| Member Profile : Martin Everett |
|
 | Contact Information | Address: -Map Me-
Martin Everett
University of Manchester, Sociology
Arthur Lewis Building
Bridgeford Street
Manchester, Lancashire, United Kingdom M13 9PL
Phone : +44 161 275 2515
E-mail : martin.everett@manchester.ac.uk
|
| Bibliographic Information |
|
| | | Software & Data | | Active Calendar Listings |
|
|
|
| | | Network Graduate Programs | | Network Courses |
|
|
|
| | | Jobs Posted | | Sunbelt Submissions |
|
| Sunbelt XXIX - March 10 to March 15, 2009 - Bahia Hotel | | Abstract : Missing data some reflections. |
Obtaining reliable, complete and accurate data has always been a challenge for social network analysts. Data is often incomplete for a variety of reasons. Typically the few studies that have been published examine two broad categories of missing data. In particular missing edges caused by non-response of survey data, simply missed in observational data or just not recorded in archival data. Secondly missing actors these were either not included in the study initially, not picked up by the initial scoping exercise (Eg missed in a snowball sample, not picked up be a name generator) or possibly excluded for practical reasons. There are three basic approaches to deal with missing data.
· Develop and or use techniques that are robust with respect to missing data.
· Try and recover the missing data either by using alternative data sources (informants, alternative data collection methods, etc) or by using techniques designed to recover the data from what is already known and/or collected.
· The third approach is to acknowledge the fact that the data is not complete but then to interpret and or use methods bearing this information in mind and interpreting and using the results accordingly.
We discuss issues connected with all of these approaches but we will look in a little more detail at data recovery and make a contribution to the question does it really help. |
| Sunbelt XXX - June 29 to July 04, 2010 - Riva del Garda Fierecongressi | | Abstract : A second look at the graph theoretic dimensions of informal organizations |
| In 1994 Krackhardt proposed four graph theoretic measures which captured the extent to which a network formed a strict hierarchical structure or in graph terms a connected out-tree. These measures were connectivity, hierarchy, efficiency and least upper boundedness. A network has all of these properties if and only if it is an out-tree. However these properties can be relaxed and we show how we can still have a similar result with less onerous conditions. In addition we look at alternative characterisations which may capture some of the properties in a more intuitive way. |
| Sunbelt XXXI - February 08 to February 13, 2011 - Trade Winds Beach Resort
http://www.tradewindsresort.com/
St. Pete Beach | | Abstract : Two-mode Projection and Data Loss |
| The standard projections take a two-mode binary matrix A and construct AAT and/or ATA and then analyze these. If our matrix A is an actor by event matrix then the former is an actor by actor matrix in which the entries are the number of events pairs of actors attended together, and the latter is an event by event matrix of the number of actors common to both events. The projections are actually similarity matrices derived from the rows and columns of the data matrix A. In the binary case these are counts of the number of times the rows (or columns) have a one in common. One of the criticisms of using projections is that there is a loss of structural information and it is true that using either AAT or ATA alone does lose structural information. However, it is not clear how much information is actually still embedded in the projections and to what extent data is actually lost. A closer examination of this issue suggests that relatively little information is lost and even less is lost if we consider both projections together. This suggests a different approach to analyzing two-mode data namely to analyze both projections and combine the solutions. |
| Workshop : Advanced Social Network Analysis using UCINET and Netdraw - PART 1 ONLY |
| This is a 1-day workshop for participants who already have some experience with network analysis, but would like to learn more. We cover advanced aspects of centrality, finding subgroups, and measuring equivalence. We also cover advanced techniques for analyzing network change and handling multiple relations, missing data, non-symmetric data, valued data and 2-mode data. Throughout, we demonstrate powerful, sometimes-undocumented features of UCINET and NETDRAW, including convenient ways of entering non-standard data. Note: what makes this workshop “advanced” is the selection of topics, not the speed or complexity of the exposition. In other words, wherever practical, all concepts are explained from first principles, making as few assumptions about prior knowledge as possible. However, we do assume basic familiarity with UCINET as a pre-requisite for the workshop. |
| Workshop : Advanced Social Network Analysis using UCINET and Netdraw - PART 2 ONLY |
| This is a 1-day workshop for participants who already have some experience with network analysis, but would like to learn more. We cover advanced aspects of centrality, finding subgroups, and measuring equivalence. We also cover advanced techniques for analyzing network change and handling multiple relations, missing data, non-symmetric data, valued data and 2-mode data. Throughout, we demonstrate powerful, sometimes-undocumented features of UCINET and NETDRAW, including convenient ways of entering non-standard data. Note: what makes this workshop “advanced” is the selection of topics, not the speed or complexity of the exposition. In other words, wherever practical, all concepts are explained from first principles, making as few assumptions about prior knowledge as possible. However, we do assume basic familiarity with UCINET as a pre-requisite for the workshop. |
| Workshop : Advanced Social Network Analysis using UCINET and Netdraw - PARTS 1 & 2 |
| This is a 1-day workshop for participants who already have some experience with network analysis, but would like to learn more. We cover advanced aspects of centrality, finding subgroups, and measuring equivalence. We also cover advanced techniques for analyzing network change and handling multiple relations, missing data, non-symmetric data, valued data and 2-mode data. Throughout, we demonstrate powerful, sometimes-undocumented features of UCINET and NETDRAW, including convenient ways of entering non-standard data. Note: what makes this workshop “advanced” is the selection of topics, not the speed or complexity of the exposition. In other words, wherever practical, all concepts are explained from first principles, making as few assumptions about prior knowledge as possible. However, we do assume basic familiarity with UCINET as a pre-requisite for the workshop. |
|
|
| |
|
