Network Analysis (Statistical Associates Blue Book Series 30)

NETWORK ANALYSIS

Network analysis has its roots in sociometry, developed by Jacob L. Moreno in the 1930s. Sociometry was closely associated with small group research and a focus on interpersonal choices within affiliation networks. While some use the term "sociometry" to refer to all research using quantitative scales, the term "sociography" is sometimes used to refer to a method of presenting data about complex individual relationships and networks in graph form. In addition to its social scientific purposes, discussed below, sociometric assessment of interpersonal choices has also played a role in therapy by helping facilitate constructive change in individuals and groups through greater interpersonal awareness. For this reasons, in some circles the term sociometry refers to a form of therapy related to psychodrama.

In modern usage, the term network analysis has largely supplanted the earlier term sociometry, though both involve analysis of social networks by statistical and graphical methods. Today, network analysis usually refers to quantitative analysis of relationships among network nodes (actors or objects) based on mathematical graph theory. The main purposes of this quantitative approach are (1) description and comparison of networks based on various coefficients such as those regarding the centrality or non-centrality of nodes; and (2) visualization of the connectedness of nodes in graphical form. UCINET is perhaps the current leading software package for network analysis of this type.

Network analysis may be applied to affiliation networks such as those connecting Facebook friends, flow networks such as commercial trading relationships, reference networks such as scholarly citation patterns, linkage networks such as webpage links, grids such as power grids, and many other network types.

TABLE OF CONTENTS

Table of Contents
Overview 4
Key Concepts and Terms 4
Sociometric tests 4
Sociometric representation 5
Network data diagrams 6
Network graphs 7
Network data matrix formats 7
Network Analysis with UCINET 11
Overview 11
Data input 12
Network visualization 15
Statistical analysis 16
Centrality measures 24
Statistical analysis 27
Assumptions 28
Measurement inerrancy 28
Correlated data 28
Group size 29
Model specification 29
Ecological fallacy 29
Frequently Asked Questions 29
Where can I find out more about network analysis? 29
What computer programs exist to generate sociograms or similar representations? 30
What is role analysis in sociometry? 31
Bibliography 31
Pagecount: 35