|
Citation analysis--the exploration of reference patterns in the scholarly and scientific literature-- has long been applied in a number of social sciences to study research impact, knowledge flows, and knowledge networks. It has important information science applications as well, particularly in knowledge representation and in information retrieval. Recent years have seen a burgeoning interest in citation analysis to help address research, management, or information service issues such as university rankings, research evaluation, or knowledge domain visualization. This renewed and growing interest stems from significant improvements in the availability and accessibility of digital bibliographic data (both citation and full text) and of relevant computer technologies. The former provides large amounts of data and the latter the necessary tools for researchers to conduct new types of large-scale citation analysis, even without special access to special data collections. Exciting new developments are emerging this way in many aspects of citation analysis 1. Foundations of citation analysis -- 1.1 Introduction -- 1.2 What is citation analysis? -- 1.3 What can we do with citation analysis, and why? -- 1.3.1 Assessing information resources and evaluating scholarly contributions -- 1.3.2 Mapping research fields -- 1.3.3 Tracking knowledge flows and the diffusion of ideas -- 1.3.4 Studying users and uses of scholarly information -- 1.3.5 Assisting information organization, representation, and retrieval -- 1.3.6 Other applications -- 1.4 Evaluation of citation analysis -- 1.4.1 Validity and reliability -- 1.4.2 Critiques and defense -- 1.4.3 Strengths, limitations, and special care required -- 1.5 Related fields -- 1.6 Scope, delimitation, and structure of this book -- 2. Conducting citation network analysis: steps, concepts, techniques, and tools -- 2.1 Field delineation: collecting scholarly publications produced in a research field or by a scholarly community -- 2.2 Selecting core sets of objects for the study -- 2.2.1 Selection criteria -- 2.2.2 Number of objects -- 2.2.3 Types of objects -- 2.2.4 Citation counting -- 2.3 Measuring the connectedness between core objects selected -- 2.3.1 Citation count -- 2.3.2 Co-citation count -- 2.3.3 Bibliographic coupling frequency (BCF) -- 2.3.4 Other similarity indicators derived from citation counts -- 2.4 Statistical analysis of citation networks -- 2.4.1 Commonly used statistical analysis methods -- 2.4.2 Factor analysis in author co-citation analysis (ACA) -- 2.4.3 Input data to statistical procedures -- 2.4.4 Determining similarity measures -- 2.5 Network analysis and visualization -- 2.6 Interpretation and validation -- 2.7 An example: mapping information science 2006-2010: an author bibliographic coupling analysis -- 2.7.1 Delineation of the information science research field -- 2.7.2 Selection of core authors to represent the research front of IS 2006-2010 -- 2.7.3 Measurement of the connectedness between core authors -- 2.7.4 Factor analysis -- 2.7.5 Visualization of factor structures -- 2.7.6 Interpretation of results -- 3. Field delineation and data sources for citation analysis -- 3.1 Commonly used approaches to field delineation -- 3.2 Effects of field delineation on citation network analysis -- 3.3 Requirements for data sources for citation analysis -- 3.4 The ISI databases -- 3.4.1 Advantages of using the ISI databases for citation analysis studies -- 3.4.2 Problems -- 3.4.3 How to delineate a research field using the ISI databases -- 3.5 Scopus -- 3.5.1 Pros and cons as a data source for citation analysis -- 3.5.2 How to delineate a field using Scopus -- 3.6 Field delineation by combining citation databases with subject bibliographic databases -- 3.6.1 Justification of combining Scopus and PubMed for delineating the stem cell research field -- 3.6.2 Process of matching between Scopus and PubMed for field delineation -- 3.6.3 Dataset obtained and its advantages for citation analysis -- 3.7 Field delineation and in-text citation analysis -- 3.7.1 Feasibility and benefits of in-text citation analysis -- 3.7.2 Limitations of in-text citation analysis -- 3.8 Field delineation with Google Scholar and other citation data sources on the web -- 3.8.1 Benefits -- 3.8.2 Problems -- 3.9 Other approaches to field delineation -- 3.10 Additional remarks -- 4. Disambiguation in citation network analysis -- 4.1 Introduction -- 4.2 Names and designations in bibliographic records -- 4.3 The name ambiguity problem -- 4.4 Ambiguity and power laws -- 4.5 Effects of ambiguity on network analysis results -- 4.6 Manual disambiguation -- 4.6.1 Small networks -- 4.6.2 Large networks -- 4.7 Algorithmic disambiguation -- 4.7.1 Author name disambiguation -- 4.7.2 Citation link disambiguation -- 4.8 Back to the future: computer-aided disambiguation -- 5. Visualization of citation networks -- 5.1 Why citation network visualization? -- 5.2 Cautionary tales -- 5.2.1 Database artifacts -- 5.2.2 Visualization artifacts -- 5.2.3 Informative vs. heavy links -- 5.3 Three decades of co-citation network visualizations of the library and information science field -- 5.4 Visualization of citation networks using Pajek -- 5.4.1 Factor analysis of bibliometric data -- 5.4.2 Conversion of factor analysis results from SPSS to Pajek network format -- 5.4.3 Visualization with loading summaries as node sizes and degree coloring -- 5.4.4 Visualization with node sizes reflecting citedness -- 5.4.5 Visualization with node color reflecting factor membership -- 5.4.6 Combining pattern and structure matrix visualizations -- 5.4.7 Fine-tuning the maps -- 5.4.8 Visualization of bibliometric networks without factor analysis -- 5.5 Concluding remarks -- Appendix 3.3 -- Appendix 5.4.2 -- Bibliography -- Author biographies
|
