Qualitative GIS

A Mixed Methods Approach

Meghan Cope, Sarah Elwood, Meghan Cope, Sarah Elwood

• Publisher: SAGE Publications Ltd
• Year: 2009
• Print ISBN: 9781412945660, 9781412945653
• eBook ISBN: 9781446244562
• Edition: 1
• Topic: Ciencias físicas
• Subtopic: Geografía

Geographic Information Systems are an essential tool for analyzing and representing quantitative spatial data. Qualitative GIS explains the recent integration of qualitative research with Geographical Information Systems

With a detailed contextualising introduction, the text is organised in three sections:

Representation: examines how researchers are using GIS to create new types of representations; working with spatial data, maps, and othervisualizations to incorporate multiple meanings and to provide texture and context.

Analysis: discusses the new techniques of analysis that are emerging at the margins between qualitative research and GIS, this in the wider context of a critical review of mixed-methods in geographical research

Theory: questions how knowledge is produced, showing how ideas of ?science? and ?truth? inform research, and demonstrates how qualitative GIS can be used to interrogate discussions of power, community, and social action

Making reference to representation, analysis, and theory throughout, the text shows how to frame questions, collect data, analyze results, and represent findings in a truly integrated way. An important addition to the mixed methods literature, Qualitative GIS will be the standard reference for upper-level students and researchers using qualitative methods and Geographic Information Systems.

1

INTRODUCTION: QUALITATIVE GIS: FORGING MIXED METHODS THROUGH REPRESENTATIONS, ANALYTICAL INNOVATIONS, AND CONCEPTUAL ENGAGEMENTS

Sarah Elwood and Meghan Cope

The title of this volume may, for some readers, suggest contradiction, incongruity, or juxtaposition. From its inception, this project has been met with some measure of all three, and with several persistent questions. Why qualitative GIS? Why a mixed methods approach? Why not just ‘mixed methods GIS’? What is ‘qualitative’ in the context of a digital technology? In spite of such questions, efforts to integrate qualitative data and techniques with GIS have been building in recent years (Kwan and Knigge, 2006). Multimedia GIS approaches embed sketches, mental maps, audio, video, or photographs into GIS, often to represent non-cartographic forms of spatial knowledge, such as emotion (Al Kodmany, 2002; Kwan, 2007; Shiffer, 2002; Weiner and Harris, 2003). A growing number of researchers use GIS-based spatial analysis in concert with methodologies more familiar to qualitative researchers, such as focus groups, ethnography, interviewing, or participatory action, thus strengthening research findings by bringing together these different ways of knowing (Cieri, 2003; Dennis, 2006; Pain et al., 2006; Weiner and Harris, 2003). Others are developing ways to use GIS as part of a suite of analysis techniques drawn from qualitative research, whether by adapting GIS software or by using it to carry out inductive interpretive visualization (Knigge and Cope, 2006; Kwan and Lee, 2004; Matthews et al., 2005). Building upon these and other examples, this collection is intended to frame the emerging field of qualitative GIS, profiling the range of ways in which researchers and practitioners are integrating GIS with qualitative research.

Qualitative GIS is one of several approaches to geographic information systems that emerged in response to critiques in the mid 1990s that cast GIS as rooted in positivist epistemologies and most suited for quantitative techniques associated with spatial science (Lake, 1993; Pickles, 1995). These critiques also raised concerns about the difficulty of incorporating non-cartographic spatial knowledge into conventional GIS, and the ensuing potential for exclusion and disempowerment (Harris and Weiner, 1998; Sheppard, 1995). Responding to these critiques, many researchers have taken on GIS in new ways, working to incorporate multiple data and forms of knowledge, extend its representational capabilities to incorporate non-cartographic information, support quantitative and qualitative forms of analysis, and illustrate that multiple epistemologies may be part of GIS-based research. These approaches, including public participation and participatory GIS, feminist GIS, and critical GIS, share an understanding of GIS as more than only quantitative in the forms of data that may be included, the analyses that may be supported, and the representational practices that may be fostered. These propositions have paved the way for qualitative GIS.¹

At the level of the technology, it is increasingly simple in most desktop GIS software to store multimedia forms of spatial knowledge (photographs, sketch maps, narrative descriptions) in a GIS database or to hyperlink to them in other locations. But qualitative GIS is more than this. It stands apart from conventional GIS practices, as well as some of the approaches described above, because of the way it engages and conceives of GIS. Specifically, qualitative GIS assumes that geographic phenomena, their relationships, and their meanings are produced and negotiated at many different moments in GIS development and application: in spatial data, in data structures, in spatial analysis techniques, in the meanings fostered or foreclosed in GIS-based maps and applications. As this collection will illustrate, this assumption enables approaches to GIS that foster new collisions with qualitative analysis techniques, qualitative forms of data, and new conceptualizations of how meaning is negotiated in and through different aspects of GIS, including its software, data structures, and visualizations (cartographic and otherwise). Further, qualitative GIS is predicated upon multiple representations and modes of analysis, hybrid epistemologies, and researchers’ critically reflexive efforts to draw on these multiplicities in ways that enable more robust explanation. In the remainder of this introductory chapter, we further detail this framing of qualitative GIS, outline some of its foundations and practices, and situate its origins in and contributions to geographers’ long-standing reliance on mixed methods.

WHAT IS GIS AND WHAT CONSTITUTES ‘QUALITATIVE’?

Identifying the emerging field of qualitative GIS, characterizing its differences from other approaches to GIS, and explaining its intellectual and political significance require some initial discussion of what is meant by ‘GIS’ and what is meant by ‘qualitative’. GIS research and practice over the past decade constitutes GIS in multiple ways, and geographers’ conceptions of qualitative research are similarly multi-faceted. This complexity produces many ways of bringing GIS and qualitative research together, and generates a diversity of sites and practices through which qualitative GIS might emerge.

For more than a decade, GIS research and critiques have conceptualized GIS in several ways (Pickles, 1995). Geographic information systems are, in one understanding, digital technologies for storing, managing, analyzing, and representing geographic information. Typically, such a system consists of data models; structures for representing geographic entities and their characteristics in digital form; data structures for storing these data; the data themselves (together with the ontologies, categorization schemes, and other elements that are part of these representations); software for query, retrieval, analysis, and mapping; and the hardware used to support these functions (Chrisman, 2002).² But simultaneously, GIS is understood as a collection of practices for producing and negotiating geographic knowledge through the representation and analysis of spatial data. These practices are constituted by GIS software producers and the predominantly private sector industry that creates GIS software; geography and other academic disciplines that create and validate certain ways of encoding spatial data in a GIS, or using GIS in research; and the ever-diversifying ‘community’ of GIS users who create various GIS practices (Pickles, 1995). From this perspective, GIS is constituted through its representational, analytical, and epistemological approaches, all of which are understood to be shaped by the social, political, and disciplinary norms and institutional practices from which they emerge. This conceptualization of GIS owes much to the efforts of researchers to respond to claims about an ‘inherent’ positivism in GIS and explain how the social and political impacts of GIS might be produced (Pavlovskaya, 2006; Sheppard, 1995), as well as to feminist geographers’ critical reflections upon the social, political, and institutional construction of knowledge in research (Lawson, 1995; Mattingly and Falconer Al-Hindi, 1995; Moss, 1995). Our notion of qualitative GIS is rooted in this hybrid understanding of GIS as technology, methodology, and situated social practice.

What, then, counts as ‘qualitative’ in our account of qualitative GIS? First, data or forms of evidence in research may be qualitative. Qualitative data are not simply those data that are non-numerical. Rather, we argue that data may be qualitative in part by virtue of the rich contextual detail they provide about social and material situations. Ethnographic interviews, for instance, tend to elicit responses from interviewees that describe conditions, relationships, and processes in detail. If we were to ask neighborhood residents how a local community development initiative has changed their neighborhood, some might describe changes to the built infrastructure, some might relate stories of how local residents’ ability to impact neighborhood planning has changed, and others might describe how new residents have moved in, altering social relations in the neighborhood. The interviewees’ responses are qualitative data because each narrative likely communicates rich descriptive detail about these shifting social and material conditions and processes.

But it is not only the presence of rich contextual descriptive detail that constitutes data as qualitative. Rather, data may also be qualitative if they contain or provide interpretations of the situations or processes that they describe. For example, in the ethnographic interview responses described above, the words chosen and the stories related by the interviewees will almost certainly evidence their own interpretations or the meanings they draw from particular events or conditions. One person may describe a newly constructed grocery store in the neighborhood as an example of positive changes that improve food security for low-income households in the community, while another may describe the same site and characterize it as a harbinger of gentrification – a potential threat to these same low-income households.

These forms of evidence are qualitative because they not only encapsulate a description of material change, but also offer interpretations of the meanings or impacts of that change.³ As well, these data are qualitative because we can use them to understand situated and negotiated knowledge. That is, the different meanings developed in the two interviewees’ characterizations of the grocery store are surely also being contested and negotiated among neighborhood residents, and so these two pieces of evidence provide a researcher insights into the social and political situations of the neighborhood. Further, the researcher might examine the identities, experiences, and interests of the two interviewees to understand how these factors affect the meanings they provide – understanding how their respective explanations provide differently situated knowledges.

Thus, it is not only data that may be qualitative, but also analysis. That is, we also understand as qualitative those forms of analysis that are intended to draw out the situated interpretive detail of qualitative data. Analytical techniques such as grounded theory, discourse analysis, or content analysis, for example, work with qualitative forms of evidence to tease out their negotiated meanings and situated knowledges. Techniques such as coding (systematically categorizing data to identify themes and patterns) and the triangulation of multiple data sources are associated with qualitative research because they enable researchers to examine the contradictions, commonalities, and nuances of data that are rich in contextual and process-based detail. Practices such as the iterative or recursive examination of multiple forms of evidence in conversation with one another are also characteristic of qualitative analysis. And finally, as has been well developed in feminist researchers’ writing on qualitative methods, a hallmark of qualitative analysis is its critical reflexivity upon the knowledge production process, specifically how research designs, forms of data, analysis techniques, disciplinary politics around epistemologies, and research relationships (such as the position of the researcher vis-à-vis the participants) tend to produce particular forms of knowledge, conclusions, politics, and power relations (Mattingly and Falconer Al-Hindi, 1995; Moss, 2003; Nast, 1994).

These understandings of GIS and qualitative forms of data and analysis foreshadow the multiplicity of engagements that might constitute qualitative GIS. We understand as qualitative GIS those approaches that seek to integrate qualitative forms of data into GIS, develop and support qualitative approaches to building knowledge and explanation with GIS, use GIS in research that emerges from multiple or hybrid epistemologies, and theorize previously unrecognized forms of social knowledge that may be present in GIS applications. These approaches are quite different in the way they bring together GIS and qualitative research, but all go beyond treating qualitative methods as ‘add-ons’ to essentially quantitative projects rendered in a GIS. Instead, they offer substantive shifts toward framing questions, collecting data, analyzing results, and representing findings in a truly integrated way. They intersect GIS and qualitative research with the goal of integrating multiple forms of evidence or ways of knowing, in order to explain how spatial knowledge, patterns, relationships, and interactions are produced, and with what sorts of social and political impacts.

POSITIONING QUALITATIVE GIS AS A MIXED METHODS APPROACH

The core commitment of qualitative GIS to integrating multiple forms of knowledge and the findings from various techniques is also at the heart of mixed methods research. This integrative way of building robust explanations in research is what positions qualitative GIS as a mixed methods approach. Distinguished from ‘multiple methods’ projects in which different methods are practiced in parallel, mixed methods projects weave together diverse research techniques to fill gaps, add context, envision multiple truths, play different sources of data off each other, and provide a sense of both the general and the particular. In these approaches, insights gained from one technique, subject group, or data source may be examined recursively with other findings, and the path of research may be shifted in response (Cresswell, 2003; Jiang, 2003; Robbins, 2003). Different techniques may produce complementary explanations for phenomena, while other times (and equally valuably) they may produce contradictory explanations, leaving the task of understanding how and why these multiple versions of ‘truth’ intersect (England, 1993; Nightingale, 2003; Pavlovskaya, 2002; Rocheleau, 1995; Tashakkori and Teddlie, 2003). Mixed methods have been especially important in geography research, given the strong presence of research questions that require investigating interrelated human and physical processes, understanding cognitive and social processes, or examining interscalar relationships and processes. For example, accounting for the role of structures such as political economy or gender, while also understanding how and why they may play out differently in various contexts, is precisely the sort of integrative project that would necessitate relying on multiple forms of evidence and diverse epistemologies (ways of knowing).

Mixed methods approaches are rooted in several assumptions about knowledge and epistemologies in research that we suggest are also critical for qualitative GIS. First, mixed methods research tends to treat knowledge as always partial (no one can know the ‘whole truth’) and situated (knowledge depends on our situations and positions), whether it is the forms of evidence that researchers ‘gather’ or the knowledge that they produce in their analysis, interpretation, and representation of these data. From this perspective, differently situated knowledges and multiple ways of examining evidence can inform more robust understandings of complex processes or phenomena. Second, mixed methods research is premised on the notion that epistemology and methodology are related, but that this relationship is neither fixed nor singular. A realist, positivist, or constructivist epistemology need not prescribe a given methodological orientation, or only one approach (qualitative or quantitative). Indeed, some scholars contend that mixed methods research forwards unique hybrid epistemologies. Inherent in efforts to bring together multiple ways of knowing, they contend, is an assumption that multiple epistemologies may be valid ways of fostering understanding and explanation for particular purposes and in specific circumstances (Elwood, 2009a; Knigge and Cope, 2006; Maxcy, 2003). Finally, a great deal of mixed methods research rests on the assumption that the knowledge making we do in research is inherently political. The manner in which researchers interpret tensions or contradictions among data or methods and weave together different approaches has social and political consequences, especially because different forms of data, representation, and analysis are frequently afforded different levels of intellectual and political authority (Elwood, 2006; McCann, 2008; McLafferty, 2002). All of these propositions underscore the importance of researchers’ critical agency in bringing together multiple epistemologies, modes of analysis, and forms of knowledge.

Emerging efforts to intersect GIS and qualitative research share many of these same commitments. The chapters in this collection illustrate the persistent methodological and epistemological multiplicity of qualitative GIS, and the authors’ efforts to work with GIS in ways that foster debate, multiple readings, and iterative interpretation, with sustained critical reflection upon methods and outcomes. Qualitative GIS involves conceiving of GIS as social and political practice, as discipline- and industryinscribed ways of making knowledge, and as an assemblage of hardware, software, data structures, and procedures for working with digital spatial data. This conceptualization of GIS allows us to consider GIS-based knowledge production as something that is occurring explicitly and implicitly in several sites: spatial data, data structure, spatial analysis, cartographic representation, and the application or use of any of these phenomena in social and political practice. This multi-layered reading of how knowledge is produced in GIS suggests that the ‘qualitativeness’ of qualitative GIS may be advanced in many ways: by integrating qualitative and quantitative representations of spatial k...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Contents
• Notes on Contributors
• List of Figures and Tables
• Acknowledgements
• 1 Introduction: Qualitative GIS: Forging mixed methods through representations, analytical innovations, and conceptual engagements
• 2 Non-quantitative GIS
• Representations
• Analytical Interventions And Innovations
• Conceptual Engagements
• Index