Ate Poorthuis

Social Media and the City: Rethinking Urban Socio-Spatial Inequality Using User-Generated Geographic Information

Big data is increasingly seen as a way of providing a more ‘scientific’ approach to the understanding and management of cities. But most geographic analyses of geotagged social media data have failed to mobilize a sufficiently complex understanding of socio-spatial relations. By combining the conceptual approach of relational socio-spatial theory with the methods of critical GIScience, this paper explores the spatial imaginaries and processes of segregation and mobility at play in the notion of the ‘9th Street Divide’ in Louisville, Kentucky. Through a more context-sensitive analysis of this data, this paper argues against this popular spatial imaginary and the notion that the Louisville’s West End is somehow separate and apart rban planning from the rest of the city. By analyzing the everyday activity spaces of different groups of Louisvillians through geotagged Twitter data, we instead argue for an understanding of these neighborhoods as fluid, porous and actively produced, rather than as rigid, static or fixed. Ultimately, this paper is meant to provide a conceptual and methodological framework for the analysis of social media data that is more attentive to the multiplicity of socio-spatial relations embodied in such data.

The New Political Economy of Geographical Intelligence

A troubling new political economy of geographical intelligence has emerged in the United States over the last two decades. The contours of this new political economy are difficult to identify due to official policies keeping much relevant information secret. The U.S. intelligence community increasingly relies on private corporations, working as contractors, to undertake intelligence work, including geographical intelligence (formally known as GEOINT). In this article we first describe the geography intelligence “contracting nexus” consisting of tens of thousands of companies (including those in the geographical information systems and mapping sector), universities and nonprofits receiving Department of Defense and intelligence agency funding. Second, we discuss the “knowledge nexus” to conceptualize how geographical knowledge figures in current U.S. intelligence efforts, themselves part of the U.S. war on terror and counterinsurgency (COIN). To analyze the contracting nexus we compiled and examined extensive data on military and intelligence contracts, especially those contracts awarded by the countrys premier geographical intelligence agency, the National Geospatial-Intelligence Agency (NGA), for satellite data. To analyze the knowledge nexus we examined recent changes in the type of geographical knowledges enrolled in and produced by the U.S. intelligence community. We note a shift from an emphasis on areal and cultural expertise to a focus on calculative predictive spatial analysis in geographical intelligence. Due to a lack of public oversight and accountability, the new political economy of geographical intelligence is not easy to research, yet there are reasons to be troubled by it and the violent surveillant state it supports.

Pacifying Babel’s Tower: A scientometric analysis of polycentricity in urban research

It is sometimes claimed that the degree of polycentricity of an urban region influences that region’s competitiveness. However, because of widespread use and policy relevance, the underlying concept of polycentricity has become a ‘stretched concept’ in urban studies. As a result, academic debate on the topic leads to situations reminiscent of Babel’s Tower. This meta-study of the scientific literature in urban studies traces the conceptual stretching of polycentricity using scientometric methods and content analysis. All published studies that either apply the concept directly or cite a work that does, were collected from the Scopus bibliographic database. This resulted in a citation network with over 9000 works and more than 20,000 citations between them. Network analysis and clustering algorithms were used to define the most influential papers in different citation clusters within the network. Subsequently, we employed content analysis to systematically assess the mechanisms associated with the formation of polycentric urban systems in each of these papers. Based on this meta-analysis, we argue that the common categorisation of polycentricity research in intra-urban, inter-urban and inter-regional polycentricity is somewhat misleading. More apt categorisations to understand the origins of polycentricity’s conceptual ambiguity relate to different methodological traditions and geographical contexts in which the research is conducted. Nonetheless, we observe a firm relation across clusters between assessments of polycentricity and different kinds of agglomeration economies. We conclude by proposing a re-conceptualisation of polycentricity based on explicitly acknowledging the variable spatial impact of these different kinds of agglomeration economies.

Follow thy neighbor: Connecting the social and the spatial networks on Twitter

This paper compares the social properties of Twitter users’ networks with the spatial proximity of the networks. Using a comprehensive analysis of network density and network transitivity we found that the density of networks and the spatial clustering depends on the size of the network; smaller networks are more socially clustered and extend a smaller physical distance and larger networks are physically more dispersed with less social clustering. Additionally, Twitter networks are more effective at transmitting information at the local level. For example, local triadic connections are more than twice as likely to be transitive than those extending more than 500 km. This implies that not only is distance important to the communities developed in online social networks, but scale is extremely pertinent to the nature of these networks. Even as technologies such as Twitter enable a larger volume of interaction between spaces, these interactions do not invent completely new social and spatial patterns, but instead replicate existing arrangements.

Offline Brews and Online Views: Exploring the Geography of Beer Tweets

This chapter analyzes the distribution of geocoded social media data (also referred to as a cyberscape) that references “beer” and related terms. Drawing upon an ongoing research project that archives every geocoded tweet in the world, this chapter explores differences in the frequency and geographic distribution of the everyday commentary made by Twitter users about beer. While the sheer volume of activity, close to a million geocoded beer tweets in 2012, is notable in its own right, it is only when comparisons between subsets of the data are made that the most intriguing spatial patterns emerge. In order to showcase these patterns of differences within online social media, this chapter compares beer tweets to twitter commentary on other topics, i.e., contrasting the geography of wine and beer tweets as well as examining differences within the online conversations about beer, i.e., how do references to light beers or regional “cheap” beers vary over space. These geographical differences (e.g., where are the hot spots for “beer” vs. “wine” or “Bud Light” versus “Coors Light”) illuminates how the commentary and views expressed online, reflect offline practices and preferences. In short, the visualization of “beer space” produced by mapping tweets represents the complex intertwining of offline preferences for specific brews which are expressed via an online practice of presenting ones views.

Christaller and “big data”: recalibrating central place theory via the geoweb

ABSTRACT This article utilizes central place theory (CPT) to navigate the “deluge” brought about by big data. While originating in the 1930s, CPT is a theoretical monument of 1960s spatial science. CPT aims to understand settlement geographies based on consumption behavior and is often presented as a singular, outdated, and rationalist theory. After critically reviewing the history of CPT, we assess the microfoundations of Christaller’s CPT – the threshold and range of goods – for various central functions in Louisville, Kentucky. The microfoundations are estimated through data from social media platforms Foursquare and Twitter. These sources alleviate many of the operationalization issues that traditionally hamper empirical use of CPT. The empirical application of CPT reveals that: (i) central functions have typical ranges and thresholds relating central places to population spread; (ii) central functions cluster based on an approximate hierarchical structure. The findings indicate the ongoing importance of CPT in shaping urban-economic geographies.

Making Big Data Small: Strategies to Expand Urban and Geographical Research Using Social Media

ABSTRACT While exciting, Big Data (particularly geotagged social media data) has proven difficult for many urbanists and social science researchers to use. As a partial solution, we propose a strategy that enables the fast extracting of only relevant data from large sets of geosocial data. While contrary to many Big Data approaches—in which analysis is done on the entire dataset—much productive social science work can use smaller datasets—around the same size as census or survey data—within standard methodological frameworks. The approach we outline in this paper—including the example of a fully operating system—offers a solution for urban researchers interested in these types of data but reluctant to personally build data science skills.

The canary in the city: indicator groups as predictors of local rent increases

As cities grow, certain neighborhoods experience a particularly high demand for housing, resulting in escalating rents. Despite far-reaching socioeconomic consequences, it remains difficult to predict when and where urban neighborhoods will face such changes. To tackle this challenge, we adapt the concept of ‘bioindicators’, borrowed from ecology, to the urban context. The objective is to use an ‘indicator group’ of people to assess the quality of a complex environment and its changes over time. Specifically, we analyze 92 million geolocated Twitter records across five US cities, allowing us to derive socio-economic user profiles based on individual movement patterns. As a proof-of-concept, we define users with a ‘high-income-profile’ as an indicator group and show that their visitation patterns are a suitable indicator for expected future rent increases in different neighborhoods. The concept of indicator groups highlights the potential of closely monitoring only a specific subset of the population, rather than the population as a whole. If the indicator group is defined appropriately for the phenomenon of interest, this approach can yield early predictions while simultaneously reducing the amount of data that needs to be collected and analyzed.

Capturing Unobserved Correlated Effects in Diffusion in Large Virtual Networks

Social networks and social capital are generally considered to be important variables in explaining the diffusion of behavior. However, it is contested whether the actual social connections, cultural discourse, or individual preferences determine this diffusion. Using discrete choice analysis applied to longitudinal Twitter data, we are able to distinguish between social network influence on one hand and cultural discourse and individual preferences on the other hand. In addition, we present a method using freely available software to estimate the size of the error due to unobserved correlated effects. We show that even in a seemingly saturated model, the log likelihood can increase dramatically by accounting for unobserved correlated effects. Furthermore the estimated coefficients in an uncorrected model can be significantly biased beyond standard error margins.