(9.6%)等校是引用光復後臺灣省級機關檔案較多的學校；4.從2006年起至2011年止出版的我國博碩士論文占74.4%，且引用光復後臺灣省級機關檔案呈現逐年增加的趨勢；5.博碩士論文研究內容時期，引用光復後臺灣省級機關檔案有56.2%最多；6.歷史類研究所產出的我國博碩士論文占51.9%，引用光復後臺灣省級機關檔案最多。 本研究建議有五項:1. 擴大檔案引用於國內各學科之系所博碩士論文;2.增廣博碩士生研究主題引用光復後臺灣省級機關的檔案;3.增進宣傳光復後臺灣省級機關檔案的典藏;4. 推廣數位化檔案檢索和引用;5.加強數位化典藏檔案的廣度。 關鍵字:引用文獻分析，檔案，博士論文，碩士論文 / Abstract The study analyzes the citations of the archives of Taiwan provincial level institutions in post Taiwan Retrocession from the National Digital Library of Theses and Dissertations in Taiwan using Bibliometrics. The purposes of the study are divided into the following two: 1. analyze the distribution of the archives of Taiwan provincial level institution Taiwan Historica in post Taiwan Retrocession, cited by Taiwan’s theses and dissertations with different degree, schools and departments, year of publication, and period of research content through the different sources of collection; 2. analyze the distribution of the archives of Taiwan provincial level institutions Taiwan Historica in post Taiwan Retrocession, cited by Taiwan’s theses and dissertations with different degree, schools and departments, year of publication, and period of research content through the various general archives. The study reaches the following six points of conclusions: 1. the majority of the archives of Taiwan provincial level institution Taiwan Historica in post Taiwan Retrocession by cited theses and dissertations are books (49.4%) and books and digital archives (20.5%);2. the majority of theses and dissertations (52.3%) cite from the preparation of the archives of Taiwan Historica;3. the majority of schools citing the archives of Taiwan provincial level institution Taiwan Historica in post Taiwan Retrocession include National Taiwan Normal University (15.4%), National Cheng Chi University (10.3%), National Taiwan University (10.3%), and National Central University (10.3%), and National University of Tainan (9.6%);4. between 2006 and 2011, about 74.4% of Taiwan’s thesis and dissertations published with citation of the archives of Taiwan provincial level institution Taiwan Historica accounts for 74.4% while the citation for the archives of Taiwan provincial level institution Taiwan Historica shows an upward increasing trend; 5. for the period of research content in theses and dissertations, about 56.2% of them cited the archives of Taiwan provincial level institution Taiwan Historica in post Taiwan Retrocession.;6. the majority of our theses and dissertations come from history institutes, accounting for 51.9% with most citations of the archives of Taiwan provincial level institution Taiwan Historica in post Taiwan Retrocession. The study makes the following five suggestions: 1. expand archives citation in these and dissertations from different departments and schools in Taiwan; 2. increase the theses and dissertation topics with citations of the archives of Taiwan provincial level institution Taiwan Historica; 3. increase collection that promotes the archives of Taiwan provincial level institution Taiwan Historica in post Taiwan Retrocession; 4. advocate digital archival query and citations; 5. strengthen the breadth of digital collection archives. Keywords : Citation Analysis; Archives; Doctoral Dissertations; Master's Theses.

O principal objetivo deste trabalho foi investigar as bases teóricas da metodologia de projeto, compreender sua evolução inicial e a influência de outras áreas nesse processo. Teve também como objetivo, descrever os primeiros métodos sistemáticos, presentes entre 1962 a 1973, os confrontado com as teorias presentes no referencial teórico. Isso foi feito através da análise de citações das seis primeiras obras publicadas sobre o tema somado ao registro da primeira conferência sobre métodos de projeto de 1962, criando uma base de dados de 1330 referências bibliográficas. O estudo identifica os principais campos que influenciaram os métodos, centrados em uma abordagem de sistemas. Além disso, revela uma lista de trabalhos e autores que precederam as publicações sobre métodos de projeto e que podem ser considerados os primórdios de um pensamento projetual. Essa lista, que pode ser considerada a principal contribuição do trabalho, foi feita através de técnicas de bibliometria, como a co-citação e o acoplamento bibliográfico, aliadas a análise de redes fazendo o cruzamento entre os autores, unindo dados quantitativos a uma análise qualitativa, esse resultado é contextualizado em uma linha do tempo e em um quadro histórico. Por fim é feita uma reflexão sobre a evolução e as tendências atuais e futuras da metodologia de projeto. / The main aim of this work was to investigate the theoretical bases of design methodology, understand their early evolution and the influence of other fields in this process. Had also aimed to describe the first systematic methods, present from 1962 to 1973, and confronting it with the theories present in the theoretical framework. This was done through citation analysis of the first six books published on the subject added to the published papers of the 1962 first conference on design methods, creating a database of 1330 references. The study identifies core fields that influenced the methods, focusing on a systems approach. Moreover, it shows a list of works and authors that preceded the publications on design methods and can be considered the beginnings of design thinking. The list, which can be considered the main contribution of the research was done through bibliometric techniques such as co-citation and bibliographic coupling, combined with network analysis making the crossing between the authors combining quantitative data with qualitative analysis, this results is contextualised into a time line and in a historical frame. At last was made a reflection about the evolution, current trends, and future of design methodology.

O principal objetivo deste trabalho foi investigar as bases teóricas da metodologia de projeto, compreender sua evolução inicial e a influência de outras áreas nesse processo. Teve também como objetivo, descrever os primeiros métodos sistemáticos, presentes entre 1962 a 1973, os confrontado com as teorias presentes no referencial teórico. Isso foi feito através da análise de citações das seis primeiras obras publicadas sobre o tema somado ao registro da primeira conferência sobre métodos de projeto de 1962, criando uma base de dados de 1330 referências bibliográficas. O estudo identifica os principais campos que influenciaram os métodos, centrados em uma abordagem de sistemas. Além disso, revela uma lista de trabalhos e autores que precederam as publicações sobre métodos de projeto e que podem ser considerados os primórdios de um pensamento projetual. Essa lista, que pode ser considerada a principal contribuição do trabalho, foi feita através de técnicas de bibliometria, como a co-citação e o acoplamento bibliográfico, aliadas a análise de redes fazendo o cruzamento entre os autores, unindo dados quantitativos a uma análise qualitativa, esse resultado é contextualizado em uma linha do tempo e em um quadro histórico. Por fim é feita uma reflexão sobre a evolução e as tendências atuais e futuras da metodologia de projeto. / The main aim of this work was to investigate the theoretical bases of design methodology, understand their early evolution and the influence of other fields in this process. Had also aimed to describe the first systematic methods, present from 1962 to 1973, and confronting it with the theories present in the theoretical framework. This was done through citation analysis of the first six books published on the subject added to the published papers of the 1962 first conference on design methods, creating a database of 1330 references. The study identifies core fields that influenced the methods, focusing on a systems approach. Moreover, it shows a list of works and authors that preceded the publications on design methods and can be considered the beginnings of design thinking. The list, which can be considered the main contribution of the research was done through bibliometric techniques such as co-citation and bibliographic coupling, combined with network analysis making the crossing between the authors combining quantitative data with qualitative analysis, this results is contextualised into a time line and in a historical frame. At last was made a reflection about the evolution, current trends, and future of design methodology.

O principal objetivo deste trabalho foi investigar as bases teóricas da metodologia de projeto, compreender sua evolução inicial e a influência de outras áreas nesse processo. Teve também como objetivo, descrever os primeiros métodos sistemáticos, presentes entre 1962 a 1973, os confrontado com as teorias presentes no referencial teórico. Isso foi feito através da análise de citações das seis primeiras obras publicadas sobre o tema somado ao registro da primeira conferência sobre métodos de projeto de 1962, criando uma base de dados de 1330 referências bibliográficas. O estudo identifica os principais campos que influenciaram os métodos, centrados em uma abordagem de sistemas. Além disso, revela uma lista de trabalhos e autores que precederam as publicações sobre métodos de projeto e que podem ser considerados os primórdios de um pensamento projetual. Essa lista, que pode ser considerada a principal contribuição do trabalho, foi feita através de técnicas de bibliometria, como a co-citação e o acoplamento bibliográfico, aliadas a análise de redes fazendo o cruzamento entre os autores, unindo dados quantitativos a uma análise qualitativa, esse resultado é contextualizado em uma linha do tempo e em um quadro histórico. Por fim é feita uma reflexão sobre a evolução e as tendências atuais e futuras da metodologia de projeto. / The main aim of this work was to investigate the theoretical bases of design methodology, understand their early evolution and the influence of other fields in this process. Had also aimed to describe the first systematic methods, present from 1962 to 1973, and confronting it with the theories present in the theoretical framework. This was done through citation analysis of the first six books published on the subject added to the published papers of the 1962 first conference on design methods, creating a database of 1330 references. The study identifies core fields that influenced the methods, focusing on a systems approach. Moreover, it shows a list of works and authors that preceded the publications on design methods and can be considered the beginnings of design thinking. The list, which can be considered the main contribution of the research was done through bibliometric techniques such as co-citation and bibliographic coupling, combined with network analysis making the crossing between the authors combining quantitative data with qualitative analysis, this results is contextualised into a time line and in a historical frame. At last was made a reflection about the evolution, current trends, and future of design methodology.

偉大的科學家牛頓曾說：「如果我能比別人看的更遠，那是因為我站在巨人的肩膀上」。任何一個學科的發展，其創始和演進的過程中，必定有一群具有前瞻性與領導風格的研究者，憑藉著他們本身的才能與智慧、努力的奉獻和卓越的理念做出具體的貢獻，透過相同的理念奠定學科的根基，並促使學科興盛發展。 本研究採用書目計量學研究方法中的共被引分析法，以美國資訊科學學會所頒發「最佳貢獻獎」之得主為研究對象，並以1900年至2007年為研究範圍，進行資訊科學家之間的主題性分析。本研究旨在探討44位資訊科學先驅的共被引情況，以瞭解各位先驅和各領域主題的群集現象，進而探討資訊科學的知識結構。本研究透過Web Of Science資料庫分別對於44位資訊科學家與個人被引用最多之著作進行共被引次數檢索，製成共被引矩陣，除了瞭解資訊科學家與其被引用最多之著作的共被引強度外，並以KNOT軟體進行路徑搜尋網路分析，進一步探討資訊科學之主題分佈情況，並與前人所做的相關研究進行比較。 本研究結果歸納如下： 一、資訊科學先驅的被引用次數和共被引次數 (一)被引用次數和共被引次數高低的差異，較高者可強化連結強度，較低者可反應出不同的學科結構。 (二)共被引次數較普遍者，具有跨領域的特性。 二、資訊科學先驅的路徑搜尋網路分析 (一)資訊科學領域以五大領域為主要範疇。分別是以Salton, G所領軍的資訊檢索領域；Lancaster, FW所代表的圖書資訊學與圖書館教育、圖書館自動化與自動化系統、圖書館與資訊服務評鑑等領域；Garfield, E為核心的資訊計量學與引用文獻分析之領域；Belkin, NJ所代表的使用者研究與資訊尋求行為領域；Saracevic, T為主的圖書資訊科學概念與理論、圖書資訊科學史的領域。 (二)單篇被引用最多之著作的網路圖無法判別。 三、本研究與White, HD (2003a)研究結果之比較結果 (一)資訊科學領域的學科結構並無太大變動。 (二)資訊科學中心點的不同。本研究所得的網路圖是以Salton, G為圖形的中心點，而White, HD (2003a)則是以Lancaster, FW為圖形中心點。 四、資訊科學領域有跨學科之性質。透過分析結果可以發現資訊科學有跨領域的特性，在許多的資訊科學先驅本身的研究範疇不只侷限於資訊科學，甚且有跨越其他學科主題，例如：電腦科學、社會網絡分析、數學、統計學等。 / Newton, one of the great scientists, said：「If I have seen further it is by standing on the shoulders of giants.」The development in any discipline, in the founding and evolution of the process , there must be a group of forward-looking with the leadership style of the researchers. They make concrete contribution by virtue of their own abilities and intelligence, outstanding efforts and dedication. They through the same idea lay the foundation of their subjects and flourish their disciplines. This study aims at using co-citation analysis of bibliometrics, and regards the Winners of Award of Merit of American Society for Information Science as the research objects to analyze the relevance between and between information scientists, and the research years from 1900 to 2007. The main purpose of this study is to explore the co-citation counts of 44 information scientists to understand the pioneers and the major subjects in information science, and further explore the structure of scientific knowledge of information science. This study through the Web Of Science database retrieve the co-citation counts of 44 pioneers of information science and the most cited of their works to build co-citation matrix. The result let us understanding the co-citation strength of 44 pioneers of information science. We also use the KNOT software in the analysis of PFNETs, try to explore the subjects how to distribute in information science, and take the results compare to the results of the study by White, HD in 2003. The results of the study reveal that: 1.Cited counts and co-citation counts of 44 pioneers of information science (1)The difference between higher counts and lower counts. The higher counts can strengthen the connections between the network, the lows counts can response to the different structures of information science. (2)The higher co-citation counts commonly have the characteristics of cross- discipline. 2.The PFNETs of 44 pioneers of information science (1)The fields of information science divide into five main areas. Salton, G lead the field of information retrieval. Lancaster, FW represent the education of library science and information science, library automation and automated systems, the evaluation for library and information services etc.. Garfield, E as the core of the field of informetrics and citation analysis. Belkin, NJ represent the field of the users research and information-seeking behavior. Saracevic, T represent the field of library information science concepts and theories, the history of library and information science etc.. (2)The network map of the most cited works of 44 pioneers of information science cannot be differentiated. 3. Take the results compare to the results of the study by White, HD in 2003 (1)The field of information science disciplines have not much change in the scientific structure. (2)Different focal point in the maps of information science. This study is based on the network map as Salton, G as focal point for the network, and White, HD (2003a) is based on the network map as Lancaster, FW as focal point for the graphics. 4. The field of information science has the nature of interdisciplinary. The results can be found through the co-citation analysis of information scientists on cross-cutting nature of information science. A pioneer in many areas of study itself is not limited to information science, and even likely to go beyond the subject of other disciplines, such as: computer science, social network analysis, mathematics, statistics etc..

L’objectif de cette étude est de saisir une image des éléments explicitement reconnaissables de la recherche en communication visibles dans les revues savantes Canadian Journal of Communication et dans Communication de 1974 à 2005. Il s’agit d’une analyse bibliométrique des articles publiés par les chercheurs d’institutions canadiennes et de leurs références bibliographiques. La bibliométrie est « l’application de méthodes statistiques aux livres et aux autres moyens de communication » (Pritchard, 1969: 348-349). C’est la première fois qu’une analyse de ce type est tentée dans ce corpus particulier. Nous nous sommes appuyés sur des postulats théoriques provenant de la sociologie des sciences et des études en communication scientifique. L’idée maîtresse est la suivante : l’activité scientifique est un « continuum de création de nouvelles connaissances » (Vassallo, 1999), dont l’organisation est basée sur l’échange d’information (Price, 1963; Crane, 1972), qui se traduit en reconnaissance sociale, en autorité scientifique, et constitue un investissement pour l’acquisition de crédibilité (Merton, 1938; Hagstrom, 1965; Bourdieu, 1975; Latour et Woolgar, 1986). À partir de l’analyse des articles, nous identifions s’ils sont le résultat de recherches empiriques ou fondamentales, ou le produit d’une réflexion critique. Il s’agit aussi de détecter les approches méthodologiques et les techniques d’investigation utilisées, ainsi que les sujets qui y sont abordés par les chercheurs. Nous détectons également les principaux lieux de recherche (universités et types de départements). Nous analysons aussi les thématiques des articles. Enfin, nous analysons des références bibliographiques des articles afin de cerner les sources d’idées qui y sont décelables. Notre corpus principal comporte 1154 articles et 12 840 titres de documents en référence. L’analyse bibliométrique des articles révèle ainsi une recherche canadienne en communication d’emblée qualitative, intéressée pour les spécificités historiques, le contexte social et la compréhension des interrelations sous-jacentes aux phénomènes de communication, en particulier, au Canada et au Québec. Au cœur de ces études se distingue principalement l’application de l’analyse de contenu qualitative dans les médias en général. Cependant, à partir de 1980, l’exploration du cinéma, de l’audiovisuel, des nouvelles technologies de l’information et de la communication, ainsi que la multiplication des sujets de recherche, annoncent un déplacement dans l’ordre des intérêts. Communication et le CJC, se distinguent cependant par l’origine linguistique des chercheurs qui y publient ainsi que dans les thématiques. L’analyse des références bibliographiques, et de leurs auteurs, met en relief l’intérêt partagé des chercheurs d’institutions universitaires canadiennes pour les agences de réglementation et les politiques gouvernementales canadiennes s’appuyant souvent sur l’analyse de documents législatifs et de rapports de diverses commissions d’enquête du gouvernement canadien. L’analyse révèle aussi les principales inspirations théoriques et méthodologiques des chercheurs. Parmi les plus citées, on voit Innis, McLuhan, Habermas, Tuchman, Bourdieu, Foucault, Raboy, et Rogers. Mais ces références évoluent dans le temps. On voit aussi une distinction relativement claire entre les sources citées par la recherche francophone et la recherche anglophone. / The aim of this study is to obtain an image of the recognizable elements of communication research visible in Canadian Journal of Communication and Communication from 1974 to 2005. This is a bibliometric analysis of the scientific papers and their bibliographies published in these journals by researchers from various Canadian Universities. Bibliometry is "the application of mathematics and statistical methods to books and other media of communication" (Pritchard, 1969). This is the first time that such analysis is attempted with this particular corpus. We based ourselves on theoretical postulates from the sociology of science and scientific communication studies, that indicate that the scientific activity is a « continuum de création de nouvelles connaissances » (Vassallo, 1999), where the organisation is based on the exchange of information (Price, 1963; Crane, 1972), and is in turn translated in social recognition, scientific authority, and is an investment to acquire credibility (Merton, 1938; Hagstrom, 1965; Bourdieu, 1975; Latour et Woolgar, 1986). Based on the article analysis, we can identify if they are empirical, theoretical or methodological researches. We can also detect the scientific methods applied by the researchers and their objects of study. Likewise, we observe the most common places to do research (universities and departments). Finally, we do a citation analysis to find the most important sources used by the researchers. Our corpus includes 1154 articles and 12 840 titles of referenced documents. The bibliometric analysis of the articles published in Canadian Journal of Communication and Communication from 1974 to 2005 shows that the Canadian communication research is qualitative in essence, interested in the historical particularities, the social context and the understanding of the underlying interrelations of the communication phenomenon, especially in Canada and Quebec. The application of content analysis to the media in general is dominant. However, since 1980, the exploration of cinema, audiovisual, internet and the multiplication of objects of study foreshadow the move in the research communication agenda in Canada. Communication and the CJC, however, differ by their thematics, and by the linguistic origin of the researchers who publish in them. The citation analysis shows the shared interest of researchers from different Canadian universities for regulatory agencies and Canadian government policies, which is often based on content analysis of legislative documents and reports of various commissions of inquiry of the Canadian government. The citation analysis also reveals the most importat theoretical and methodological research influences. Among the most cited, we see Innis, McLuhan, Habermas, Tuchman, Bourdieu, Foucault, Raboy, and Rogers. But these references are evolving over time. We see as well as the distinctions between French and English Canadian communication research traditions.

本研究利用書目計量學與引用文獻分析法分析引用1998年至2008年資訊科學重要期刊之特性。四種代表性期刊分別為Journal of the American Society for Information Science and Technology、Information Processing and Management、Journal of Information Science以及Journal of Documentation，研究樣本取自WOS資料庫，共計9579篇書目資料。希冀藉由分析探討引用資訊科學文獻之分佈現象與主題範疇，以觀察其他學科與資訊科學學科的互動關係。 本研究結果歸納如下：(1)資訊科學期刊被引用文獻篇數數量隨年代增長。(2)研究型文章為引用資訊科學期刊文獻最主要的資料類型。(3)本研究不完全符合布萊德福與布萊德福-齊夫定律，但可經由布萊德福-齊夫定律圖解，觀察出核心期刊有七種，核心期刊主題以圖書館學與資訊科學為主。 (4)引用資訊科學期刊文獻之學科大類分佈情況，以引用圖書館學門以及科學類為主，JOD是圖書館學門重要的參考期刊。(5)期刊及圖書主題分佈以被圖書資訊學主題和資訊檢索引用最多。(6)高被引用文獻以JASIST所佔比例最多，顯示JASIST是高生產力與高品質兼具的期刊，是資訊科學最重要且最受關注的文獻來源。(7)高被引主題以全球資訊網最多，其次為線上資訊檢索以及搜尋。(8)高被引作者共21位，目前皆任職於學校機構，以從事資訊科學、資訊研究領域為主，研究主題則是資訊檢索、資訊計量學等為多。這21位核心作者可以說是目前資訊科學界的核心領導作者。 本研究結果可應用於館藏發展與管理之參考，界定重要的期刊與圖書文獻來源，有助於圖書館或相關研究單位規劃、建立完整的資訊科學研究架構，以支援研究。同時，可將本研究之分析應用於資訊科學領域研究者，提供投稿之參考、了解目前研究主流以及未來的發展趨勢等，並可提供於資訊科學期刊編輯者，讓其了解該期刊之表現，適時調整，作為期刊長期追蹤的參考。 / Through bibliometric approach and citation analysis, this study analyzed the related disciplines and subjects of literature citing the important Information Science journals during 1998 to 2008. The four leading Information Science journals are Journal of the American Society for Information Science and Technology, Information Processing and Management, Journal of Information Science, and Journal of Documentation. A total of 9579 bibliographic data were from WOS database. Results show that：(1) The cited times of Information Science journals increased steadily. (2) Journal article is the main documentation type citing Information Science journals. (3)The distribution of literature citing Information Science journals does not exactly fit Bradford’s Law and Bradford-Zipf Law. However, by observation on Bradford-Zipf Law graph, there are seven core journals. The chief fields of core journals are Library Science & Information Science, and Information Science. (4) Primary disciplines of literature citing Information Science journals are Library Science & Information Science, and Science. JOD is the important reference journal in Library Science & Information Science discipline. (5) The subjects of journals and books highly citing Information Science journals are Library Science & Information Science, and Information Retrieval. (6) JASIST is highly cited, showing that it is both a high productivity and high quality journal. (7) The highly cited subject is World Wide Web, and followed by information retrieval and search. (8) The highly cited authors, who mainly serve in college and most of whom hold the post in the related Information Retrieval and Information Study departments, defined in this study amount to 21. Most of them specialize in information retrieval and informetrics. Finally, the results of this study can be applied to collection development and management, and identify the core journal and books. It assists libraries or information centers in planning and establishing complete research framework on Information Science.

本研究旨在應用共被引分析法與社會網絡分析法，以資訊社會代表文獻為研究對象，進行文獻間的主題相關性分析，希望透過對資訊社會代表文獻之共被引分析與社會網絡分析，能瞭解資訊社會研究所涵蓋的學科領域、文獻之間的集群性、及歷年研究趨勢。本研究以1962年至2010年的資訊社會代表文獻為研究範圍，研究資料取自Webster於2006年所撰寫的《Theories of Information Society》與《The Information Society Reader》二本著作，經整合後得出89篇代表文獻做為本研究之研究樣本。其次，再以WOS資料庫對此89篇代表文獻進行共被引次數檢索，製成共被引矩陣。除了瞭解代表文獻間的共被引強度外，並以SPSS軟體進行相關係數分析與集群分析，此外，利用UCUNET軟體計算出文獻中心性，及將文獻間的關係繪製成多元尺度圖與社會網絡圖，最後，透過研究年代的區隔，計算不同時段共被引情形的變化，以觀察資訊社會領域發展趨勢。 本研究結果歸納如下：1.資訊社會代表文獻被引用次數概況：（1）整體而言，資訊社會代表文獻歷年被引用次數趨於穩定；（2）資訊社會代表文獻受到社會學與地理學領域高度引用。2.「資訊社會」之知識地圖：（1）資訊社會代表文獻共被引次數普遍偏低；（2）資訊社會領域之核心文獻；（3）資訊社會領域可分為社會學、地理學及資訊科學與圖書館學子群。3.資訊社會之發展趨勢：（1）資訊社會知識地圖演變；（2）資訊社會共被引次數衰退。 本研究結果可應用於館藏規劃與評估，界定重要的圖書文獻來源，有助於圖書館或相關研究單位評估資訊社會相關館藏是否足以支援研究。同時可將研究之分析應用於資訊社會學術研究，提供研究人員近年來資訊社會共被引學科之變化及研究主題趨勢等相關資訊，作為資訊社會學者進行研究規劃之參考。 / The purpose of this study is to find out what disciplines Information Society covers. By Co-citation Analysis, this study highlighted the major disciplines in the Information Society field and identified the main literature and their relationship. This research takes representative Information Society literature from 1962 to 2010 as research scope. Firstly, by Bibliometrics, the total of 89 representative literature of Information Society was extracted from “The Information Society Reader” and “Theories of Information Society” written by Frank Webster in 2005 and 2006 respectively. Secondly, the co-citation times between 89 literature was investigated through Web of Science and thus a co-citation matrix was build to reveal the co-citation strength of literature. Co-relation and cluster analysis between literature were also explored by SPSS software. In addition, this study uses UCUNET software to analyze centrality and plot knowledge map on the subject of Information Society. The major findings are as follows: 1. On the citation strength: (1) Cited times of Information Society literature have been growing stable in recent years. (2) Information Society literature is highly cited by disciplines of Sociology and Geography. 2. Knowledge map of Information Society: (1)Co-citation frequencies of Information Society literature are low;(2) The core literature of Information Society field; (3)Sociology, Geography and Information Science are main disciplines in Information Society area. 3. Development of Information Society: (1) Development of knowledge map of Information Society; (2) Co-citation frequencies of Information Society literature have been declining in recent years. Finally, the results of this study can be applied to collection planning and assessment, and identify the core journals and books. It assists libraries or information centers in evaluating the adequacy of Information Society collections to support research. Meanwhile, it provides researchers with recent trends of Information Society research.