In Cooking Data Crystal Biruk offers an ethnographic account of research into the demographics of HIV and AIDS in Malawi to rethink the production of quantitative health data. While research practices are often understood within a clean/dirty binary, Biruk shows that data are never clean; rather, they are always “cooked” during their production and inevitably entangled with the lives of those who produce them. Examining how the relationships among fieldworkers, supervisors, respondents, and foreign demographers shape data, Biruk examines the ways in which units of information—such as survey questions and numbers written onto questionnaires by fieldworkers—acquire value as statistics that go on to shape national AIDS policy. Her approach illustrates how on-the-ground dynamics and research cultures mediate the production of global health statistics in ways that impact local economies and formulations of power and expertise.
|Publisher:||Duke University Press Books|
|Series:||Critical Global Health: Evidence, Efficacy, Ethnography Series|
|Product dimensions:||6.00(w) x 9.00(h) x (d)|
About the Author
Crystal Biruk is Assistant Professor of Anthropology at Oberlin College.
Read an Excerpt
THE OFFICE IN THE FIELD
Building Survey Infrastructures
It is market day at Mangochi turnoff in southern Malawi, and the trading center is bustling with activity. Buyers and sellers of kaunjika (secondhand clothes), sneakers, vegetables, printed fabrics, and batteries bargain over prices and socialize, creating a low buzz of voices against a backdrop of persistently blaring minibus horns. On a sunny June morning in 2008, I walk a short distance away from the busy trading center. Passing an open-air butcher shop where young men sit beneath a tall tree hung with two goat carcasses, I arrive at a large compound. Surrounded by walls hand painted with bright advertisements for Boom washing powder and Panadol pain relievers, a squat rest house sits back from the open gates: a favored stop for truck drivers, the rustic motel is called Mpaweni, or Other People's Place.
There is no vacancy at Mpaweni. Its rooms have been taken over by the fieldwork teams — American researchers and graduate students and Malawian fieldworkers, data entry clerks, and drivers — of the Longitudinal Study of AIDSin Malawi (LSAM), a cohort study that has collected demographic data in villages nearby since 1998. For the next two months, fieldworkers will survey and HIV test about a thousand Malawians. From a vantage point in the dirt courtyard, a visitor might not notice that one of the motel's conference rooms has been converted into a makeshift field office. Data entry teams tap at the keyboards of LSAM-owned laptops, manually transferring data codes from the dusty pages of completed surveys administered the day before to a growing database. Boxes of Lifebuoy body soap and Sunlight laundry soap are piled neatly around the periphery of the room, gifts that will compensate research participants for answering the questions that make up this year's twenty-five-page survey. A photocopier and printer whir quietly, printing off endless copies of questionnaires, consent forms, and log forms that will soon be filled in with data and information. Electrical cords snake underfoot, ending in overworked power strips that protect the electronic devices in the room from the periodic power surges and outages so common in Malawi. Parked helter-skelter around the compound are minibuses that carry fieldworkers to the project's sample villages, all within an hour's drive of Mpaweni: one by one, fieldworkers will visit the households where the members of the study sample live.
Mpaweni is the temporary headquarters for LSAM for the duration of data collection fieldwork. In the words of local residents who notice the visitors around town, "Akafukufuku abweranso! [The researchers have come again!]"
* * *
The scene at Mpaweni hints at the massive human and material infrastructure that must be built in order for large-scale survey research to be carried out in a corner of Malawi far from LSAM's home office in the Population Studies Center at an elite research university in the United States. Reams of paper, laptops, and extension cords must be carried to the field from abroad or from Lilongwe; minibuses must be rented to ferry field teams to and from rural households; fieldworkers must be hired; housing must be found for researchers and fieldwork supervisors for the duration of data collection; and green bricks — in 1,000-kwacha increments rubber-banded together — must be withdrawn periodically from cash points to pay the salaries and per diems of fieldworkers employed by the project. Trips to the airport to pick up arriving researchers or imported items, such as weight scales to collect anthropomorphic data and HIV test kits to collect samples from respondents, were a weekly occurrence. Sometimes items such as the test kits would get tied up in customs bureaucracy, necessitating complex efforts to free them. Building the temporary infrastructure of people and things necessary to carry out peripatetic survey research in one of the poorest countries in the world is a Herculean task.
This chapter shows how planning and designing field survey projects entails imaginative work on the part of researchers who aim to translate standards — conjured in the office — into clean, high-quality data produced in the messy space of the field. Adopting the position of an anthropologist among the demographers, as discussed in the introduction, I first elaborate how the human infrastructure for survey research, made up of foreign and Malawian experts who bring different expertise to the table, is built in difficult conditions. I draw attention to the disparate material and academic investments of foreign and Malawian researchers in data collection, often obscured by the discourse of partnership or collaboration central to development, humanitarian, and global health worlds today (Mercer 2003; Crane 2010b; Watkins and Swidler 2012; Kenworthy 2014; Thoreson 2014; Brown 2015; Gerrets 2015b). In the second half of the chapter, I articulate the epistemological dreams and standards that call into being the infrastructure for data collection in the field. In analyzing debates between Malawian and foreign collaborators around cultural and linguistic translation and the fine-tuning of survey concepts, instruments, and questions, around plans for where surveys should be administered, and around what should be the objectives of research, the chapter excavates the multiple interests and forms of expertise that coalesce in the pages of a survey, even before it is administered to the first household in the field.
The survey questionnaire is the tool at the core of data production and operates as a framing device that aspires to make Malawi visible and intelligible as data or numbers that circulate among demographers or policy makers: "The world appears to the observer as a relationship between picture and reality, the one present but secondary, a mere representation, the other only represented, but prior, more original, more real" (Mitchell 1991, 60). The survey — as the key mechanism of ordering, counting, and framing the division between real and represented — plays a central role in effecting what demographers experience as a good-enough representation of the really real: data. As I show, the questionnaire itself and data practices in the field reflect a fundamental distrust of data on the part of the most diligent demographers, who recursively ask themselves and others, Are the data good? Is this the right question to answer our research problem? Are people lying? Are fieldworkers cooking the data? These questions arise in the pre-fieldwork meetings discussed in this chapter, but, more importantly, they are a quality of data themselves. If one assumes one can collect data that transcend these questions and the uncertainty they signal, posing such questions indicates that uncertainty is inherent to data themselves. This chapter's central interest is in how questions, standards, and tools that eventually produce quantitative data that are devised in the office are translated into and for the field. I theorize translation as an ongoing and improvised practice that privileges the epistemic investments of those who design the survey, one that betrays their shared imagination of a cultural Other who will answer their questions, leaves the culture of demography itself unmarked, and prefigures the nature of data to be collected.
From the Office to the Field: The Spatial Politics of Data Production
I sometimes get depressed when I come to Malawi. I'm used to sitting in my office crunching numbers and having the categories be anonymous, not personified. ... But I'm pretty wedded to coming to check up on things. ... If you don't come now and then you have no idea what is going on in the field if you don't hover over people's shoulders there. — Dr. Jones, economist and MAYP coprincipal investigator, September 20, 2007
The way you enter the village the first time, will remain in the minds of the people and will also determine the success or failure of your objectives. ... As a fieldworker [you] should know that the [villagers'] culture has been there for ages. ... To effectively work with the community you also have to be participative in the community, attend funerals, attend village meetings to show you are not just there to work, but you are one of them. However ... attending political rallies [is not advised]. ... You might only be a part of one group thereby losing the other. ... Refrain from any political gatherings or debates to be part of the whole community. — LSAM Fieldwork Manual 2008
Reading these two texts alongside each other — the first an excerpt from an interview with a Marriage and Youth Project (MAYP) researcher and the second an excerpt from a fieldwork manual designed by Malawian supervisors working with LSAM for many years — I am struck by their shared construction of a place called "the field." Dr. Jones sets up a clear contrast between being in her office "crunching numbers" and being in the field. Implicit in this contrast is an assumption that the office is a clean space for data analysis and tinkering with numbers while the field is a messy place where numbers become people. In the office, it is easy to "forget that the numbers once represented people with real communities and real histories and complex genealogies" (Jain 2013, 36), but this becomes more difficult when researchers like Dr. Jones confront poverty and suffering firsthand on a visit to Malawi. Jones acknowledges, however, the importance of visiting the field now and then to check up on the activities transpiring there, hinting at their potential influence on the data that wind up in the office. Although when we met she had been in Malawi for only a few days, she asserted her difference from other economists who never set foot "on the ground" (in the field). Her insinuation that things might go awry in the field if one doesn't "hover over people's shoulders" connotes epistemological and structural hierarchies that characterize survey projects: she looks over the shoulders of potentially unreliable Malawian fieldworkers on the front lines of data collection, implicitly acknowledging their ability to mess up or dirty the data to be ferried to the office.
Finally, the distinction she draws between anonymous numbers and categories and personified realities indexes the interest of this chapter in how abstract standards and ideals for clean data translate into the field and hints at how subjective practices in the latter might erode the objective status often granted to statistical data. Across a large body of published work on guidelines, methods, and survey design across cultural contexts, the construction of the field as a place of "difficult geographic topography" rife with "weather and seasonal impediments" and "danger[s]" that threaten to "bungle" a survey is consistent (Pennell, Levenstein, and Lee 2010; see also Bulmer and Warwick 1983). "The field" compels the translation work needed to link standard survey methodology and procedures to "environments of stringent budgetary constraints in countries with widely varying levels of survey infrastructure and technical capacity" (Yansaneh 2005, 5). To manage impediments to smooth and timely data collection in remote or rustic locales, survey projects sometimes selected the sites for their data collection based on their proximity to the office. For example, MAYP's research proposal notes that researchers selected Salima District to administer its surveys because working in only one district (as opposed to several) would allow the field staff to monitor data quality. As an added benefit, Salima is close to the national capital, thus reducing project transport and infrastructure costs. Even before the first survey is administered, then, behind-the-scenes decisions determine and delimit the nature and quality of data to be collected, in this case, via convenient bounding of the sample area.
The Malawian supervisors who authored the fieldwork manual (cited above) meant to provide guidelines to fieldworkers implementing LSAM's 2008 survey similarly construct the field as a place of difference, distance, and complexity. They cast it as foreign to the fieldworkers who will enter it for the first time and attempt to prepare the teams for the culture they will find there, presumably more pronounced, dense, traditional, visible, and different than their own culture, which, of course, is not recognized as such. Fieldworkers are advised to walk a fine line between being participative and maintaining proper distance from the villagers they will interact with in the field. They are encouraged, for example, to attend funerals and community meetings for the duration of data collection, but discouraged from getting involved in local politics, which might serve to alienate some research subjects and make them less willing to answer survey questions. For Dr. Jones and fieldworkers alike, the field is a place whose uncertainties and stumbling blocks must be imagined prior to fieldwork such that their influence on data quality can be minimized. The talk and practices of researchers and fieldworkers make the field intelligible by inventing it, facilitating their ability to imagine themselves and the data collection tools they employ as translators between the field and the office (Wagner 1981).
Holding steady a vision of the field as container of data facilitates the collaborative effort to assemble high-quality data. Whereas chapter 2 explores this imaginative labor and its entailments specifically from the perspective of fieldworkers, this chapter focuses on how the field undergirds and directs the efforts of researchers in the office to design survey questions, tools, and research plans that self-consciously aim to manage the messiness and unpredictability of the field. Before data are collected, this chapter shows, rural Malawi and its residents must be recast as "the field" and "research participants," respectively, enabling researchers to translate their epistemological dreams into a contained — and manageable — space of difference.
Demographers leading survey projects in Malawi were very clear about the simultaneously marginal and core role that the field played in their research efforts. On the one hand, they agreed that survey researchers "rarely, if ever, step foot in the field" and don't see the fieldwork component of research as important to their work. Dr. Payson, MAYP demographer, suggested that her disciplinary kin tend to "parachute in and out of countries," echoing critiques by anthropologists and others that "the demographer could study a society without ... knowing much of anything about it. ... Visits to the country, if required at all, could be confined to short stays in western luxury hotels" (Kertzer and Fricke 1997, 11). Payson suggested that for those who work on survey projects in Africa, doing fieldwork is actually detrimental to furthering one's career in academia: disciplinary norms — and, by proxy, tenure expectations — see a researcher being too heavily involved in the field side of things as a waste of time that could be instead directed toward writing new research proposals, publishing results, or analyzing data. She was frustrated that her investment in qualitative methods and longer-term fieldwork as accompaniments to collecting numbers was squashed by disciplinary norms and structures. Dr. Canton, a Canadian social demographer leading projects in Burkina Faso, Kenya, and South Africa, echoed Payson's claim that the disciplinary norms of demography disallow long-term fieldwork: "Fieldwork is seen as a vacation; its point is not understood at all."
In such disciplinary renderings, the field becomes a distant and exotic site that is hierarchically situated far beneath the space of calculation, intellect, and analysis that is the office. This spatialization likewise grafts on to the actors who are expected to populate each of these spaces: fieldworkers and villagers in the field and expert demographers crunching numbers in the office. Indeed, the space between these two sites is crucial to producing the kinds of knowledge expected by the epistemic community of demographers: dispassionate, objective, and universally circulating numbers. Dr. Matenje, a Malawian demographer based in South Africa, emphasized the ways in which number crunching simultaneously made him aware of harsh on-the-ground realities and made him feel helpless: "As a demographer, when I started analyzing the DHS data, I realized what was killing people was AIDS. ... I understand how important that data is, but it just incapacitated me. I couldn't do anything about [the people dying]." Matenje, like Dr. Jones, suggests that crunching numbers in the office — the everyday labor of the demographer — necessitates an emotional distance between himself and distant realities, one that nonetheless compels him to consider the moral implications of his work. Numbers, as portable placeholders for people themselves and stand-ins for human suffering, operate to make realities appear as taken-for-granted givens to be measured or enumerated rather than structurally produced inequalities and suffering to be meaningfully ameliorated.
Excerpted from "Cooking Data"
Copyright © 2018 Duke University Press.
Excerpted by permission of Duke University Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
Acknowledgments ix Introduction. An Anthropologist among the Demographers: Assembling Data in Survey Research Worlds 1 1. The Office in the Field: Building Survey Infrastructures 31 2. Living Project to Project: Brokering Local Knowledge in the Field 67 3. Clean Data, Messy Gifts: Soap-for-Information Transactions in the Field 100 4. Materializing Clean Data in the Field 129 5. When Numbers Travel: The Politics of Making Evidence-Based Policy 166 Conclusion. Anthropology in and of (Critical) Global Health 200 Appendix. Sample Household Roster Questions 217 Notes 223 Bibliography 237 Index 269
What People are Saying About This
"This is not a simple revelation story in which we learn that data in research projects is socially contingent. It is a cultural study of demography research in the field, and the end product is the best we can do in anthropology—familiar things are made unfamiliar, conditional, and fragile. Crystal Biruk's work is quite simply fantastic."
"This book is going to find a wide audience throughout and beyond global health and anthropology—Crystal Biruk's attention to language and metaphor makes Cooking Data eminently teachable. This is superior scholarship that is very well grounded in everyday life and the peculiar world of research. I learned a great deal."