Arab World English Journal (AWEJ) Special Issue on CALL No.2 July, 2015 Pp. 226 – 229
PDF
Book Review
Social Network Analysis
Author: John Scott
Title of the Book: Social Network Analysis
Year of Publication: 2013
Publisher: Sage, United KingdomNumber of Pages: 201
Reviewer: Dr. Nadia IDRI, University of Bejaia.
In such a digitalized and modernized world, social relations have known a new development and became characterized by online world-wide levels. This digital form of the world is what we concretize by means of social networking and social media that are becoming omnipresent in the individual’s daily life. Hence, methods and approaches to analyze such new social structures of this new social life are permanently required. The existence of Social Network Analysis (SNA) becomes a necessary issue for not only researchers but also for practitioners in many disciplines. Stemming from this principle, we consider Scott’s third edited book on SNA after its publication in 1991 and 2000 as an updated resource for all social network analysists.The book is written in nine chapters and 185 pages. Scott, in his third edition updated and revised its content according to the changes that took place during the last decade in the field of SNA. The author facilitated its usage through its well-organized form and simplified language. That is, the book contains helpful indexes for authors and subjects, provides supplement of comments and notes on each chapter as well as the usefulness and the objective of each chapter. The book includes diversified references that are more likely helpful in understanding social relations from both a diachronic and synchronic standpoints. Scott’s analysis of the phenomenon stems from the origin of social relations’ analysis and structures of these relations since the issue is not novel in social anthropology mainly. Then, Scott expanded his explanations to online social structures; a phenomenon that knows a rapid change in the last decade. That what makes his references stem from 1908 to 2012; where he presented the evolution of the phenomenon through a diachronic analysis of social networking. For this reason, he added a totally new chapter compared to the two first editions where he focused on recent changes on network dynamics and change over time.
The book represents and intellectual enterprise in the field of SNA. Scott reported historical knowledge about social networks’ development. To achieve such a goal, he enriched the related sources to this phenomenon, questioned the way social networks has been analyzed putting focus on the rapid development in this concern. To achieve such an aim, the author offered an explicit explanation of key concepts employed in assessing network structure such as density centrality, cliques and blocks. In addition, he tried to translate mathematical discussion into more comprehensible words using a less technical form this is more likely to broaden the book’s readership especially that Scott used multidisciplinary resources and illustrations that ranges from nursing, biology, anthropology, economics, etc. It is evident that all disciplines make use of social networks since the 1970’s where SNA appeared. In this epoch, a wide number technical work and specialist applications saw light to analyze the nature of social actions, social structure and relations that characterize social networking websites like facebook and twitter. However, the provided analysis methods required mathematical knowledge and highly technical literature existed and this led Scott to write the book and help researchers with little mathematical knowledge. In this book, Scott presented different types of data appropriate to SNA. Here, details about kinship patterns, community structure, interlocking directorships are highlighted.
In his first chapter, Scott introduced the many types of data that call upon the use of various types of analysis methods; “relational data”, “ideational data” “attribute data”. The book provides illustrations about existing program for analysis such as UCINET and PAJEK. Scott also put focus on socio-metric analysis using graph theory methods (Chapter Two). Then, the author has gone through interpersonal configurations and cliques. In this, he focused on different approaches from various disciplines naming formal models, the Harvard breakthrough and social physics (Chapter Two). All the above content makes part of the two first chapters. The first is devoted to networks and relations whereas the second deals with the development of SNA.
After the presentation of SNA and its development, Scott moved in his third chapter to relational data and its analysis. According to Scott, SNA is more appropriate for relational data and the methods depend on the availability of these data rather than attribute data. The author worked this chapter to help the user of his handbook learn how to collect this kind of data, store and prepare it for SNA. Sampling is highlighted in relation to specific areas before dealing with the preparation and the organization of relational data; which according to Scott, are not covered in existing texts the thing that led him cover a number of such issues as gaining access, designing questionnaires, sampling, non-response treatment, etc. (Chapter 3). All in all, this chapter is devoted to handling relational data in a matrix form focusing upon transposing, adding and multiplying of matrices.
The handbook offers a simplified presentation of the graph theory which is based on the sociogram use where matrices are used and translated into formal concepts and theorems. This is the starting point of any SNA. The graph theory is generally used to simplify the complex mathematical nature of the analysis to describe and analyze networks and their features in a more general manner. Scott illustrated his explanations related to graph theory and sociograms with different types of matrices and graphs such as adjacency matrix, alternative graph diagrams, directed graph (Chapter 4). In doing so, Scott has gone through the explanation of related terms like the concept of density; as crucial in the theory. This concept was elucidated from ego-centric and socio-centric perspectives. Here, the needed formulas with examples were presented the thing that makes the book serves as a guide for its user. Then, the author explained density by presenting the digression on absolute density for the problem of size and mass measurement with the needed formulas to do that. He ended the chapter with community structure and density where facebook, twitter were given as examples of social structures and interpersonal relations.
Given that social networks are bound to the principle of sociometry, and that there is always a centre of interest or the “star”, centrality or a point of centrality is part of SNA (Chapter 5). This centre can either be an individual person or an organization. In this chapter, Scott distinguished between local centrality and global centrality. When measuring the point of centrality, it is then the overall “centralization” of the graph; which is another concept clarified by Scott.
Social networks are also divided into cohesive sub-groups and “cliques”. They generally inform about the person’s or the group’s identity and sense of belonging. However, the different social worlds give permanently birth to a huge amount of emerging sub-groups and, hence, the appearance of a wide range of theoretical models for analysis. These are described as “cliques”, ‘clusters’, ’components’, ‘cores’ and ‘circles’ (Chapter 6). Scott presented the theoretical bases in this chapter where components, cycles and knots; cliques’ intersections; and components in citation circles were tackled.
As a follow up of the theoretical explanation of cliques, clusters and components; the author went further to detail the analysis of positions within social structures. In SNA, dealing with structural equivalence is needed when dealing with roles and social positions. In addition, although many concepts that describe these social structures appear, they are sometimes misused or used and starts identifying the concepts and their uses. A case in point is the “cluster” and the “clique”. Hence, the author focused on clusters in terms of combining and dividing points, the block modeling wit CONCOR; the ‘CONVergence iterated CORelations’; the algoritm that uses correlation coefficients as measures of similarity. Here, Scott presented various models like the simple block model, the hierarchical block model. There is also a section in this chapter about corporate interlocks and participations
As a living phenomenon, social networks can in no way remain treated as static since they are dealing with social structures and interpersonal relations. However, SNA has moved beyond descriptive ends in the analysis and took a cross-sectional dimension. Scott did not ignore this important development in the methods and approaches in SNA and attempted to present this change in the processual models that rely on statistical techniques and relate them to the recent work on the ‘small-world’ issues (Chapter 8).
A detailed last chapter with illustrative diagrams about networks to facilitate the analysis through pictorial representations is provided. Scott enriched this ninth chapter with figures that clarify any kind of analysis. He also provided the reader with explanations about the ‘multi-dimensional scaling’ mathematical approach where concepts like distance, space and metrics are clarified. Then, principal component analysis or the so called ‘factor analysis’; though different they are, is also tackled. Scott presented the non-metric methods and went through the advances in network visualization. Here, many programs are available like VIEWNET, MOVIEMOL, MAGE, SONIA, PAJEC. The UCINET, for instance, the program helps to visualize the network, you can pull up the data in NetDraw (the network visualization program packaged with UCINET) and code for each relationship and show all of the relationships together in one sociogram or separately
In his first chapter, Scott introduced the many types of data that call upon the use of various types of analysis methods; “relational data”, “ideational data” “attribute data”. The book provides illustrations about existing program for analysis such as UCINET and PAJEK. Scott also put focus on socio-metric analysis using graph theory methods (Chapter Two). Then, the author has gone through interpersonal configurations and cliques. In this, he focused on different approaches from various disciplines naming formal models, the Harvard breakthrough and social physics (Chapter Two). All the above content makes part of the two first chapters. The first is devoted to networks and relations whereas the second deals with the development of SNA.
After the presentation of SNA and its development, Scott moved in his third chapter to relational data and its analysis. According to Scott, SNA is more appropriate for relational data and the methods depend on the availability of these data rather than attribute data. The author worked this chapter to help the user of his handbook learn how to collect this kind of data, store and prepare it for SNA. Sampling is highlighted in relation to specific areas before dealing with the preparation and the organization of relational data; which according to Scott, are not covered in existing texts the thing that led him cover a number of such issues as gaining access, designing questionnaires, sampling, non-response treatment, etc. (Chapter 3). All in all, this chapter is devoted to handling relational data in a matrix form focusing upon transposing, adding and multiplying of matrices.
The handbook offers a simplified presentation of the graph theory which is based on the sociogram use where matrices are used and translated into formal concepts and theorems. This is the starting point of any SNA. The graph theory is generally used to simplify the complex mathematical nature of the analysis to describe and analyze networks and their features in a more general manner. Scott illustrated his explanations related to graph theory and sociograms with different types of matrices and graphs such as adjacency matrix, alternative graph diagrams, directed graph (Chapter 4). In doing so, Scott has gone through the explanation of related terms like the concept of density; as crucial in the theory. This concept was elucidated from ego-centric and socio-centric perspectives. Here, the needed formulas with examples were presented the thing that makes the book serves as a guide for its user. Then, the author explained density by presenting the digression on absolute density for the problem of size and mass measurement with the needed formulas to do that. He ended the chapter with community structure and density where facebook, twitter were given as examples of social structures and interpersonal relations.
Given that social networks are bound to the principle of sociometry, and that there is always a centre of interest or the “star”, centrality or a point of centrality is part of SNA (Chapter 5). This centre can either be an individual person or an organization. In this chapter, Scott distinguished between local centrality and global centrality. When measuring the point of centrality, it is then the overall “centralization” of the graph; which is another concept clarified by Scott.
Social networks are also divided into cohesive sub-groups and “cliques”. They generally inform about the person’s or the group’s identity and sense of belonging. However, the different social worlds give permanently birth to a huge amount of emerging sub-groups and, hence, the appearance of a wide range of theoretical models for analysis. These are described as “cliques”, ‘clusters’, ’components’, ‘cores’ and ‘circles’ (Chapter 6). Scott presented the theoretical bases in this chapter where components, cycles and knots; cliques’ intersections; and components in citation circles were tackled.
As a follow up of the theoretical explanation of cliques, clusters and components; the author went further to detail the analysis of positions within social structures. In SNA, dealing with structural equivalence is needed when dealing with roles and social positions. In addition, although many concepts that describe these social structures appear, they are sometimes misused or used and starts identifying the concepts and their uses. A case in point is the “cluster” and the “clique”. Hence, the author focused on clusters in terms of combining and dividing points, the block modeling wit CONCOR; the ‘CONVergence iterated CORelations’; the algoritm that uses correlation coefficients as measures of similarity. Here, Scott presented various models like the simple block model, the hierarchical block model. There is also a section in this chapter about corporate interlocks and participations
As a living phenomenon, social networks can in no way remain treated as static since they are dealing with social structures and interpersonal relations. However, SNA has moved beyond descriptive ends in the analysis and took a cross-sectional dimension. Scott did not ignore this important development in the methods and approaches in SNA and attempted to present this change in the processual models that rely on statistical techniques and relate them to the recent work on the ‘small-world’ issues (Chapter 8).
A detailed last chapter with illustrative diagrams about networks to facilitate the analysis through pictorial representations is provided. Scott enriched this ninth chapter with figures that clarify any kind of analysis. He also provided the reader with explanations about the ‘multi-dimensional scaling’ mathematical approach where concepts like distance, space and metrics are clarified. Then, principal component analysis or the so called ‘factor analysis’; though different they are, is also tackled. Scott presented the non-metric methods and went through the advances in network visualization. Here, many programs are available like VIEWNET, MOVIEMOL, MAGE, SONIA, PAJEC. The UCINET, for instance, the program helps to visualize the network, you can pull up the data in NetDraw (the network visualization program packaged with UCINET) and code for each relationship and show all of the relationships together in one sociogram or separately
(Durland, 2013). Final notes about elites, communities and influence are also presented in this book in its last section.
To conclude, the book is an excellent resource that serve workers in SNA whether a beginner or not in this field or in mathematical knowledge. It is an updated source and contain authentic resources.
To conclude, the book is an excellent resource that serve workers in SNA whether a beginner or not in this field or in mathematical knowledge. It is an updated source and contain authentic resources.
Reference
Durland, M. (2013, May, 3). SNA TIG Week: Maryann Durland on Analyzing Multiple Relationships in UCINET [Web log post]. Retrieved fromhttp://aea365.org/blog/?p=8933&utm_source=feedburner&utm_medium=email&utm_campaign=Feed%3A+aea365+%28AEA365%29
Reviewer:Dr. Nadia Idri
Faculty of Arts and LanguagesUniversity of Bejaia, Algeria