Some results from Scottish data The Statistical Analysis of the Dynamics of Networks and Behaviour: An Application to Smoking and Drinking Behaviour among.

Slides:



Advertisements
Similar presentations
Framing Processes in Social Dilemmas. Formal Modelling and Experimental Validation. Christian Steglich, ICS Groningen presentation prepared for the Mathematical.
Advertisements

Statistical Social Network Analysis - Stochastic Actor Oriented Models Johan Koskinen The Social Statistics Discipline Area, School of Social Sciences.
Analysing network-behavioural co-evolution with SIENA Christian SteglichUniversity of Groningen Tom SnijdersUniversity of Groningen Mike PearsonNapier.
Assessing influence and selection in network-behavioural co-evolution with an application to smoking and alcohol consumption among adolescents. Christian.
Where we are Node level metrics Group level metrics Visualization
The Statistical Analysis of the Dynamics of Networks and Behaviour. An Introduction to the Actor-based Approach. Christian Steglich and Tom Snijders ——————
Introduction and Aim Group identification describes our sense of belonging to the group and of commonality with other ingroup members. Research has shown.
SOCI 5013: Advanced Social Research: Network Analysis Spring 2004.
Some results from Scottish data Topic smoking behaviour and friendship Problem influence and/or selection Theory drifting smoke rings (Pearson, West, Michell)
Outline input analysis input analyzer of ARENA parameter estimation
Section 1.3 Experimental Design © 2012 Pearson Education, Inc. All rights reserved. 1 of 61.
Advanced model specification for SIENA models: score tests and forward model selection Mark HuismanUniversity of Groningen Christian SteglichUniversity.
An RG theory of cultural evolution Gábor Fáth Hungarian Academy of Sciences Budapest, Hungary in collaboration with Miklos Sarvary - INSEAD, Fontainebleau,
Analysis of frequency counts with Chi square
Directional triadic closure and edge deletion mechanism induce asymmetry in directed edge properties.
Link creation and profile alignment in the aNobii social network Luca Maria Aiello et al. Social Computing Feb 2014 Hyewon Lim.
Analysing the co-evolution of social networks and “behavioural” dimensions with SIENA Christian SteglichUniversity of Groningen Tom SnijdersUniversity.
NEIGHBORHOOD EFFECTS BY Steven N. Durlauf Ania Bonarska & Okafor Luke Emeka Development Workshop, 2007.
FRIENDSHIP AND DELINQUENCY OF ADOLESCENTS FRIENDSHIP AND DELINQUENCY OF ADOLESCENTS First results of a four wave study on adolescent’s behavior and relations.
Centrality and Prestige HCC Spring 2005 Wednesday, April 13, 2005 Aliseya Wright.
Co-evolution of Members’ Attachment to the Team and Team Interpersonal Networks Chunke Su Noshir Contractor University of Illinois at Urbana-Champaign.
Social Contexts: Peers Peer group: A collection of individuals approximately equal in age, social status, ability, and other characteristics.
Joint social selection and social influence models for networks: The interplay of ties and attributes. Garry Robins Michael Johnston University of Melbourne,
Friends and social networks. Strong and Weak Ties Strong ties: Intense attachment Often reinforced by other ties Weak ties: Acquaintances May link otherwise.
Analysing the co-evolution of social networks and “behavioural” dimensions with SIENA Christian SteglichUniversity of Groningen Tom SnijdersUniversity.
References References (continued) ANOMALIES Risk-taking Network Density Paradox Apparent contradictions in research findings Network density is an.
Program-stimulated change in network composition and behavior related to family planning in Ghana Marc Boulay Dynamics of Networks and Behavior Symposium.
The Structure of Consensus: Cohesion and Hierarchy in Peer Networks G. Robin Gauthier Duke University Partial support for this project thanks to NSF/HSD:
The Neymann-Pearson Lemma Suppose that the data x 1, …, x n has joint density function f(x 1, …, x n ;  ) where  is either  1 or  2. Let g(x 1, …,
Exploring the dynamics of social networks Aleksandar Tomašević University of Novi Sad, Faculty of Philosophy, Department of Sociology
Composition of environmental decision-making networks – a case study in Helsinki, Finland M.Soc.Sc. Arho Toikka HERC Seminar Series Department.
Experimental Design 1 Section 1.3. Section 1.3 Objectives 2 Discuss how to design a statistical study Discuss data collection techniques Discuss how to.
Objectives: TOBACCO NETWORKS IN A VOCATIONAL EDUCATION SCHOOL IN THAILAND. Introduction : Methods: Conclusions: To study a social network of smokers and.
Friends (Temporarily) Forever: Frequency of Facebook Use, Relationship Satisfaction, and Perception of Friendship Zack Hayes, Jerad Hill, Heather Jacobson,
Copyright © 2015, 2012, and 2009 Pearson Education, Inc. 1 Chapter Introduction to Statistics 1.
Various topics Petter Mostad Overview Epidemiology Study types / data types Econometrics Time series data More about sampling –Estimation.
Mixture Models, Monte Carlo, Bayesian Updating and Dynamic Models Mike West Computing Science and Statistics, Vol. 24, pp , 1993.
Coevolution in mutualistic communities Plants Animals Network structure for a plant-frugivore community in southeastern Spain. Bascompte and Jordano, 2007.
V13: Causality Aims: (1) understand the causal relationships between the variables of a network (2) interpret a Bayesian network as a causal model whose.
Social Network Analysis Prof. Dr. Daning Hu Department of Informatics University of Zurich Mar 5th, 2013.
Susan O’Shea The Mitchell Centre for Social Network Analysis CCSR/Social Statistics, University of Manchester
Vulnerability to addiction and the role of the media Can I: describe risk factors in the development of addiction including stress, peers, age and personality?
Presentation of the Groningen group plan Marijtje van Duijn ECRP Meeting Ljubljana, Feb. 2, 2012.
A two minute introduction to: Exponential random graph (p*)models for social networks SNAC Workshop, Illinois, November 2005 Garry Robins, University of.
1 Components of the Deterministic Portion of the Utility “Deterministic -- Observable -- Systematic” portion of the utility!  Mathematical function of.
MODELING MATTER AT NANOSCALES 6.The theory of molecular orbitals for the description of nanosystems (part II) The density matrix.
KPS 2007 (April 19, 2007) On spectral density of scale-free networks Doochul Kim (Department of Physics and Astronomy, Seoul National University) Collaborators:
SOCIAL LEARNING THEORY As an explanation for Substance Misuse.
Introduction to Statistical Models for longitudinal network data Stochastic actor-based models Kayo Fujimoto, Ph.D.
Peer Relationships.
Transitions in Conjoint Alcohol and Tobacco Use among Adolescents Kristina M. Jackson University of Missouri, Columbia & Missouri Alcoholism Research Center.
Peer-Pressure & Risk-Taking Behaviour. The Peer Group “Peer” – anyone who has one or more characteristics or roles in common with one or more other individuals.
Wander Jager 1 An updated conceptual framework for integrated modelling of human decision making: The Consumat II.
Introduction to ERGM/p* model Kayo Fujimoto, Ph.D. Based on presentation slides by Nosh Contractor and Mengxiao Zhu.
Section 1.3 Objectives Discuss how to design a statistical study Discuss data collection techniques Discuss how to design an experiment Discuss sampling.
Appropriate use of Design Effects and Sample Weights in Complex Health Survey Data: A Review of Articles Published using Data from Add Health, MTF, and.
Modelling Complex Systems Video 4: A simple example in a complex way.
Why do people take risks? The examples of smoking and bad driving
An Introduction to Stochastic Actor-Oriented Models (aka SIENA)
Social Networks Analysis
Chapter 4: The Nature of Regression Analysis
Adoption of Health Information Exchanges and Physicians’ Referral Patterns: Are they Mutually Reinforcing? SAEEDE EFTEKHARI*, School of Management, State.
Social Balance & Transitivity
Adjustment of Temperature Trends In Landstations After Homogenization ATTILAH Uriah Heat Unavoidably Remaining Inaccuracies After Homogenization Heedfully.
Ennett, Bauman and Koch 1994.
Modeling Peer Influence
An Introduction to Latent Class Analysis (LCA)
Social Network Analysis
Chapter 4: The Nature of Regression Analysis
Longitudinal Social Network Data
Presentation transcript:

some results from Scottish data The Statistical Analysis of the Dynamics of Networks and Behaviour: An Application to Smoking and Drinking Behaviour among School Friends. Christian Steglich Tom Snijders ICS / Department of Sociology University of Groningen Mike Pearson Centre for Mathematics and Statistics Napier University, Edinburgh

some results from Scottish data Topic smoking behaviour and friendship Problem influence and/or selection Theory drifting smoke rings (Pearson, West, Michell) Data three wave panel ’95’96’97, school year group, age Method actor-driven modelling

some results from Scottish data Literature S. Ennett & K. Bauman (1993). Peer Group Structure and Adolescent Cigarette Smoking: A Social Network Analysis. Journal of Health and Social Behavior 34(3): E. Oetting and J. Donnermeyer (1998). Primary Socialization Theory: the Etiology of Drug Use and Deviance. Substance Use and Misuse 33(4): M. Pearson & L. Michell (2000). Smoke Rings: Social Network Analysis of Friendship Groups, Smoking, and Drug-Taking. Drugs: Education, Prevention and Policy 7(1): M. Pearson & P. West (2003). Drifting Smoke Rings: Social Network Analysis and Markov Processes in a Longitudinal Study of Friendship Groups and Risk-Taking. Connections 25(2):59-76.

some results from Scottish data Problem Empirical “network autocorrelation”: Friends of smokers are smokers, friends of non-smokers are non-smokers. Why that? Various theoretical accounts influence selection

some results from Scottish data Problem refined influence selection What is the role of cohesion ? Influence is expected to be strongest in cohesive subsets of the network. Selection mechanisms can generate such cohesive subsets. selection influence cohesion autocorrelation

some results from Scottish data Modelling Actor-driven, dynamic model: actors are assumed to take two types of decisions: network decisions (whom to call a friend) behavioural decisions (own smoking). The interplay of both generates the evolution process of network and behaviour. What is modelled are structural and other determinants of the actors’ preferences.

some results from Scottish data Modelling It is assumed that the network and behaviour evolves in continuous time between the observation moments. Network & behaviour evolve in mini steps, in which one of the actors is permitted (but not required)…  to make a change in one friendship tie: network mini step, or  to make a change in his/her behaviour: behaviour mini step.

some results from Scottish data Modelling When actor i is allowed to make a network mini step, (s)he can change one tie variable, maximizing an objective function + random disturbance: The objective function expresses the actor’s preferences as a function of network position and own & others’ behaviour. i = ego, j = alter, x = network, z = behaviour, t = time,  = parameter,  = random influence. (Behavioural mini steps are modelled analogously.)

some results from Scottish data Modelling The network objective function includes: network structure, own behaviour, others’ behaviour, and interactions. The behavioural objective function includes: network structure, own behaviour, others’ behaviour, and interactions. Interdependence between network and behaviour is accounted for !! 

some results from Scottish data Modelling Model specification: Spell out the two objective functions as weighted sums of network and behaviour effects. Weights  are parameters estimated from data. Here (smoking of adolescents): model actors’ preferences…  for cohesion,  for adapting to their friends’ behaviour,  for choosing friends that behave the same,  etc., …in both types of decisions / objective functions.

some results from Scottish data Modelling In SIENA, include measures of cohesion as well as measures of selection and influence, plus interaction terms. cohesion reciprocitytransitivity # reciprocal pairs # peripheral to dense triads # transitive triplets # actors at distance 2 # dense triads –+++++ – + local density

some results from Scottish data Influence and selection are based on a measure of behavioural similarity : Friendship similarity of actor i : Actor i has two ways of increasing friendship similarity: by adapting own behaviour to that of friends j, or by choosing friends j who behave the same. Modelling

some results from Scottish data Stepwise increase of model complexity Start with simple cohesion measures… reciprocity effect measures the preference difference of actor i between right and left configuration transitivity effect i i i i j j j j kk

some results from Scottish data Stepwise increase of model complexity … and with simple measures of influence and selection. friendship similarity effect “classical” selection “classical” influence

Results SIENA parameter estimates: basis model some results from Scottish data

Results SIENA parameter estimates: basis model some results from Scottish data

Results SIENA parameter estimates: basis model some results from Scottish data

some results from Scottish data Stepwise increase of model complexity Add simple interaction. reciprocity × similarity effect selection × reciprocity influence × reciprocity

Results SIENA estimates extended models: similarity × reciprocity in network model (all other parameters barely change) some results from Scottish data

Results SIENA estimates extended models: similarity × reciprocity in behavioural model: Standard errors of all behavioural parameters become high – no meaningful estimates ! some results from Scottish data

Results: frequency of decision types SIENA parameter estimates: basis model some results from Scottish data

some results from Scottish data Stepwise increase of model complexity Add cohesion measures based on group positions (approximated as specific configurations of the neighbourhood). group member belongs to “dense triad” peripheral is unilaterally attached to group isolate has no incoming ties

some results from Scottish data Stepwise increase of model complexity For example: peripheral × similarity effect selection × peripheral influence × peripheral

some results from Scottish data Results SIENA parameter estimates: a complex model network part of the model (1):

some results from Scottish data Results SIENA parameter estimates: a complex model network part of the model (2): (other network effects remain as were before)

some results from Scottish data Results SIENA parameter estimates: a complex model behavioural part of the model: (again, standard errors are quite high)

some results from Scottish data Results Selection effects are strong. Cohesion effects also. Interaction with cohesion reduces selection effect: the more cohesive a group, the less important similarity to these friends. Influence effects are weak or even spurious: controlling for cohesion, there is no influence effect. Q: Is smoking no ‘social thing’, while other activities like drinking are ?  run a parallel analysis of drinking behaviour !

Second analysis – drinking SIENA parameter estimates: basis model some results from Scottish data

Second analysis – drinking SIENA parameter estimates: basis model some results from Scottish data

Second analysis – drinking SIENA parameter estimates: basis model some results from Scottish data much higher t-score than in smoking analysis A: Drinking indeed seems to be more of a ‘social thing’, than smoking (influence parameter significant).  follow up on this, increase model complexity…

some results from Scottish data Summary simultaneous statistical modelling of network & behavioural dynamics for longitudinal panel data allows for disentangling selection and influence effects special positional effects can be investigated software SIENA 2.0 is available from (beta version, final version comes soon)