Missing Data: Conceptualizing and making sense of the absence of data in a data-abundant age

Type: Paper
Theme:
Sponsor Groups: Digital Geographies Specialty Group, Geographic Information Science and Systems Specialty Group, Spatial Analysis and Modeling Specialty Group
Poster #:
Day: 4/5/2019
Start / End Time: 5:00 PM / 6:40 PM
Room: Hoover, Marriott, Mezzanine Level
Organizers: Monica Stephens, Sara Peterson
Chairs: Monica Stephens

Call for Submissions

Data is critical to geographers' work to describe and better understand complex world phenomena. For this reason, Bennett, Haining, & Griffith (1984) described missing data as "the terra incognita of geographical inquiry." In some regards, the rise of 'data driven geography' (Miller & Goodchild, 2015) has mitigated concerns over missing data. While new technologies and the abundance of spatial data collected from state efforts, sensors, surveillance mechanisms, and social media have altered our epistemology of social sciences, they have done little to illuminate the places and spaces absent from other datasets (or the map.) The terra incognita remains. Moreover, it demands critical examination and conceptualization. In a data deluge, missing data – and the areas that remain unilluminated – assume new significance

The long-established methodologies (e.g. interpolation of missing values and models for predicting missing attributes) used to overcome the quantitative concerns of working with missing or incomplete data remain relevant and widely used (ibid; Fotheringham, Brunsdon, & Charlton, 2000; Goodchild, Haining, & Wise, 1992). Yet even as such methodologies offer an invaluable means to address the technical challenges posed by missing data, there are broader considerations – e.g. the contexts and the consequences of missing data – that must be considered as well.

The term missing data emphasizes the absence of information, but the very fact that data is missing is information in itself. Understanding the context of missing data can yield insight into the ways missing data can reflect, recreate, reinforce, and reproduce existing social divisions and power dynamics. In a world where data is a prerequisite of visibility, missing data can erase groups from the conversation, propagate and perpetuate misrepresentations of people and places, and justify inequitable practices or provide unearned power.


This session seeks to broaden the way in which missing data is conceptualized. We are interested in empirical, theoretical, critical, and methodological contributions related to missing data. We are particularly concerned with the social contexts and consequences of missing data, the power dynamics that both contribute to and arise from missing data, and the real-world implications of missing data in issues of social equity and policy formation.


If you are interested in presenting in this session, please send your abstract to Sara Peterson (sarapete@buffalo.edu) and Monica Stephens (mgstephe@buffalo.edu) by Friday, October 19th, 2018.

---------------------------------

Bennett, R. J., Haining, R. P., & Griffith, D. A. (1984). The problem of missing data on spatial surfaces. Annals of the Association of American Geographers, 74(1), 138-156.

Fotheringham, A. S., Brunsdon, C., & Charlton, M. (2000). Quantitative geography: perspectives on spatial data analysis. Sage.

Goodchild, M., Haining, R., & Wise, S. (1992). Integrating GIS and spatial data analysis: problems and possibilities. International journal of geographical information systems, 6(5), 407-423.

Miller, H. J., & Goodchild, M. F. (2015). Data-driven geography. GeoJournal, 80(4), 449-461.


Description

The term missing data emphasizes the absence of information, but the very fact that data is missing is information in itself. While new technologies and the abundance of spatial data collected from state efforts, sensors, surveillance mechanisms, and social media have altered our epistemology of social sciences, they have done little to illuminate the places and spaces absent from other datasets (or the map.) The terra incognita of geographical inquiry (Bennett, Haining, & Griffith, 1984) remains essential to a variety of research questions. Established methodologies (e.g. interpolation of missing values and models for predicting missing attributes) used to overcome the quantitative concerns of working with missing or incomplete data remain relevant and widely used, but tell little about the contexts and the consequences of missing data. This session seeks to broaden the way in which missing data is conceptualized through examining empirical, theoretical, critical, and methodological contributions related to missing data.


Agenda

Type Details Minutes Start Time
Presenter Sara Peterson*, University at Buffalo, Abigail Cooke, University at Buffalo, Sara Metcalf, University at Buffalo, Interrogating the “Murder Centre of South Africa”: Estimating Homicide Rates for Small Area Geographies in Cape Town, South Africa 20 5:00 PM
Presenter Ryan Hruska, Idaho National Laboratory, Rob Edsall*, Idaho National Laboratory, Constructing Datasets for Dependency Analysis of Critical Infrastructure: Bridging Knowledge Gaps through Spatial and Semi-Automated Analysis 20 5:20 PM
Presenter Mariana Costa Lima*, Centro Universitário Christus, Clarissa Figueiredo Sampaio Freitas, Universidade Federal do Ceará, Daniel Ribeiro Cardoso, Universidade Federal do Ceará, Opaque informality and differentiated legitimacy: The Case of Fortaleza-Brazil 20 5:40 PM
Presenter Nicholas Nagle*, University of Tennessee, Parmanand Sinha, University of Louisville, David Folch, Florida State University, Seth Spielman, University of Colorado, Christopher Fowler, Pennsylvania State University, Address-level spatial structure of U.S. households 20 6:00 PM
Discussant Joshua Comenetz U.S. Census Bureau 20 6:20 PM

To access contact information login