DRWA Abstract

From InterSciWiki
(Redirected from DRW Abstract)
Jump to: navigation, search




Methods for valid statistical inferences drawn from samples of well-described ethnographic cases have become available in the last decade that solve the problems of peer effects (endogeneity) and missing data for many different kinds of inference (Eff and Dow 2009). Here the goal is to draw networks of causal inference among variables, taking peer effects into account, using the multiple regression method of Pearl (2009) for causal graphs.

The experiment. The experiment carried out in two UCI undergraduate classes in 2009-2010 was for each student to choose a dependent variables to try to explain from the SCCS (Standard Cross-Cultural Sample) database using exploratory causal analysis software by Eff and Dow (2009).

SCCS. The Standard Cross-Cultural Sample was created by (Murdock and White 1969) as the basis for a cumulative database (White 2008) for cross-cultural studies of a sample of the earliest best-described ethnographic cases in each of as many major cultural provinces in the world. Researchers at the Cross-Cultural Cumulative Coding Center (CCCCC) funded by NSF (1968-1974) coded 580 variables (approximately 1/3rd of the 2000+ codes in the database as of 2010) and other contributors coded double that in subsequent years.

SCCS Precursors - SCCS codes

Intermediate results.

Expected final results.


Three theories new to cognitive anthropology are proposed and either tested or supported with empirical evidence. First, predictive cohesion theory suggests the cohesion-consensus hypothesis of cultural sharing. Structural cohesion is a formal network measure that identifies the group boundaries for which the redundancy or multiconnectivity of ties is greatest and a group is least likely to be separated. It is thus an identifier of groups within which culture is most likely to be shared. Second, ecological psychology suggests that the way perceptions are stored episodically in memory lends itself to network coding of links representing people and interactions: multiple dyadic bonds at a temporal scale of experiential events and contexts. Kinship networks, for example, can be drawn to represent a more macro time scale of dyads and events such as marriage, childbirth, death, migration, and proximal interactions. At this scale there are culturally recognized and individually perceived event boundaries and time-scales of event sequences but other contexts that have different times-scales and event sequences. Third, research on animal and primate perceptions suggest that the information processing capacities of humans are exceptional in the perception of event and interaction structures such as might be coded for different aspects of social network interactions and contexts. Elements of network structure may thus be perceived or experienced (as with membership in a structrually cohesive group) and have causal efficacy without necessarily being named entities. Taken together, these theoretical viewpoints provide new approaches to study of the relation between social networks, cognition and culture. Approaches and findings are explored in a series of methodological tutorials, examples and case studies in which the hypotheses and theoretical frameworks are supported.




Analyzing Social Media Networks with NodeXL From: "Roumeliotis, Rachel (ELS-BUR)" <R.Roumeliotis@elsevier.com> Date: Wed, August 11, 2010 7:58 am To: "Roumeliotis, Rachel (ELS-BUR)" <R.Roumeliotis@elsevier.com> Options: View Full Header | View Printable Version | Download this as a file | Add to Address Book | View Message Details | View as HTML

Dear Colleagues,

We're pleased to announce that our new book "Analyzing Social Media Networks with NodeXL" will be available September 2010 from Elsevier/Morgan Kaufmann Publishers:

See the book's web page here! <http://www.elsevierdirect.com/product.jsp?isbn=9780123822291&dmnum=CWS1 >

We know it is late, but we'd like you to consider including this book as part of your fall semester teaching plans. We've had satisfying success in getting information, computer science, communications, and business students to do substantive network analysis and create revealing visualizations with 2-5 week segments of existing courses.

The NodeXL tool, a plug-in for Excel 2007/2010, is free to download at www.codeplex.com/nodexl. Users can download datasets used as examples in the book, or use NodeXL to get network data from email, twitter, facebook, flickr, youtube, and other sources. The book describes the power of social media, gives a solid introduction to network analysis, and then leads readers through the steps of using NodeXL.

If you want to get an e-copy of the book, please send an email to our editor, Rachel Roumeliotis (R.Roumeliotis@elsevier.com). Let us know if you have any questions, and please pass this on to appropriate colleagues.

     Derek Hansen  shakmatt@gmail.com 
     Ben Shneiderman ben@cs.umd.edu
     Marc Smith  marc@connectedaction.net 

Book Description: Businesses, entrepreneurs, individuals, and government agencies alike are looking to social network analysis (SNA) tools for insight into trends, connections, and fluctuations in social media. Microsoft's NodeXL is a free, open-source SNA plug-in for use with Excel. It provides instant graphical representation of relationships of complex networked data. But it goes further than other SNA tools -- NodeXL was developed by a multidisciplinary team of experts that bring together information studies, computer science, sociology, human-computer interaction, and over 20 years of visual analytic theory and information visualization into a simple tool anyone can use. This makes NodeXL of interest not only to end-users but also to researchers and students studying visual and network analytics and their application in the real world.

In Analyzing Social Media Networks with NodeXL, members of the NodeXL development team up provide readers with a thorough and practical guide for using the tool while also explaining the development behind each feature. Blending the theoretical with the practical, this book applies specific SNA instructions directly to NodeXL, but the theory behind the implementation can be applied to any SNA.

  • Walks readers through using NodeXL while explaining the theory and

development behind each step, providing takeaways that can apply any SNA

  • Demonstrates how visual analytics research can be applied to SNA tools

for the mass market

  • Presents readers with case studies using NodeXL on popular networks

like email, Facebook, Twitter, and Wikis

See the book's web page here! <http://www.elsevierdirect.com/product.jsp?isbn=9780123822291&dmnum=CWS1 >

Rachel Roumeliotis

Senior Acquisitions Editor


Morgan Kaufmann, an Imprint of Elsevier Science

TEL: 781-313-4774

FAX: 781-221-1615