Susan OShea The Mitchell Centre for Social Network

Susan O’Shea The Mitchell Centre for Social Network Analysis CCSR/Social Statistics, University of Manchester susan. o’shea@manchester. ac. uk

5 th ESRC Research Methods Festival, 2 -5 July 2012 What is Network Analysis? Network analysis is the study of social relations among a set of actors. Social relations can be thought of as dyadic attributes. Types of Network Analysis Ego network analysis: Respondents are asked about the people they interact with and the relationships between them. Data can be collected via observation, archives (including facebook) and surveys. Used in conjunction with random sampling, which enables classical statistical techniques to be used to test hypotheses. Complete network analysis: tries to obtain all the relationships among a set of respondents, e. g. Friendships of co-workers Difficulties with missing data and autocorrelation. Statistical modelling can help with these issues. Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 2

5 th ESRC Research Methods Festival, 2 -5 July 2012 Connections Homophily: Birds of a feather flock together. The extent to which actors form ties with similar versus dissimilar others. Similarity can be defined by gender, race, age, occupation, educational achievement, status, values or any other salient characteristic. Mutuality/Reciprocity: Is the relationship between two actors reciprocated, for example friendship. Possibility of cognitive dissonance. Common concept in the analysis of social capital. Network Closure: Friends of friends are my friends too. This is known as transitivity. Network closure is a measure of the occurrence of triads within the network. Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 3

5 th ESRC Research Methods Festival, 2 -5 July 2012 Distributions Density: Direct ties in a network relative to the total number possible Centrality: a set of measures examining the influence or importance of a particular node within the network. Distance: The smallest number of ties required to connect two actors The idea of ‘six degrees of separation’ or small world effect Bridge: An actor providing the only link between two individuals or clusters Structural holes: The absence of ties between two parts of a network. Useful concept for social capital; has business applications Tie Strength: Strong ties associated with homophily; weak ties with bridges Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 4

5 th ESRC Research Methods Festival, 2 -5 July 2012 Visualizing Networks Exploratory analysis Provides qualitative information Allows us represent complex information visually However: Must be effective Should be valid Can be misleading Open to interpretation Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 5

5 th ESRC Research Methods Festival, 2 -5 July 2012 Nodes: Represent actors Edges: Represent relationships (Ties) Attribute based We use the attributes of the nodes to position them e. g. gender Scaling methods We use known or derived distances, e. g. Cities, countries Graph theoretic We use properties of the network (density, closure etc. ) Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 6

5 th ESRC Research Methods Festival, 2 -5 July 2012 Blocks and cut-points Affiliation network 129 Riot Grrrl bands 534 actors Courtney Love link between dense sub-cliques. Patty Schemel link between groups. Played with Hole and is currently involved with Rock Camp for Girls. Corin Tucker, Kathleen Hanna, Allison Wolf and Tobi Vail bridging the network. Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 7

5 th ESRC Research Methods Festival, 2 -5 July 2012 Who is important in a network & does structure matter? Centrality is a measure of how a node’s importance is determined by network structure and position https: //sites. google. com/site/ucinetsoftware/home With UCINET Under Network | Centrality Degree Closeness Betweeness Eigenvector With Net. Draw Under Analyze | Centrality Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 8

5 th ESRC Research Methods Festival, 2 -5 July 2012 What are Ego Networks? An ego network is a network focused on a single actor (e. g. a person), referred to as the ‘ego’ Information on his/her ‘alters’ is gathered (e. g. Friends) Usually alter to alter relationships are collected Where possible attribute data on ego and alters is gathered Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 9

5 th ESRC Research Methods Festival, 2 -5 July 2012 Ego networks Left - Courtney Love & Patty Schemel; Right – Corin, Kathleen, Allison & Tobi This reflects a division between Riot Grrrl Activism and DIY music and more ‘professional’ commercial music represented by Patty and Courtney. There also more men in the ego networks of these two actors. Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 10

5 th ESRC Research Methods Festival, 2 -5 July 2012 Cohesive subgroups Areas of the network in which actors are more closely related to each other than actors outside the group It is a concept not a formal definition Based on relations rather than attributes e. g. Collaboration networks, close friends Clique is a cohesive subgroup with at least 3 members and Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 11

Friends T 1 Close friends T 1 Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 12

Friends T 3 Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 13

Ladyfest Ten network all nominations at 3 time points including shared activities and events after the festival Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 14

5 th ESRC Research Methods Festival, 2 -5 July 2012 Analysing your own Facebook Ego Network 1. Go to http: //apps. facebook. com/namegenweb 2. Click on UCInet - log into your facebook account 3. The app will download a file, this takes a few minutes 4. Open up notepad, this is in All Programs>Accessories (you can use word but you must save the file as a plain text file) 5. Copy and paste the downloaded file (which should be open) into notepad. Save the file as something like myfacebook. 6. Start up UCINET and click Data>import text file>dl and enter your file, that is myfacebook 7. Your data should now be in UCINET and you can view it in Netdraw. Susan O'Shea, University of Manchester. susan. o'shea@manchester. ac. uk 30/11/2020 15