Vladimir Barash

Analyzing (social media) networks with NodeXL

We present NodeXL, an extendible toolkit for network overview, discovery and exploration implemented as an add-in to the Microsoft Excel 2007 spreadsheet software. We demonstrate NodeXL data analysis and visualization features with a social media data sample drawn from an enterprise intranet social network. A sequence of NodeXL operations from data import to computation of network statistics and refinement of network visualization through sorting, filtering, and clustering functions is described. These operations reveal sociologically relevant differences in the patterns of interconnection among employee participants in the social media space. The tool and method can be broadly applied.

Chapter 10 – Twitter: Conversation, Entertainment, and Information, All in One Network!

Publisher Summary This chapter introduces the social media network Twitter and discusses how one can tap into and map parts of the global Twitter conversation. Twitter can be thought of as a conversational microblog. Like bloggers, Twitter users post messages that show up in the streams of all the people who are subscribed to them. Twitters infrastructure in this way mirrors the functionality of really simple syndication (RSS) feeds. Analyzing Twitter users and conversations is more than tabulating counts and trends—its about connections and interactions between people. It describes how social network analysis techniques can help to examine Twitter network and how the flow of the global conversation can be tracked. Collecting Twitter data with NodeXL is a simple and straightforward process. Collecting data from Twitter is generally a slow process; therefore, when collecting data from Twitter, one needs to be aware of “rate limits.” Twitter limits the amount of data collected by any one user to 150 “requests” per hour. NodeXL provides several options for speeding up data collections, such as artificially limiting the number of users in the data set. NodeXL helps to extract and analyze many different social networks from Twitter.

Critical phenomena in complex contagions

Why do some contagions “go viral” and others do not? Research on “small world” networks (Watts and Strogatz, 1998) shows how a very small number of long-range ties that bridge between clusters can allow contagions to spread almost as rapidly as on a random network of equal density. Recent research shows how long-range ties that accelerate the spread of information and disease can impede the spread of complex contagions—behaviors, beliefs and preferences that diffuse via contact with multiple adopters (Centola and Macy, 2007). In confirming this result analytically and extending the analysis from small world to power law networks, we discovered that complex contagions require a critical mass of infected nodes that corresponds to a phase transition in the ability of the contagion to take advantage of the “shortcuts” created by long-range ties. We demonstrate how this critical mass is related to the dynamics of the contagion process and identify implications for modeling behaviors that spread via social influence, such as viral marketing and social movements.

Whither the Experts? Social Affordances and the Cultivation of Experts in Community Q&A Systems

Community based Question and Answer systems have been promoted as web 2.0 solutions to the problem of finding expert knowledge. This promise depends on systems’ capacity to attract and sustain experts capable of offering high quality, factual answers. Content analysis of dedicated contributors’ messages in the Live QnA system found: (1) few contributors who focused on providing technical answers (2) a preponderance of attention paid to opinion and discussion, especially in non-technical threads. This paucity of experts raises an important general question: how do the social affordances of a site alter the ecology of roles found there? Using insights from recent research in online community, we generate a series of expectations about how social affordances are likely to alter the role ecology of online systems.

Complex contagions and the diffusion of popular Twitter hashtags in Nigeria

Social media sites such as Facebook and Twitter provide highly granular time-stamped data about the interactions and communications between people and provide us unprecedented opportunities for empirically testing theory about information flow in social networks. Using publicly available data from Twitter’s free API (Application Program Interface), we track the adoption of popular hashtags in Nigeria during 2014. These hashtags reference online marketing campaigns, major news stories, and events and issues specific to Nigeria, including reactions to the kidnapping of 276 schoolgirls in Northeastern Nigeria by the Islamic extremist group Boko Haram. We find that hashtags related to Nigerian sociopolitical issues, including the #bringbackourgirls hashtag, which was associated with protests against the Nigerian government’s response to the kidnapping, are more likely to be adopted among densely connected users with multiple network neighbors who have also adopted the hashtag, compared to mainstream news hashtags. This association between adoption threshold and local network structure is consistent with theory about the spread of complex contagions, a type of social contagion which requires social reinforcement from multiple adopting neighbors. Theory also predicts the need for a critical mass of adopters before the contagion can become viral. We illustrate this with the #bringbackourgirls hashtag by identifying the point at which the local social movement transforms into a more widespread phenomenon. We also show that these results are robust across both the follow and reply/mention/retweet networks on Twitter. Our analysis involves data mining records of hashtag adoption and of the social connections between adopters.

Respondent-Driven Sampling – Testing Assumptions: Sampling with Replacement

Abstract Classical Respondent-Driven Sampling (RDS) estimators are based on a Markov Process model in which sampling occurs with replacement. Given that respondents generally cannot be interviewed more than once, this assumption is counterfactual. We join recent work by Gile and Handcock in exploring the implications of the sampling-with-replacement assumption for bias of RDS estimators. We differ from previous studies in examining a wider range of sampling fractions and in using not only simulations but also formal proofs. One key finding is that RDS estimates are surprisingly stable even in the presence of substantial sampling fractions. Our analyses show that the sampling-with-replacement assumption is a minor contributor to bias for sampling fractions under 40%, and bias is negligible for the 20% or smaller sampling fractions typical of field applications of RDS.