Research materials, analysis and ethics

The collection of research material was conducted by three researchers including myself, each of whom had her own subproject.⁴⁹ My project was based on ethnographic work at

49 The research project had, in fact, two parent subprojects from different disciplines (social sciences and psychology) that were further divided into subprojects. The multidisciplinary research project (2006–08) on child healthcare clinics was calledEnnaltaehkäisevä terveystyöneuvolassa (responsible leaders: Docent J.

Ruusuvuori and Professor R-L. Punamäki, University of Tampere). Near the end of the fieldwork period, a fourth researcher joined the social science project.

four different maternity and child healthcare clinics⁵⁰ for approximately three months, and included multiple methods of data production such as video recording, observations of nursing work activities, interviews and documentary material from various sites of maternity healthcare practice, as is common in multi-sited ethnographic inquiry (see Hammersley & Atkinson 1995; Marcus 1995; Harbers & Mol & Stollmayer 2002;

Huttunen 2010). After the fieldwork was carried out, we arranged workshops in each clinic for the nurses who had participated in the study in order to communicate preliminary findings in lay terms and to offer them a space to reflect on their working methods.

The empirical fieldwork was mainly conducted by me over a three month period in the course of 2006–2008.⁵¹ A relatively long period of fieldwork was conducted in order to explore the dynamics of a variety of care practices that could not be grasped through short visits to the field (e.g. Hammersley & Atkinson 1995; Harbers & Mol & Stollmayer 2002).

The fieldwork was conducted in short periods of four to five days, so that I spent about one working week a month in each clinic. This was done so that the fieldwork would not be too exhausting to any of the participants. After approximately three months had been spent in the clinics I decided to finish the fieldwork, because it seemed that I had covered in my observations all the different activities, settings and temporal cycles related to doing pregnancy and maternity healthcare (see Diamond 2006, 58; Hammersley & Atkinson 1995, 23–54). Further, through video recordings, I had collected a huge amount of data-intensive material in a relatively short period of time. Also, it would seem that ethnography done in one’s own culture requires shorter fieldwork in that the researcher is familiar with the basic social conduct and language of the field (Eriksen 1995, 19–20).⁵²

50 Only additional video recording from a fourth clinic was conducted to obtain welfare assessment form interview appointments.

51The research material will be stored at the Department Social Research for 25 years, after which it will be destroyed. Negotiations about storing the data in some form at the Finnish Social Science Data Archive have been initiated to ensure further use.

52 Smith gives little attention to particular methods of collecting material, so I have had to explore other literature resources here. Interviewing is the only exception in Smith’s coverage. InInstitutional

Ethnography. A Sociology for People, for example, she makes reference to participatory observation on five separate pages. On one page it is referred to as a method in an illustrative research setting (30), on another it is mentioned in reference to comparisons made between IE and extended case study method (35), and in the remaining references it is discussed with regard to attaining experience-based knowledge as data through dialogue between the ethnographer and field observations (125, 150, 160).

In detail, the main body of data consists of partially transcribed videotapes and observations from maternity healthcare appointments (69), professional team planning meetings (11), training sessions for nurses (four during the spring of 2007), a training afternoon for the members of multi-professional teams, and family counselling classes (eight). Pregnant women's and public health nurses' interviews (seven of each), guides and handouts distributed to families, forms on pregnancies kept by public health nurses, and local and nationwide guideline material for care work were also collected. (For further details seeResearch materials).

The decision to use videotapes was originally made for quite practical reasons. As mentioned above, I collected my data as part of a larger research project, and the other researchers’ methodological orientation was towards conversation analysis. They therefore used videotapes as their main form of data. Since I saw no great problems in using videotapes as a source, it was decided that the appointments and team meetings would be video recorded without a researcher present in the room. Although the use of video has not always been really common in ethnographic fieldwork, I do not see it as a particularly problematic method. In fact, the videotapes worked as a sort of a memory aid for me: I could always go back to the detailed recordings of appointments in a way that mere memory and field notes did not allow. This is something that is described as an epistemological advantage of video recordings as ethnographic data: they allow the observation of details that are not visible or memorable to participants (including the researcher) in a given situation, and provide access for researchers who have not participated in the collection of the material (Peräkylä 1997; Knoblauch 2006, 73).

Further, there is an intimate bond between video and ethnography: video is audio-visual observation automatised into video technology. Both video and observation allow the addressing of ‘natural situations’ and the conduct of people in those situations (Knoblauch 2006, 71; Knoblauch & Schnettler & Raab 2006).

However, there are certainly methodological consequences of having a camera present at an appointment or a team meeting rather than having an observer present. For instance a camera can capture a limited range of the environment in comparison to an observer, and

is a different kind of co-producer of knowledge than an observer in each given situation.

There is a debate over the degree to which video is participatory and contributes to the actions compared to observation (Knoblauch 2006, 71). Certainly, video equipment can be obtrusive to action, but there are situations where the presence of an observer is even more so (Lehn & Heath 2006). In my own study, I made observations supporting both views.

When I asked pregnant women and couples to participate in the study before they went in the appointment rooms, the use of video quite often aroused doubt and suspicion. People seemed to feel uncomfortable about their actions being recorded, and told me that they would not be able to act ‘naturally’. Quite soon I noticed that mentioning early on that I, myself, would not be present in the room and that the cameras were quite small and situated high up in the corners of the room seemed to ease their minds about giving consent. Often, after the appointments were over and I went in to turn off the cameras, they commented on how fast they forgot about the cameras and, contrary to their prior suspicions, did not think that the cameras distorted the situation at all. I suppose, the default opposition to camera use has to do with people not just being used to such a method of data collection and, possibly, to cultural imagery related to surveillance cameras. Despite this, people gave their consent in over 90 per cent of cases, which is an extremely high rate of consent.

In addition to partial transcriptions of the appointments (1420 pages) and team meetings (311 pages) that I have used, I have also written separate field notes on the video recordings (272 hand written pages on appointments and 63 pages on team meetings: for detailed data information of the recordings see Research materials). The reason for this was that transcriptions of all the recordings, in addition to the other material, would have been too vast to handle. A full transcription of a first time appointment lasting 1 hour 38 minutes, for example, covered 89 pages, and a 64 minutes team meeting took 106 pages.

Because my research interests cover the whole range of maternity healthcare as a field of making unborn–maternal relations, I did not see it as necessary to get more transcribed material, which would have been the case for a study focused more on the particular and analysing structures and patterns in interaction of a kind typical for social science using video material, e.g. conversation analysis (Erickson 1988).

In the fourteen interviews I conducted, I applied different kinds of interview methods to try to go beyond articulations of abstract institutional discourse (Smith 1987; 2005), such as statements by nurses that they work in line with particular principles or recommendations set for maternity care work like ‘client-centredness’ or ‘family orientation’. (For the outlines of the interviews see Appendix I, and for data information see Research materials). Generally I tried to interview in an open-ended manner. I first asked the nurses and pregnant women to talk about their work and pregnancy in their own words, then asked them to elaborate on different issues such as what different principles meant for them in practice and to give concrete examples. (Smith 1987, 187–190; 2005, 135–142; see also DeVault & McCoy 2006.) Finally, if there was time, I introduced themes that were not touched upon earlier but were of interest to my research. Although the nurses and pregnant women responded to this line of interviewing, at some points it was still hard for me to gain an understanding of what they actually did when performing particular discursive principles, such as client-centredness, and what significance doing whatever they did had from their viewpoint. Thus, in spite of my efforts to apply ethnographic interviewing techniques, the nurses and pregnant women sometimes adhered to abstract descriptions and short examples within the terms of administrative and policy discourses on maternity and child healthcare. The interviews conducted took from 1 to 1.5 hours and they were transcribed in full (138 pages of nurses’ interviews and 132 pages of pregnant women’s interviews, single line spacing).⁵³

In contrast to the difficult moments in the interviews, the activities and discussion that took place in the workshops, particularly around video recordings of the nurses’

appointments, seemed to evoke specific stories and explanations of the nurses’ own experiences of doing maternity healthcare. The workshops, then, also became sites for collaborative knowledge production, as the nurses’ experiential knowledges were articulated alongside the scientific knowledge implicit in my comments and questions. I took notes of the workshops.

53 A few people employed by and/or working on their master’s thesis in the research project have transcribed video recordings and interviews collected and used by me: Sanni Tiitinen, Aku Kallio, Merja Viteli, Henna Kinnunen, Sanna Juurakko, Katri Keskinen, Sanna Tuominen and Niina Taimisto.

Observations and talks I had with nurses and pregnant women and their partners in the waiting rooms, hallways and coffee rooms of the clinics, family counselling classes (eight) and training events held for nurses on the nursing reform in Spring 2007 (four seminar days) were documented in field notes (500 hand written pages in total, including 48 pages of seminar handouts with my notes: see also Research materials). The documentary material consists, first, of the main guides for nursing in maternity healthcare identified by the nurses themselves: the latest national guide at the time,Screening and collaboration in maternity healthcare (1999), and a locally produced guide on the intranet, Handbook of Maternity Healthcare (2007) that was compiled precisely because at the time there were no up-to-date nursing guides for maternity heathcare. Further, I included everything the nurses introduced to me as material they use in their work: forms used in gathering information from pregnant women and their partners for assessment, guides and handouts about baby care and parenthood and Files for family counselling classes that introduced issues and materials for counselling classes. Finally, I familiarised myself with a few recent publications on maternity healthcare from the Finnish Ministry of Social Affairs and Health and the National Institute for Health and Welfare,⁵⁴ as well as municipal publications on the new nursing reform and some literature on which the new working methods were based. (For a complete list of documentary material see Research materials). The documentary material is part of the social setting under investigation (Hammersley & Atkinson 1995, 158), and can be seen as textually coordinating everyday work at the clinics (Smith 2005, 170–179).

Multiple sites and logics

In addition to considering my position in the field of research as dialogic, my fieldwork method in this study can be characterized as multi-sited. I follow Bruno Latour’s (2004a) suggestion to turn matters of fact into matters of concern and look at the common world as constitutive of contestation and debate. With this he is suggesting a turn to the politics of

54 The Institute is a result of merging the National Public Health Institute (KTL) and the National Research and Development Centre for Welfare and Health (STAKES) in 2009. Thus, publications from an earlier era are STAKES publications.

nature and science, and to the nature and science of politics. Ingnunn Moser (2008, 98) further develops Latour’s concern by saying that we do not understand the ‘working of power and dominance in the situations of multiplicity’. In the spirit of Michel Callon and Vololona Rabeharisoa (2003) she works ‘in the wild’ by showing how one matter of concern, namely Alzheimer’s disease, is being crafted in a range of locations and practices, from laboratory settings to clinical practices and civil movement networks. In a similar manner I worked with each location in maternity healthcare to show that they present different versions of what pregnancy concerns are and how they should be dealt with.

These locations include the site of doing policy and protocol, which in most cases is informed by statistical and quantitative research results. In other words, it is informed by a measurable change in a sufficiently large number of people, and is intended to alter the situations of individuals (cf. Mol 2002, 127–142). However, the relationship between individuals and populations, or care practice and policy practice, is not one of straightforward inclusion or causality. This has been noted in several prior studies, and it seems obvious in light of my own research as well (e.g. Berg 1998; Mol 2002; 2008a;

Ruppert 2011). Indeed, the (work) object in focus in these practices is not the same.

Rather, the sites of doing protocol and doing care work are complexly intertwined. This intertwining is also of interest to my research. It seems that there is much more involved in implementing protocol than adding the order embedded in protocol to the messiness in the clinical practice or replacing the former with the latter (see also Berg 1998, 228). In the following analysis, this is often realized in terms of the relations between the formal and informal aspects of work.

There are also the sites for enacting pregnancy in practising new and old working methods at the clinics: appointment work, home visits during pregnancy, multi-professional team meetings, and interviewing with the new ‘welfare assessment’ forms. The practice of working methods can be further divided empirically into more that one site of enacting pregnancy. Furthermore, there seems to be a (historically) relevant and quite specific division of labour between care work provided by the state and that provided by third-sector actors, and their coexistence in various forms of relation in turn shapes maternity

healthcare and thus concerns over pregnant women’s lives in Finland. These relations are such that both parties are able to rely on the division of labour and keep acting accordingly and are hence determined by their relations to each other. Exploring how different locations, matters and knowledges interfere with one another, in dialogue, then, is a method of bringing together different kinds of research material (see also Gordon &

Lahelma & Holland 2004).

In line with my methodological orientation, I have teased out and listed the material and discursive enactments of pregnancy concerns that relate to the relationship between pregnant women and the unborn, as well as the sites of those enactments in my fieldwork material. That is, I have explored the multiple relations that are done differently in screening for diseases or problems, treatment and advice on nutrition, the mental transition to parenthood and so on – in interviews, documentary material and videos and fieldnotes on observations. This has also involved analysing who or what does the enacting in a given site. After differentiating between the enactments and their spatial specifications I will map the ways in which these multiple relations are related to each other, i.e. how are they coordinated. In other words, the aim is to show how, with all the different enactments, frictions and differences in practices, in the end each and every pregnant woman is supplied with, if not a single diagnosis or assessment, at least a single treatment or a decision not to treat. Objects and relations may be done multiply in a given institution, but they are not fragmented: ‘they hang together’ (Mol 2008a) according to some logic.

The logic of coordination or ‘hanging together’ refers to both temporal and thematic rationales of doing care and pregnancy. It comes close to what Foucault (2007;

2008/1977–1979) means by biopower and biopolitics as forms of power, governance and politics obsessed with life, health and well-being of modern individuals as they appear in biological and social scientific cohorts or populations of people. My study is, then, informed by Foucault’s discussion of how, in modern liberal thought and power/knowledge regimes, potentials of people as ‘species’ are managed, regulated and maximised by means of mechanisms or ‘apparatuses’ that deal with rates, patterns and probabilities about a population and its ‘nature’ (Foucault 1990/1976; 2007; 2008/1977–

1979). I will refer to Foucauldian conceptualisations of biopower/politics where relevant along the way (see also Haraway 1997, 11–14; Deleuze 1990).⁵⁵

I have, however, decided to stick with the analytic concepts of logic and rationale (and synonyms alike). Logic as an analytical tool is derived originally from philosophy, and it seeks to formulate rational rules of reasoning (Mol 2008a, 8). The way I use it in my thesis is quite different and has fortunately been made easier by writers who have, first, undermined the seemingly universal rationalist logic (e.g. Nye 1990; Mol 2008a) and, then, addressed practices with it (Mol 2008a). Thus the concept is ‘stretched’ here, for the purpose of applying it to practices that are by no means coherent.

I will also talk about discourses and modes and styles of ordering, but they will always come after the local, fragile and pertinent rationales that have been uncovered as they are realised in practice. This is in line with my Smithian and STS project that is interested in how discourses and orders do not simply ‘speak over our intentions’ but also how the dialogue between intentions and givens of a (Foucauldian) discourse is fluid (Smith 2005, 127). Smith (2005, 127) claims that applying a Foucauldian concept of discourse seems to result in according it an overpowering role. In my view, this is the same argument made by STS writers like Annemarie Mol (2002), when she says ‘[Foucault] has been abandoned [by some writers] in his suggestion that society mimics organisms and thus hangs together in a single epistème’.⁵⁶ The solution Smith (2005, 127–143) offers is an adoption of Bakhtinian conceptualisation, where ‘each moment of discourse in action can be seen as both reproducing and remaking the discourse’. Mol (2002; 2008a), instead, abandons

55 I say ‘Foucauldian’ and not ‘Foucault’s’ conceptualisations, here, because it has been claimed that life and politics in the present are not precisely as Foucault conceptualised them. According to Deleuze (1990), for example, institutions and society as a whole are going through a change that situates them beyond Foucault’s

55 I say ‘Foucauldian’ and not ‘Foucault’s’ conceptualisations, here, because it has been claimed that life and politics in the present are not precisely as Foucault conceptualised them. According to Deleuze (1990), for example, institutions and society as a whole are going through a change that situates them beyond Foucault’s