Geo-ethnography: Coupling Geographic Information Geo-Enthnography.pdf · project (henceforth referred…

Download Geo-ethnography: Coupling Geographic Information Geo-Enthnography.pdf · project (henceforth referred…

Post on 05-Jun-2018




4 download

Embed Size (px)


<ul><li><p>JOURNAL ARTICLE 05-09 </p><p>STEPHEN A. MATTHEWS JAMES E. DETWILER LINDA M. BURTON </p><p> Geo-ethnography: Coupling Geographic Information Analysis </p><p>Techniques with Ethnographic Methods in Urban Research </p><p>The article is available below and at, University of Toronto Press, 40(4): 75-90. </p><p></p></li><li><p>Geo-ethnography: Coupling GeographicInformation Analysis Techniques with</p><p>Ethnographic Methods in Urban Research</p><p>Stephen A. MatthewsDepartment of Sociology / The Pennsylvania State University /</p><p>University Park / PA / USA</p><p>James E. DetwilerDepartment of Geography and World Campus Certificate Program in GIS / The Pennsylvania State</p><p>University / University Park / PA / USA</p><p>Linda M. BurtonCenter for Human Development and Family Research in Diverse Contexts / The Pennsylvania State</p><p>University / University Park / PA / USA</p><p>Abstract</p><p>This research article focuses on the coupling of geographic information system (GIS) technologies with ethnographicdata, an approach we refer to as geo-ethnography. The data used here were gathered in an ongoing, multi-site study oflow-income families and their children. Throughout our work, the goals have been to think creatively about how GIScan be used in welfare research, to stretch the technology, and to revise the methodologies we currently use. Wespecifically discuss the ways in which the ethnographic data on families and neighbourhoods have been integrated withina GIS and how these two methods, alone and in combination, help situate families actions and experiences in timeand space and enhance data analysis and interpretation. More specifically, we focus on conceptual and methodologicalissues we have faced in the process of this integration and on practical strategies for combining qualitative andquantitative research.</p><p>Keywords: Geo-ethnography, low-income populations, welfare, family and neighbourhood research</p><p>Resume</p><p>Larticle porte sur lassociation des technologies liees aux systemes dinformation geographique (SIG) et des donneesethnographiques, une approche que les auteurs appellent geo-ethnographie. Les donnees de larticle ont ete recueilliesdans le cadre dune etude multicentrique encore en cours sur les familles a faible revenu et leurs enfants. Durant letude,les auteurs ont tente de trouver de nouvelles manieres demployer les SIG dans les recherches sur laide sociale, afin derepousser les limites de la technologie et de revoir les methodes actuelles. Plus particulierement, ils ont examine lesmoyens par lesquels les donnees ethnographiques sur les familles et leur voisinage ont ete integrees dans le SIG etcomment ces deux methodes (donnees ethnographiques et SIG), seules ou en association, aident a situer dans le tempset lespace les actions et les experiences des familles et ameliorent lanalyse et linterpretation des donnees. En outre,ils se sont concentres sur les problemes conceptuels et methodologiques decouverts dans le processus de cette integrationet sur les strategies pratiques visant a combiner les donnees qualitatives et les donnees quantitatives.</p><p>cartographica (volume 40, issue 4) 75</p></li><li><p>Mots cles: geo-ethnographie, populations a faible revenu, aide sociale, recherche sur les familles et leur voisinage</p><p>Introduction</p><p>The effect of recent welfare reform on low-income womenand children is the substantive focus of the Welfare,Children and Families Study, a complex, multi-siteproject (henceforth referred to as the Welfare Project).This article focuses on how geospatial data andgeographic information systems (GIS) have beenvariously adopted in the ethnographic component of theWelfare Project. It is important to note that GIS was notpart of the original design of the Welfare Project, andthus, throughout this article, we identify some of thechallenges and opportunities faced in promoting GISwithin multi-site ethnography and describe some of theways we have used GIS to facilitate distance-basedethnography. Our goals for the Welfare Project havebeen to think creatively about how GIS can be used tostudy the geographies of families, to stretch the GIStechnology, and to revisit and challenge importantconceptual and methodological issues regarding studiesof families and definitions of neighbourhood or context.</p><p>An advantage of the mapping and data visualizationcapabilities of a GIS is that the system can handle datacollected on multiple spatial scales. Moreover, the dataobjects need not be restricted to numerical data, as a GIScan handle hot links to a variety of audio, video, image,and text files. In this way an ethnographic researcher cancombine, using a geographic framework, different formsof data (narrative text, photographs, audio and videofiles) and data layers in previously impossible ways. Inthe context of our work we use GIS to explore and betterunderstand the lives of the low-income families beingstudied. Our work suggests that combining GIS andethnography, an approach we call geo-ethnography,has many advantages for ethnographic research on low-income families. We caution, however, that GIS and theresulting maps and/or derived new information are notenough by themselves. In the context of our study, we stillneed ethnographic data to understand the child andfamily factors, the cultural meanings of place(s), and thepolitical and sociocultural influences on the day-to-daylives of low-income families. We still need ethnography tohelp us piece together how many of the contextual datasets that can be spatially joined are relevant to low-incomefamilies and low-income communities.</p><p>We begin this article with description of the WelfareProject, with an emphasis on the main ethnographiccomponents. Next we describe the ethnographic dataproducts generated by the Welfare Project, therebypreparing the ground for a description of how GIS wasintroduced to the project and the examples, appearinglater in the paper, of how GIS is being used to facilitatemulti-site ethnographies of families and neighbourhoods.</p><p>Lastly, we reflect on practical strategies for combiningqualitative and quantitative research.</p><p>An Overview of the Welfare, Childrenand Families Study</p><p>In 1996, the Personal Responsibility and WorkOpportunity Reconciliation Act (PRWORA) changedthe welfare landscape in the United States. PRWORAand the accompanying state legislation have been describedas the greatest single shift in social policy for low-incomefamilies since the passage of the Social Security Act of 1935;in President Clintons words, they represented the endingof welfare as we know it.1 The Welfare Project is anintensive study in Boston, Chicago, and San Antoniodesigned to evaluate the consequences of welfare reform forthe well-being of children and families and to follow thesefamilies as welfare reform evolves (see Welfare, Childrenand Families 2002 for additional resources).</p><p>The Welfare Project includes a focus on children andchild development, includes a disability component,incorporates qualitative data from an extensive ongoingfamily and neighbourhood ethnography, includes bothwelfare recipients and non-recipients, is based in threedifferent geographical contexts, and is longitudinal indesign (see Winston and others 1999 for specific details).The project design has three main components: a survey,an embedded development study (EDS), and an ethnog-raphy (see Figure 1). All GIS activities in the WelfareProject are linked directly to the ethnography but not tothe survey or EDS components.</p><p>The survey component includes interviews with some 2400households with children in low-income neighbourhoodsin the three selected cities. Approximately 40% of thefamilies were receiving cash welfare payments at the timeof the first interview in 1999. Each household includeda child aged 04 or 1014 at the time of the interview;that child and his or her primary caregiver are the focalparticipants. A second and third survey were completedin 2000/2001 and 2003/2004. The embedded developmentstudy (EDS) component incorporates the videotaping andcoding of caregiverchild interactions, time-study diaries,and observations of childcare settings. The approximately700 young children in the EDS are drawn from a sub-sample of families participating in the survey.</p><p>The ethnography component, the primary focus of the GISactivities on the project, is ongoing in 25 neighbourhoodsacross the three cities; the ethnography team is tracking256 African American, Latino, and non-Hispanic whitefamilies residing in, or living in close proximity to, theseneighbourhoods. The ethnography includes three mainparts: family ethnography, disability ethnography, and</p><p>Stephen A. Matthews, James E. Detwiler, and Linda M. Burton</p><p>76 cartographica (volume 40, issue 4)</p></li><li><p>neighbourhood ethnography. The 211 participants in thefamily ethnography were recruited through formal andinformal sources such as community groups and serviceagencies as well as through introductions from otherstudy participants. Most families included at least onechild aged 24, and all were low income, with approxi-mately half receiving welfare at the time of recruitment.Forty-five families included a child under age eight witha disability (the disability ethnography). Data collectionfor the family and disability ethnographies includedparticipant observation and taped, in-depth interviewswith families over a period of 18 months (completedin 2002/2003), continuing with follow-up interviewsonce every six months through 2004. The neighbourhoodethnography, which was designed to help us betterunderstand the role of institutional resources and socialnetworks, typically involved collecting data via key-informant interviews, attending local neighbourhoodmeetings, monitoring local newspapers, and descriptiveaccounts of neighbourhoods via walkthroughs andinterviews at community-based organizations, especiallythose involved in child services and health care.</p><p>The specific aim of the ethnographic component (family,disability, and neighbourhood ethnographies) is to betterunderstand how welfare recipients experience changes inwelfare regulations and how the decisions and behavioursof low-income families are influenced by the welfare</p><p>system. The primary research questions arising from theethnography focus on the transition from welfare to workand back, making it financially, health and health care,parenting, social networks, caring for the disabled, andneighbourhood effects on all of the above.</p><p>Structured Discovery, Ethnographic Data Journeys,and Ethnographic Data Products</p><p>Within the Welfare Project, we characterize the ethnog-raphy as being one of structured discovery, an approachthat focuses on primary research topics while building insufficient flexibility to capture emergent themes andunanticipated information (Burton and others 2001).That is, many of the data on families take the form offield notes describing naturalistic, loosely structuredencounters in which field ethnographers observebehaviours and ask questions about primary researchthemes based around modules developed by seniorresearchers on the project. These modules cover discus-sions of the typical day, welfare reform experiences,household/family structure/power, residential mobilityand housing, education, family routines and rituals,family and social support networks, childcare, child andmaternal health, and child development, as well asdiscussions of neighbourhood characteristics.</p><p>Thus, the approach to fieldwork revolves around struc-tured opportunities for discovery. For example, in the</p><p>Ethnography Survey</p><p>SurveyEthnographyFamily</p><p>Ethnography</p><p>Disability</p><p>Ethnography</p><p>Neighbourhood</p><p>Ethnography</p><p>EmbeddedDevelopmental</p><p>Study(EDS)</p><p> 2,400 Families</p><p> Estimated 80% of families with current incomes below poverty; 20% between the poverty line and 200% of the poverty line</p><p>Estimated 40% on TANF</p><p>Embedded Developmental</p><p>Study</p><p> Subsample of the families in the survey</p><p> 700 young children age 2 to 4 and their care givers</p><p> Home observations and child care observations every 18 months</p><p> 215 African-American, Mexican-American, Puerto Rican, and Non-Hispanic White Families interviewed 18 to 24 months</p><p> Follow-up interviews will take place every 6 months until completion of study</p><p> TANF and Non-TANF</p><p> 45 Families with a child under 8 years old with a disability</p><p> Key informant interviews, attending neighborhood meetings, neighborhood walkthroughs</p><p>Figure 1. The design of Welfare, Children, and Families: A Three City Study.</p><p>Geo-ethnography: Coupling Geographic Information Analysis Techniques with Ethnographic Methods in Urban Research</p><p>cartographica (volume 40, issue 4) 77</p></li><li><p>context of health-related issues, ethnographic data collec-tion includes multiple encounters with families and greatlyfacilitates an exploration of multiple health events or issuesfor any family member. Such health events or issues mightinclude routine events such as health clinic visits andexperience of common ailments, issues such as domesticviolence and substance abuse, and discussions of concernsover local environmental risks such as industrial pollutantsor crime. By way of contrast, the survey component of theWelfare Project includes an array of questions andinstruments for gathering valuable data on the primarycaregiver, such as whether or not she has health insurancethrough her employer and whether or not her children arecovered, but these questions are for the most part closed-ended in format, precluding the ability to explore withrespondents additional questions on health-related topics.</p><p>ETHNOGRAPHIC DATA JOURNEYS</p><p>Ethnographic data are gathered from families andneighbourhoods in Boston, Chicago, and San Antonioand then transferred to two additional sites: the Universityof North Carolina (UNC) at Chapel Hill and thePennsylvania State University (PSU). Data on familiesthat include a child with a disability, a special feature ofthe Welfare Project, are transferred to UNC, and allethnographic data are delivered to the coordinating siteor depository at PSU. To gather, organize, analyse, andinterpret the data for this project requires a large team.The ethnographic team consists of more than 80 seniorethnographers and research scientists, family and special-ized disability ethnographers, neighbourhood ethno-graphers, coders, and programmers across five sites.All ethnographic data (tapes, transcripts, field notes,documents) from the three cities are archived at thecoordinating site, where coders read field notes ontheir assigned families and recode data in NUD*IST(Non-numerical Unstructured Data Indexing Searchingand Theorizing; see Gahan and Hannibal 1998). Eachmeeting between an ethnographer and a family (typicallythe primary caregiver) is written up, coded, and recordedin separate data files, with all text coded into categoriesor nodes. Each node is a three-letter identifier used tohighlight text, typically a paragraph within a...</p></li></ul>


View more >