Methodenveranstaltungen der Brückenprofessur

Vorträge im Wintersemester 2014/15.

27.11.2014Methoden-Workshop

Einführung in die Software f4/f5 und f4 Analyse

Thorsten Dresing (Philipps-Universität Marburg)
Veranstaltungsort:
IW3
Am Biologischen Garten 2
28359 Bremen
Uhrzeit:
16:15 Uhr
Ansprechpartner/in:
Veranstaltungsreihe:
Methodenveranstaltungen der Brückenprofessur
Semester:
WiSe 2014/15

Abstract:
Thorsten Dresing stellt die von ihm mitentwickelte Software f4/f5 sowie die neue Analysesoftware f4 Analyse vor. f4/f5 gehört zum Standardrepertoire der qualitativen Sozialforschung, das Programm erleichtert die Transkription von Audio- und Videodaten (z.B. Interviews, Gruppendiskussionen, Filme) erheblich. f4 Analyse unterstützt die Auswertung qualitativer Daten kategorienbasierter Auswertungen (wie z.B. Inhaltsanalyse, Grounded Theory). In der Einführungen wird der Einsatz der Software exemplarisch vorgestellt. Dabei werden Sie mit einzelnen Arbeitsschritten vertraut gemacht. Die Einführung ist ein ausgezeichneter Einstieg für Anfänger/innen und Neugierige.

Weitere Informationen: audiotranskription.de

Veranstaltungsort:
COGNIUM - Rotunde
Enrique-Schmidt-Straße 5
28359 Bremen
Uhrzeit:
18:15 Uhr
Ansprechpartner/in:
Veranstaltungsreihe:
Methodenveranstaltungen der Brückenprofessur
Semester:
WiSe 2014/15

Peter V. Marsden, Edith and Benjamin Geisinger Professor of Sociology, Harvard University. Marsden's research interests are centered on social organization, especially formal organizations and social networks. He has studied survey measurement of social networks and research methods for establishment surveys. Marsden is involved in the ongoing data collection efforts of the General Social Survey and has been a lead investigator of three National Organization Studies conducted between 1991 and 2003. Since 2011 he serves as Dean of Social Science at Harvard University.

Abstract:
Based on a comparison of the 2004 and 1985 General Social Surveys (GSS), a pro-minently-reported finding suggested that a dramatic decline in the availability of confiding relationships to adult Americans had taken place. This presentation reports on two studies suggesting that such a conclusion may not be warranted. One focuses on trends in the frequency of socializing with others between 1974 and 2008; it finds no general trend toward more or less social activity, though some over-time differences in types of socializing are evident-in particular, toward less socializing with neighbors. The second study suggests that the 1985-2004 difference in confiding may be due to differences in questionnaire placement of the "name generator" items on which the finding rests.

Veranstaltungsort:
SOCIUM Forschungszentrum Ungleichheit und Sozialpolitik
Raum: 5.4680
Mary-Somerville-Straße 7
28359 Bremen
Uhrzeit:
9:00 - 15:30 Uhr
Ansprechpartner/in:
Veranstaltungsreihe:
Methodenveranstaltungen der Brückenprofessur
Semester:
WiSe 2014/15

Program:
9 a.m.
Welcome

9.15 a.m. 
Christof Wolf (Mannheim U): Challenges of egocentric network instruments

10.05 a.m. 
Valentina Hlebec (Ljubljana U): Methodological considerations in the collection of ego-centered network data

10.55 a.m. 
Coffee break

11.10 a.m. 
Beate Völker (Utrecht U): Measuring Social Networks, Social Capital and Social Cohesion

12.00 p.m. 
Lunch Break

1.15 p.m. 
Peter Marsden (Harvard U): Social perception and social networks

2.05 p.m. 
coffee & tea pick up

2.10 p.m. 
Betina Hollstein, Michael Windzio & Uwe Engel (Bremen U): A longitudinal study to develop and test items for ego-centered network support

26.03.2015Bridge-Lecture

The settings model. A stochastic actor-oriented model for dynamics of large networks

Prof. Tom Snijders (Rijksuniversiteit Groningen)
Veranstaltungsort:
Gästehaus der Universität Bremen
Auf dem Teerhof 58
28199 Bremen
Uhrzeit:
13:15 Uhr
Ansprechpartner/in:
Kooperation:
Veranstaltungsreihe:
Methodenveranstaltungen der Brückenprofessur
Semester:
WiSe 2014/15

About person: Tom A.B. Snijders is Professor of Statistics and Methodology at the Dept. of Sociology, University of Groningen, Emeritus Fellow, Nuffield College, University of Oxford and an Associate Member at the Dept. of Statistics, University of Oxford.

26.03.2015Methoden-Workshop

Analysis of longitudinal social network data with R Siena

Prof. Tom Snijders (Rijksuniversiteit Groningen)
Veranstaltungsort:
SOCIUM Forschungszentrum Ungleichheit und Sozialpolitik
Raum: 5.4680
Mary-Somerville-Straße 5
28359 Bremen
Uhrzeit:
10:00 - 17:30 Uhr
Ansprechpartner/in:
Veranstaltungsreihe:
Methodenveranstaltungen der Brückenprofessur
Semester:
WiSe 2014/15

About person: Tom A.B. Snijders is Professor of Statistics and Methodology at the Dept. of Sociology, University of Groningen, Emeritus Fellow, Nuffield College, University of Oxford and an Associate Member at the Dept. of Statistics, University of Oxford.

Abstract:
This workshop is about statistical inference for longitudinal observations on social networks. Longitudinal social network data are understood here as two or more repeated observations of a directed graph on a given node set (usually between 30 and a few hundred nodes). The workshop teaches the statistical method to analyze such data, for which a tutorial is given in Snijders, T.A.B., Steglich, C.E.G., and van de Bunt, G.G. (2010), Introduction to actor-based models for network dynamics (Social Networks), and implemented in the RSiena program.

The statistical model is the actor-oriented model where the nodes are actors whose choices determine the network evolution. This allows to include various network effects (reciprocity, transitivity, cycles, popularity, etc.), effects of individual covariates (covariates connected to the sender, the receiver, or the similarity between sender and receiver), and of dyadic covariates.

An important extension is to have, in addition to the network, one or more actor variables that evolve in mutual dependence with the network; an example is a friendship network of adolescents where drinking behavior is a relevant actor variable which influences, and is influenced by, the friendship network. This leads to models for the simultaneous dynamics ('co-evolution') of networks and behavior, which are a special option in RSiena. Further information about this method can be found at the SIENA website (see below).

The statistical analysis is based on many repeated Monte Carlo simulations of the network evolution model and therefore is a bit time-consuming. The computer program RSiena is a package in the statistical computer system R. The workshop will demonstrate the basics of using RSiena. Attention will be paid to the underlying statistical methodology, to examples, and to the use of the software.

The first session is intended for those without previous experience with this method, and will focus on the intuitive understanding of the model and operation of the software.

The second session will present models for the simultaneous dynamics of networks and behavior and other more advanced topics such as model specification, multivariate networks, structurally determined values, and goodness of fit checking.

Participants are requested to check the SIENA website (Courses-activities tab) in the week before the workshop to download the workshop materials. For optimal benefit, it is advisable to bring an own laptop with R and RSiena already installed, such that some steps of data manipulation and analysis can be followed hands-on. Participants for whom R is new are requested to learn the basics of R before the workshop: how to run R and how to give basic R commands. This is to reduce the amount of new material to digest at the workshop itself. The Siena website (RSiena tab) has some links which can be helpful for this purpose: it's not hard!