Copyright Sociological Research Online, 1998

Lomax, H. and Casey, N. (1998) 'Recording Social Life: Reflexivity and Video Methodology'
Sociological Research Online, vol. 3, no. 2, <http://www.socresonline.org.uk/3/2/1.html>

Abstract

The degree to which researcher generated visual records (for example video texts) may be used to collect valid information about the social world is subject to considerable academic debate (cf. Feld and Williams, 1975; Gottdiener, 1979 and Grimshaw, 1982). On the one hand the method is assumed, by implication, to have limited impact on the data, the taped image being treated as a replica of the unrecorded event (Vihman and Greenlee, 1987; Vuchinich, 1986). On the other, it is suggested that the video camera has a uniquely distorting effect on the researched phenomenon (Gottdiener, 1979: p. 61; Heider, 1976: p. 49). Research participants, it is argued, demonstrate a reactive effect to the video process such that data is meaningful only if special precautions are taken to validate it. Strategies suggested include a covert approach to the data collection itself (cf. Eibl-Eibesfeldt and Haass, 1974; Gottdiener, 1979; Albrecht, 1985) or the application of triangulative techniques such as respondent validation (Gottdiener, 1979; Albrecht, 1985 and Arborelius and Timpka, 1990). In this paper we suggest that both these views are problematic. The insistence of one on marginalising the role of the research process and the other on attempting to separate the process from the research data is at the expense of exploring the degree to which the process helps socially and interactionally produce the data. As we demonstrate, the activity of data collection is constitutive of the very interaction which is then subsequently available for investigation. A reflexive analysis of this relationship is therefore essential. Video generated data is an ideal resource in as far as it can provide a faithful record of the process as an aspect of the naturally occurring interaction which comprises the research topic.

Keywords:

Conversation Analysis; Ethnomethodology; Reflexivity; Validity; Video Methods; Visual Analysis

Introduction

1.1

The value of audio-visual material has been well documented across various methodological perspectives (cf. Albrecht, 1985; Erikson and Schultz, 1982; Gottdeiner, 1979; Grimshaw, 1982) and across various disciplines, most notably anthropology in its use of photography, film and video (Feld and Williams, 1975; Heider, 1976). The ability to record the minutia of social life makes it an ideal method for a number of research objectives and theoretical approaches. It is, though, particularly appropriate for the naturalistic study of situated social interaction using the methodology of conversation analysis (cf. Goodwin, 1981; Heath, 1986; Mallett, 1990, 1993; Lomax, 1994; Lomax and Robinson, 1996). The main advantages can be briefly summarised. It is relatively easy to use and adaptable to most research situations (Goodwin, 1981; Heath, 1986). Moreover, it is argued that audio-visual recordings of human social activity provide a record which is more accurate, more detailed and more complete than that obtainable by unaided human observation (Gottdiener, 1979; Grimshaw, 1982; Hanson, 1994). Because of this it gives access to the richness and complexity of human interaction (Gottdiener, 1979; Grimshaw, 1982; Albrecht, 1985: pp. 307 - 8; Heath and Luff, 1993). The reliability of the data is, in a sense, self-evident because the recorded image may be repeatedly re-played, enabling analysis to be delayed until the researcher has left the field (Gottdiener, 1979; Albrecht, 1985) and allowing other researchers to conduct their own analysis (Gottdiener, 1979; Grimshaw, 1982; Arborelius and Timpka, 1990; Lundevall et al, 1994).

1.2

Despite these apparent advantages there is a relative scarcity of sociological research in which video has been used to generate data (though note Goodwin, 1981; Grimshaw, 1982; Albrecht, 1985 and Heath, 1986). As Heath (1986: p. 4) notes, much of the work involving the creation of visual data specifically for analysis has been developed by psychologists working within an experimental framework. Similarly, anthropologists have recognised the inestimable value of ethnographic film. The deficiency of video-based research in sociology has been attributed to a lack of information about how to analyse the data (Heath, 1986: p. 4). While we would not disagree with this, we suggest, in addition, that sociologists have been discouraged by the practical and theoretical difficulties associated with using video in natural settings, a situation which is compounded by the lack of clarity in the literature regarding the validity of video based methods. In particular there is a shortage of material which discusses the practical aspects of 'doing videoing' and the implications of this process for the production of data. Despite anthropologists' use of audio-visual data there has been a similar uncertainty or even silence in the relevant literature, Heider noting perhaps a little unfairly, that 'Few ethnographers have ever tried to answer the question of the effect of their own presence.' (Heider, 1976: p. 51).

1.3

A cursory review of the relevant literature identifies two main approaches to questions around the validity of the data. The first effectively denies that the researcher and the video camera (with its properties of preserving interaction for other contexts) have any effect or influence on the phenomenon being researched. Thus, it is presumed that the method in no way threatens the epistemological status of the data. Thus Margaret Mead, writing of a study undertaken with Gregory Bateson in 1936-38 was confident that the Balinese were unaffected by the presence of cameras; 'They were un-selfconscious about photography, accepting it as a part of a life which was in many ways always lived on a stage' (quoted in Heider, 1976: p. 51). The second approach, in contrast, is characterised by a suggestion that the research method must inevitably 'alter ... the representation of reality' (Heider, 1976: p. 51). The video recording process, it is claimed, intrudes upon social reality such that data has to be obtained covertly, or supplemented with respondent or comparative techniques.

1.4

This paper critically reviews these two perspectives and suggests that they take insufficient account of the research process as a social interaction. It then sets out an alternative approach to the use of video data. Based on our research which explores the interactional accomplishment of midwifery postnatal examinations and consultations (Lomax, 1994; Lomax and Robinson, 1996) we address the way in which the process of data collection helps socially construct and produce the data that is collected. Using examples from video transcripts and field note records, our analysis focuses upon the way in which the researcher and the research participants negotiate and manage the video based research process. We conclude that rather than being a source of bias to be ignored, marginalised or 'validated' out, the situated activity of the participants as research subjects is a valuable source of information and insight in to the research phenomenon. Further it is suggested that this reflexive methodological stance can enhance ethnographic and ethnomethodological research.

Ignoring and Marginalising the Video-Based Research Procedure

2.1

The view that the process of observing and recording research subjects with a video camera has little or no influence on the data which is collected, is not an uncommon position. Specific examples may be found in Vihman and Greenlee's (1987) study of phonological development for which mother-child interaction was recorded in the home, and Vuchinich's (1986) study of verbal family conflict in which family meals were video taped. Neither give an account of the process by which they obtained their data, claiming, in the case of Vihman and Greenlee, that the recorded talk is 'spontaneous'. The method, by implication, is presumed to have no impact on the data that is collected.

2.2

A further perspective within this general approach, is that which purports to have considered the effect of the method, finding it negative or unimportant. This position is taken by Bergstrom et al (1992) and Starr (1987) in their analyses of second stage labour during childbirth, and parent-child 'free-play' interaction respectively. Underpinning their position is the view that participants are so engrossed in what they are doing that, though they know they are being taped they become unaware, ignore or forget the presence of the camera and researcher. For example, Starr claims that 'participants knew they were being video-taped but did not appear uncomfortable' (Starr, 1987: p. 88) while Bergstrom states that 'all participants soon became used to the presence of the camera and usually ignored the researchers' (Bergstrom et. Al, 1992: p. 11). Neither offer empirical interactional evidence to support their assertions such that, while their claims may appear reasonable, there is no practical way of verifying them. An alternative interpretation which hints at the significance of the process is available. Bergstrom appears, by implication, to consider 'ignoring' the camera as an indicator of participant behaviour unaffected by the research presence. However, it seems reasonable to suggest that participants' 'ignoring' might be interpreted as an active state of not paying attention. It could be that participants, in the presence of the camera and/or researcher have interpreted their research role as one in which 'doing ignoring' was appropriate. The problem now is not the 'impact' of the camera but that it is not acknowledged nor explored empirically and reflexively.

2.3

A further assumption underpinning the 'no effect' position is that the 'behaviours' being researched are not under conscious control and thus are not easily altered by a research subject (cf. Heider 1976: p. 52). Weimann (1981) includes in this category, behaviours such as 'head nodding, gestures, foot movements, amount of other-directed gaze, time talking, sound-silence patterns and the like' (Wiemann, 1981: p. 302). However, ethnomethodologists have demonstrated empirically that these 'behaviours' are actually fundamental and mundane methods which members of society use to constitute an orderly world. As such they are employed in precise ways in particular social environments (cf. Livingstone, 1987: p. 10). Other-directed gaze, for instance, is a significant means by which actors display that they are listening and understanding, and secure a turn at talk (cf. Frankel, 1983: p. 29 and Goodwin, 1981: pp. 29 - 33).

2.4

This last point is explored elegantly by Heath (1986: pp. 11 - 13) in relation to video-based research. His analysis of a sequence in which participant attention is drawn to the partially hidden camera by his daughter's gazing and pointing activities, is important for two reasons. Firstly it demonstrates the way in which gaze, directed alternately at a co-participant and at the object of interest, may be a specific interactional device for eliciting a co-participant's 'noticing' and response. Secondly, it provides empirical evidence that the video camera, for at least the length of this sequence, is part of the situated activity of the consultation; the participants notice it, orient to its presence and properties, and structure responses accordingly. Thus, in contrast to the view that 'other-directed gaze' is a behaviour not subject to conscious control, Heath demonstrates that it is locally managed, has precise, context-specific interactional roles, and may be organised in response to the presence of the camera.

2.5

In short then, it may be methodologically problematic to simply ignore or negate the role of the research process in video-based methodology.

Video Research as Distorting

3.1

The second perspective holds that the video camera has a uniquely distorting effect on the researched phenomenon (Gottdiener, 1979) and implicitly, the 'naturalness' of social reality. Edmund Carpenter provides an anthropological example in suggesting that the behaviour of the inhabitants of the Sepik area of New Guinea changed completely once they became camera-conscious (Heider, 1976). There is therefore a choice. Firstly, the research could be done covertly, which immediately raises traditional ethical questions whatever the nature of the research. Relatedly, it has been suggested that 'distortion' can be eradicated by such means as erecting screens to hide the camera (Hanson, 1994: p. 221). Clearly, this would not hide members' awareness of the video equipment and as we shall see their orientation to the process of recording is of central importance.

3.2

A second remedy argues that validity can be restored and promoted by combining video-based research with other methods, such as respondent interview (Gottdiener, 1979) or a combination of observational techniques. For example, comparative techniques typically involve measuring participant behaviours across two or more different research conditions, such as audio-recording and naked-eye observation. Carpenter and Merkel (1988) compared the effects of three different research techniques - audio-recording, video-recording, and a one-way mirror, on the interactions of thirty couples. They claim that no significant differences were found between the three recording devices when gauging the couples' interactions, anxiety and feelings expressed towards partners. This triangulative exercise is used as evidence for the research's validity. However, as the authors themselves note, the research is only able to demonstrate that participants respond similarly to each of the interventions. It does not explore how observed participants' activities compare with unobserved activities, nor does it treat the processes as topics of research in their own right. The limitations of the experimental design are highlighted further by the authors' own acknowledgement that the research appeared to have constituted an intervention in itself, as one third of couples continued the therapeutic exercises after the observational period.

3.3

Respondent validation techniques are a second method by which it is purported that the validity of video methods can be increased. These usually involve interviewing or surveying participants post video-taping, asking, for example, how they felt about the process and whether the camera presence caused them to behave 'differently'. This technique is used by Martin and Martin (1984) who surveyed over 250 patients video-taped while consulting their general practitioner. Of this group, thirty percent reported that they had forgotten the camera, compared to six percent who claimed to be constantly aware that they were being recorded. Less than two percent felt that the camera presence made a difference to the way they were treated by the doctor and only six percent claimed that they were less willing to talk about embarrassing problems because of the presence of the camera.

3.4

Similarly Erikson and Schultz (1982) in their smaller qualitative study, report that two of the four counsellors whom they had video-taped felt 'slightly nervous' at the beginning of the first video taped consultation but then became 'not camera conscious' as the research progressed (Erikson and Schultz, 1982: p. 56). A further validation procedure was carried out by showing the films to other school counsellors and administrators who 'uniformly agreed' that the films were 'true to life for the kind of interview that was being conducted' (Erikson and Schultz, 1982: p. 56).

3.5

The assumption underpinning each of these strategies is that participants' subsequent accounts of being researched are superior to the recorded social reality. This seems problematic for several reasons but we will focus on one specific flaw. The assumption that participants will have a privileged insight into social interactions of which they form a part is questionable not least because it ignores the point of video-based research. Video- taping is usually undertaken with the express purpose of identifying actual behaviour as opposed to reported behaviour in situations where participants are not always aware of their actual activities and the activities of others. It seems inconsistent, therefore, to assume that for the purposes of the research findings, observational analyses are superior to self-reports but that in validating those findings the latter are more significant. Thus, respondent validation, rather than helping analysis of the empirical reality, merely produces a reconstituted second-order account. This may be worthy of analysis in its own right but it is a separate phenomenon.

3.6

To argue that video methods distort reality and must be compensated for, is to lose sight of the very point of collecting and analysing video taped interaction. As we shall go on to show, the process of preservable and re-presentable data collection helps to actively constitute the interaction being recorded and analysed and thus, a reflexivity is essential in analysing that data. Far from being a distraction or unimportant, a reflexive analysis of the research process can contribute to an understanding of the phenomenon under investigation.

Video Methods and Reflexivity

4.1

In the second part of this article we suggest that it is more instructive to take a reflexive approach to the data. Our approach is informed by the perspective that the researcher is inevitably part of the social world that is being studied (Hammersley and Atkinson, 1991). Thus, in relation to video-based methods, the researcher is an active participant in the situated activity that is being recorded. But the camera too, is socially significant given both its ability to preserve interaction for re-presentation and participants' awareness of that ability (a characteristic which sets visual data apart from other forms of observational data). Both Goodwin (1981) and, more particularly Heath (1986) acknowledge this in their work which uses video:

If we are to make an empirical case for the effects of recording on interaction, then we need to demonstrate an orientation by the participants themselves to the production of their action and activity to some aspect of the recording equipment. (Heath, 1986: p. 176)

4.2

Our approach involves firstly an acknowledgement that the video text (ie. the collection of visual and aural images evident on a monitor) is a product of the occasioned activities of the researchers and participants, and secondly, a resource for exploring the interactional production of those activities. The facility of video to record, albeit partially, a version of the research process is unique in that it enables an analysis of the contribution of that process to the production of the data. Thus, here we demonstrate and explore the possibilities of a reflexive video-based methodology.

4.3

With reference to our data on midwife-client interaction we propose two key points. Firstly, that the researcher, by doing 'being a researcher', reflexively contributes to the videoed definition of the encounter. This is not least because she must make choices within the context of interaction about when and how to film the event and when it is more felicitous not to film. These actions are influenced, in part, by what is 'appropriate' behaviour in the field. Our second point is that the participants themselves display an orientation to the research process and the recording equipment and its properties. That is, at any given moment they are both midwives or clients and research participants. Thus the suggestion is that participants' talk and activity exhibits a self awareness and orientation to themselves as both the objects of research and to the preservability of the encounter.

4.4

In the remainder of the article we discuss ways in which the research process helps constitute the data that is collected, and in so doing, provides further analytic resources. Specifically, this is discussed in relation to data on: initiating the consultation; researcher self- management and the video-taping of the midwife's examination of the client's body. Data is presented in the form of transcribed excerpts from the video tapes, sound files, video stills and extracts from the researcher's field notes. Sound files have been marginally distorted, faces on stills blurred and all names changed in order to preserve participants anonymity.

1. Initiating the Consultation

5.1

We have suggested that the role of the video-researcher in the research setting is relatively unexplored. One response is to advocate that she ought to minimise both her presence and the presence of the recording equipment, through, for example, the use of one-way screens or covert recording. Because of the nature of our research, which necessitated gaining ethical approval, and recording consultations in clients' homes, often at short notice, both of these were untenable and, in terms of the former, methodologically problematic. Invariably, the equipment would be set up in full view of the participants at the beginning of the consultation. Here, we want to discuss the implications of this practice for the data that was produced using examples from the video data.

5.2

Starting the video- taping, that is, switching on the video camera, an activity which might, at first, seem a relatively straight forward business, became, in our research, a matter of some complexity and analytic interest. Aside from the practical obstacles of recording the interaction from the doorstep, it became apparent during data collection, that, even with specific arrangements, it is not possible to enter a person's home and set up camera without becoming interactionally involved. At the very least, the researcher is required to do greetings to co-ordinate entry into a conversation (Schegloff, 1968) and accomplish the more complicated matter of entering a persons' home in an appropriate way, which in turn relates to the need to accomplish an amenable research context. The implications for the data are that opening sequences are likely to include the researcher. In the following extract the client (C) at line 3 elicits a return greeting from the researcher (R, line 5).

Sound clip     Help

5.3

In addition the researchers' presence may generate talk unrelated to the business of the consultation. For example, in this visit the researcher at lines 32 to 34 is drawn into an exchange with the client and her husband (H) concerning the noticeably long wait she has experienced in her car. This talk, around an observable and, in its most mundane sense, remarkable phenomenon provides a source of informal topic prior to the start of business.

Sound clip     Help

5.4

Looking at the physical location of this talk and activity, the video-tape reveals that participants are, at this juncture still engaged in 'preliminaries' (Heath, 1986). That is, they are still standing in the hall, having yet to organise themselves into an appropriate location for the commencement of 'business proper'. Visual analysis of start sequences elsewhere (Lomax, 1994) indicates that the 'start proper' occurs after participants are seated, the medical records are in the midwife's possession and is frequently associated with a summons to start, for example:

'Right so how are you feeling then?' (S19)

and

'Right so how's everything going? (S2).

5.5

Thus before the visit proper may commence, participants must organise themselves in to an appropriate location such as a sitting room or bedroom. This apparently unremarkable activity involves some considerable verbal and spatial managing. Visitors and partners must be ushered out along with any older children. Partners too are invariably relegated to making tea or looking after older children. During these preliminaries, greetings are exchanged, invitations to come in are proffered and refreshments may be offered. In addition, while the midwife organises herself and her equipment she may offer a number of mundane comments around local, observable phenomenon such as the baby's sunshade: 'I like his sunshade' (S30) or 'grandma doing the ironing' (S30). As Bergmann (1990) notes, the topicalisation of silent objects and events in this way is a means by which participants ensure the continuation of interaction (1990: p. 211). Our analysis demonstrates further that these items are not unremarkable 'mundane' utterances, rather, this talk has a precise function within the encounter. It facilitates a particular physical activity associated with preparing for 'midwifery' activity but at a juncture where 'midwifery' talk is not yet appropriate or possible.

5.6

Our point here therefore is that the researcher herself is part of a continuing interaction but her presence is not a problem of distortion for the analysis. Neither can it be ignored. Returning to our original sequence (S27), for example, the video tape reveals it is several minutes before the 'business of the visit' may commence, participants having yet to organise themselves in the lounge, a delay which is, in part, precipitated by the client making tea. Analysis of the interaction occurring during these delays demonstrates that one of the ways that this is managed is by talk with the researcher about the data collection for the research project:

Sound clip     Help

5.7

This fragment is proceeded by talk concerning a programme hearable on the television concerning home births, talk which is closed by the midwife's utterances in lines 120 - 121:

Sound clip     Help

5.8

As can be seen from the data, the client's response to the midwife in lines 120 - 121:

is evidence that she hears the midwife's utterance as closure. She takes a final step towards establishing an appropriate context for the visit (turning off the television) and then waits for the midwife to speak (pause between lines 124 and 125). The midwife's proceeding utterance:

marks the start of the business of the visit, constituting a summons for the client to talk on 'business'. In common with research on other health professionals (ten-Have, 1991; Robinson, 1987) our data reveals that midwives exercise the right to open the visit in this way, preferring a client response which co-operates in starting business (Lomax and Robinson, 1996).

S27 Video still

122 C Do you want that off is it better with that off?

5.9

To summarise the examples discussed so far. In addition to providing evidence that maintaining a 'low profile' as advocated by some of the literature discussed earlier is practically difficult to accomplish, analysis of the researcher's elicited involvement in the talk at these junctures and the way it helps organise a non-formal environment prior to business, is inferentially rich. Our next example, again focusing on beginning the midwifery consultation, emphasises the role of researcher as participant in the interaction and the way in which all participants may orient to the camera and its properties. It also reveals the way in which the reflexively handled video text provides a multi- faceted resource for researchers.

5.10

The corpus of data revealed various examples in which the midwife and the researcher mutually negotiated the beginning of the consultation where, for the midwife, the beginnings of the interaction and the consultation were evidently different. On several occasions midwives made it clear that, in their view, the work of initiating the consultation, that is, doing greetings, informal 'chat', deciding 'what needs doing' and explaining the purpose of the consultation, were not 'important' and thus would not need to be recorded. This provides an interesting contrast with the professional midwifery literature which highlights the importance of 'talking work' while downplaying that of technical care (Rider, 1984 and Murphey-Black, 1994).

5.11

Looking at the data more closely, in excerpt five for example, the midwife enters the hospital bay and begins talking with the client about what 'needs doing'. Their talk, which involves a fairly lengthy discussion about breast feeding in order to establish whether the baby needs vitamin supplements, is recorded by the microphone, the camera being placed ready outside the curtains. The researcher, on returning to the bay attempts to put the camera in, in response to which (lines 34 and 36) the midwife tells the researcher that the consultation has not in fact started yet.

Sound clip     Help

5.12

This definition of the encounter by midwives became increasingly evident during the course of the data collection such that, as the following data from the field notes illustrate, it became, on occasion, socially problematic to turn on the camera:

Videoed S7 on the 7th March. This was the first in a series of three videos carried out on the same day. Present were 'Sarah' [midwife], 'Mary' [mother], 'Veronica' [student midwife], a friend of Mary's and the baby. Mary's husband was working in the shop at the front of the house and did not participate in the visit . I arrived after everyone else and was directed to the lounge by him. On entering I was surprised to find everyone eating cake and drinking tea. The atmosphere was very jolly and there was a lot of laughter. The initial impression was that this was more like a party than a visit by a midwife. It was the tenth day so Sarah was discharging Mary, although this did not happen until later in the visit. As I arrived Mary got up and fetched me tea and cake which she was quite insistent that I accept. They were talking about all sorts of things not immediately identifiable as 'midwifery' concerns such as Mary saying the cake was baked by a friend and that people were especially welcome when they brought food. The 'business of the visit' - examination of mother and baby and accompanying paper work - did not start until 5 - 10 minutes after I arrived. I found this really problematic because I wanted desperately to record the interaction but felt socially unable to. Reflecting on this experience, while outwardly I accepted the cake, thanking Mary profusely for it and behaving (I think!) as if the situation I found myself in was un- problematic, inwardly my head was whirling with ideas: What are the implications for my data and most of all for my methodology chapter? On the positive side - data collection is going well - participants seem very ready to accept me. On the negative side, the apparent socially inappropriateness of switching on my camera means that I am missing some really interesting data. For better or worse I decided to go with my 'gut instinct' and leave the camera in its case. My feelings that it was inappropriate to video at this juncture appear to be confirmed by the way in which the initiation of the 'midwifery' activity of the visit was organised. The start of the encounter proper was characterised by Sarah draining her tea, putting down her cup, and getting out the patient notes, actions which were accompanied by her physical transition from the settee to the floor. (S7, home visit, field note diary)

5.13

The decision as to when to turn off the video camera generated similar issues. When planning a data collection strategy it had been proposed to video the entire sequence, including the midwife's closure and departure. In practice, as the following video and field note data illustrate, the same concerns which emerged around when to turn the camera on influenced the decision to turn off the camera.

Once I had set up the camera I left the room and went downstairs. After about 5-10 minutes Sarina (the midwife) shouted to me 'we've finished', a statement which I interpreted as a summons to switch of the video camera, which I did. Reviewing this experience in the context of both the events which occurred during the visit after the camera had been switched off and the consultations videoed so far, it is becoming apparent that there are certain activities which the midwives construe as 'midwifery' and that they perceive I will be interested in researching (the physical examination, bathing the baby, helping mum breast-feed) and others which, although observable in each of the visits I will not be interested in ('social' talk occurring at the beginning and end of a visit, making arrangements for a subsequent visit). On this occasion, it being difficult to explain in the context of the visit that in fact I was interested in interaction other than that around the physical examination of the mother and child, I went along with the assumption and switched off the camera. However, my feelings of disappointment (that this data was not being video taped) grew as Sarina discussed with Paula the arrangements for the following visit, an activity which to an outsider like myself clearly constituted a midwifery activity. (S12, home visit, field notes)

And the video transcript from the same visit:

Sound clip     Help

5.14

In summary, these data indicate that the researcher, in order to maintain an amenable research context was obliged to accept the midwives' definition of the encounter, at least while in the field. This in itself is methodologically interesting. In the context of visual research it is of further interest, not least because this insight is a direct result of the data collection process; that is, the researcher's use (and non-use) of a video camera. It is doubtful that taking notes would have elicited the same commentary from participants. Of further methodological interest is that the process by which researcher and midwife negotiate what may or may not be recorded is captured on video and available for reflection and analysis. Thus, while those sceptical of the method might hold that the phenomenon of the negotiation around the opening and closure of the encounter is either insignificant or an example of data being made 'unnatural' by the intrusion of the researcher, we suggest that a reflexive analysis of the interaction is beneficial both in terms of the original goal of the research (the interactional accomplishment of midwives' work) and informing researchers about their own position and the role of the equipment in the research process. Thus, whilst work on medical consultations might recognise the wider structure of the overall visit (see for instance Heath, 1986 and Robinson, 1987) the midwife's utterances mark her distinction between 'proper' activity- based midwifery (yet to take place) and organising and talking about this with the client. Analysis of the research process, then, can tell us something about how midwives define their work and indeed how they perceive what is appropriately researchable (the physical, practical examination), a definition which is reflexively sustained by the researchers' activities in the field.

2. Managing Non- Participation

6.1

Unlike either the participant observer or the field note observer, the video researcher has no obvious role in the setting. We have already suggested that there are difficulties about being present but communicatively dis-engaged because of interactional pressures to talk around greetings, eye contact, etc. In this section we discuss reflexively, ways in which the researcher managed this expectation during the encounter and the implications this had for the production of the data.

6.2

One method involved the researcher extracting herself from the setting, an approach which had been partially prompted by sources which had advocated minimal researcher involvement (cf. Robson, 1993: pp. 208 - 209; Hanson, 1994: p. 219; Bergstrom, 1992: p. 11 and Erikson and Schultz, 1982: p. 55). However this was not always possible to implement in orderly ways. Whereas in the public space of the hospital, it was possible, and to a large extent expected, that the researcher would not stay in the confined hospital bay, as a visitor to a client's home it quickly became apparent that, in order to 'exit' from the research setting one had to present a legitimate 'social' reason for doing so. One means of achieving this was to offer to help with tea. However, this was not always possible, being contingent upon someone offering refreshments and then accepting the researcher's offer of assistance. The point here is that in order to achieve 'invisibility' the researcher has to accountably withdraw. Data extract seven, taken from the opening sequence of home visit S5, provides such an example.

Sound clip     Help

6.3

Looking at this sequence in detail, the researcher, in response to midwife's initiation of the 'business of the visit' in line seven, begins to leave the sitting room:

her footsteps can be heard on the stairs during the client's utterance at lines nine to ten:

6.4

In fact, her strategy was not entirely successful. As can be seen from the data, the researcher departs only to re- emerge a few moments later in line 19:

6.5

The field note entry for this visit is of interest here:

Video-taped my fifth home visit today...... Having already experimented already with several ways of managing my presence during videoing, decided to try and leave the setting using the same departure strategy as I had with S3. On that occasion I discovered inadvertently that once out of the immediate environment it is possible to become engrossed in conversation with others until almost the end of the visit. In this way I am able to avoid becoming involved with the visit itself. Did not go entirely to plan. Decided to 'leave' by going to the bathroom. Leaving was okay, didn't feel it necessary to say anything as Isobel (midwife) and Carol (mother) were deep in conversation and any explanation would have constituted an interruption to proceedings. However, when I got to the top - no bathroom !! Rather embarrassingly, I had to make my way back down the stairs (which were in the lounge) and interrupt them in order to offer an explanation for my actions (in fact neither Carol or Isobel asked me what I had been doing however I felt obliged to account for wandering uninvited around a relative stranger's house). In fact, after being told where the bathroom was I managed to engage the client's mum (who had been hanging out washing) in conversation - she seemed happy to talk at length about the birth of her grandchild. In this way I managed to remain outside the lounge where the majority of the visit took place and uninvolved in the interaction for the most part of the consultation. (S5, home visit, field notes)

6.6

Reflecting on this diary entry some months later raises a number of issues. Firstly, the researcher's criteria of 'successful' and 'unsuccessful' data collection strategies needs considering in the light of the methodological literature. As we have already suggested, the decision to depart from the setting was informed by the 'minimal involvement' approach which suggested that researcher presence was inappropriate. This raises an interesting dilemma. While on the one hand, those using such a method, know, experientially, that this is neither realistic nor possible, the literature encourages the view that examples of the researcher's presence in the data can only be viewed negatively, as an example of poor fieldwork, for example. This may account for the fact that such examples are rarely seen in published work and, where they are present, tend to be consigned to a footnote explicating the 'limitations' of video. It is our view that, rather than dismissing such sequences, it is more helpful to explore their implications for the research topic. How, for example, do the participants respond to the researcher's presence, in this case her interruption? What does this tell us about the particular visit, the methodology and the research question? As we argue elsewhere in this paper, one of the advantages of video is that these activities are available for analysis on the video tape.

6.7

Taking the example illustrated here, it is possible to see from the video and the transcript that both midwife and client respond to the researcher's utterance with a verbal and non-verbal description of where the loo is:

S5 Video Still: Midwife and client gesturing to the back of the house.

6.8

Interestingly, the midwife then offers:

an utterance which both accounts for her knowledge of the layout of the house and her ability therefore to offer directions. But, importantly, it also offers a justification for her giving such information when she, like the researcher, is a guest in the client's house. In addition, both participants deal positively with the interruption. Directions to the loo are given, the midwife responds 'yeah that's okay', both laugh and, most importantly, the client's account of her labour is resumed moments later prompted by the midwife in line 24:

6.9

Thus, in summary, analysis of this sequence reveals firstly that respondents deal positively with interruption and, secondly, that they are able to return to the business at hand relatively smoothly. This is perhaps not surprising given the nature of midwifery work. It is evident from the data set as a whole that midwives visiting mothers are routinely interrupted by unexpected visitors (the milkman and the catalogue company, for example). An example from the data set is provided in the following sequence:

Sound clip     Help

6.10

Looking at this sequence in more detail we can see that, following an interruption by the toddler (T) and the client's actions to deal with this at lines 58-60:

6.11

The midwife is able to resume the original topic (the baby's trip to casualty) in lines 66 and 68. Her summary of the story so far provides an opportunity for the client to finish the story, an offer which she takes up in line 69:

6.12

Thus, prompted by the 'problem' of the researcher involvement, analysis of these sequences provides a further means of exploring 'what midwives do'. As we can see from these data, accomplishing a home visit is, in part, about dealing with interruption in a socially appropriate way while still accomplishing 'accountable' midwifery work. One final point regarding these sequences, it was interesting that the midwife in S5 felt able to offer an explanation (and by implication permission to use) about the whereabouts of the loo. While she qualifies this with an explanation in her proceeding utterance, her statement is indicative of her status in the household. It would be difficult to imagine other visitors (the milkman, the catalogue company) offering such invitations. Her statement tells us something about her own perception of her status in the household, while the fact that this is unchallenged by the client indicates at least verbal acceptance of this assumed status.

6.13

Returning to the main argument, the data set reveals that, in the majority of cases there was little alternative but for the researcher to remain in the room with participants. Borrowing from the 'minimal involvement' literature already cited, the researcher could dis-attend from the interaction, for example, by sitting away from the camera and attending to the interaction only when called to do so. Although in many ways this is not unlike the strategy that a field-note observer might adopt, there is one significant difference: in contrast to the researcher who is busy composing a field narrative and is thus observably and accountably engaged, the video researcher appears to be 'doing nothing'. This presents problems for remaining dis-engaged from the interaction. The silent observer is available for inclusion in the activity she/he is observing and at the very least may be expected to demonstrate some non-verbal involvement. In our data, for example, occasional glances toward the researcher from both midwife and client, meant that, at the very least, appropriate facial responses were required, for example a visual acknowledgement of the client's discomfort at witnessing her baby's crying during a blood test (S30) or, as in the following example, an invitation to laugh by the client. As can be seen from these data, the client gives the researcher a sideways glance and silent 'giggle' following a verbal admonishment from the midwife about posture during breast feeding:

Sound clip     Help

6.14

A further example of a response to an invitation to laugh occurs later in the same visit. As can be seen in the following excerpt, the researcher is invited by both the midwife and client to acknowledge a joke which involves the construction of the baby as a potential footballer:

Sound clip     Help

6.15

Specifically, the researcher's response is solicited by the client at lines 1- 3:

and 5;

6.16

Both of the clients' utterances are completed by a shift in gaze from the object of the talk (the socks) to a designated recipient of the 'joke', the researcher. Because of the position of the camera it is not possible to see whether the researcher responds non-verbally but the researcher does laugh audibly in line 12 following the midwife's uptake of the 'joke' in line 6. The midwife's utterance 'are you gonna be a footballer' culminates in laughter and a shift in her gaze from the baby to the researcher. At the same time the client, responding to the midwife's utterance, laughs and also looks over at the researcher. Their laughter utterances provide the researcher with a turn in which she can accept or decline the opportunity to laugh (Jefferson, 1979). The researcher takes up the preferred option of co-operating in laughter.

6.17

The involvement of the researcher might, as we discussed earlier, be perceived as a problem. However, as we demonstrate with the following example, it is our view that these sequences, as long as they are reflexively handled, can be a useful resource. The requirement for the researcher to make specific verbal responses became more apparent towards the end of visits, after most of the specifically task-centred activities had been completed. The following excerpt, for example, occurs during 'chat' towards the end of the same visit while the midwife is writing in the notes.

Sound clip     Help

S27: Midwife writing in the notes

6.18

Looking at this sequence in detail, the client's utterance in line one is preceded by a pause of 30 seconds, a significantly long time in 'mundane' conversation but one which is made accountable by the note-writing which requires the midwife's visual attention. Immediately prior to her utterance in lines 1-2, the client glances over at the midwife and then looks at the fish tank opposite her. As she looks a smile spreads across her face and she comments about the fish tank;

6.19

Initially the midwife responds to the client's utterance, breaking off from her paper work to comment about the fish tank in lines 3 and 5. The client's shift in gaze during her laughter in line 6 invites a response from the researcher which is forthcoming in the form of a short laugh in line 7. A pause of 4.5 seconds then occurs during which the midwife resumes her paper-work. The client then adopts a middle-distance gaze but continues to giggle silently to herself. The pause is then broken by the researcher who offers a further comment on the 'fish' topic:

an utterance which generates a sequence of talk extending over several conversational turns during which the client and herself talk on the topic of fish- keeping.

6.20

The researcher's participation here is again interesting rather than contaminating. It can be observed that the client's proffered topic at line one centres on a 'noticing' (Sacks, 1995; Bergmann, 1990), her comments serving to 'trigger' an extended sequence of talk on this topic between herself and the researcher during a lengthy period of administrative work by the midwife. As we have noted earlier in the paper, silent local objects may be used by participants to maintain interactional involvement at junctures were talk might potentially falter. It was noted earlier that this is frequently where overtly 'midwifery' talk is either not appropriate or not possible. For example, looking at the data set as a whole, it would appear that preparing for the physical examination, physical transitions from one room to another, and organising equipment are interactionally managed by participants through recourse to 'noticing' based topics such as house decor, furnishings, baby clothes and equipment (Lomax, 1994).

6.21

Accomplishing paperwork presents similar interactional problems for participants in that one participant is potentially interactionally unavailable. The midwife's actions necessitate a potentially dis-preferred silence which she may manage with prefatory comments about 'all this paper work' as in excerpt 12 and 13 (lines 171 and 342 respectively) and the accompanying video clips. However, where the researcher is also present with the client, a dis-preferred silence can develop in that two speakers are unaccountably silent. Thus, returning to the 'fish' example (S27) cited earlier, it can be seen that this sequence is a means by which the researcher and client manage any interactional awkwardness, an activity which in turn facilitates the accountable midwifery activity of accomplishing record keeping prior to the closure of the visit.

Sound clip     Help

171    M     Sorry it's a little bit lengthy with the paper work today

Sound clip     Help

342    M     pieces of paper everywhere

6.22

The general point is that the research is not marred by the necessary involvement of the researcher but, conversely, she is a contributor to the constitution of the interaction. Further, the involvement of the researcher in the interaction can be analysed and understood from the video text. The analysis, in turn, is informative about 'normal' consultations; ie. how midwives organise an overall structure of the visit.

3. Managing Body Taboos

7.1

In this section we suggest that participants' talk and activity (including that of the researcher) displays an orientation to themselves as participants within a research process and to the preservability of the video taped encounter. Again it is further suggested that reflexive analysis of this phenomenon enhances the research.

7.2

The principle goal of the original research had been to explore the ways in which midwives and clients interactionally organise the examination of the client's body; that is, to explicate the ways in which body taboos associated with exposure and touch, particularly to do with sexual areas of the body, are managed. However, further significant insights were obtained by exploring ways in which the participants orient to the camera during the consultation and specifically for the physical examination. Most obviously it was recurrently noticeable that whilst no clients expressed difficulties at being video-taped for the examination of the abdomen, legs and other non-intimate body parts [see S27 and S30 examination sequences overleaf], several of the clients requested that the perineal examination, which requires exposure of the anus and genitalia, should not be video-taped. For example;

Sound clip     Help

3    C     Can we ur have the video off for this?

7.3

Similarly, of the midwives in the study, at least one indicated great reluctance in allowing the perineal examination to be video-taped. Others sought to create privacy against the camera, for instance by holding up a sheet while examining the client's nipples (see S12 still below) or, as the following two sequences demonstrate, standing between the client's buttocks and the camera so that the perineum would not be video taped (at lines nine and 18 respectively).

S12 Breast examination

Sound clip     Help

1    M     that's lovely your womb dear

9    M     I'll stand in front of the camera

Sound clip     Help

18    M     Let's have a look I'll stand in front of the camera so that (no one) can see ahahaha

7.4

Looking at the midwives' activities in the context of the data set as a whole, their actions are comparable with midwives' creation of privacy from non-medical others. A consistent feature of the organisation of the physical examination is that it was rarely conducted in front of other adult members of the client's family, including those whom we might expect to be accustomed to the client's nudity. In the hospital this was policed by the drawing of curtains around the client's bed and the ushering out of visitors by the consulting midwife. In the home it was observed that if husbands were present, the consultation would take place in clients' bedrooms. In the majority of cases, however, this was pre-empted as husbands and others left the house or at least the sitting room upon the arrival of the midwife. Thus, while it was acceptable for nurses and trainee midwives to observe the examination, others were actively excluded from proceedings such that failure to leave by these others resulted in the relocation of proceedings to a different room. The following excerpt seventeen is a case in point. Participants (midwife, client and student midwife) relocate themselves to the bedroom, the client's husband is not invited, neither does he invite himself to take part in proceedings. Indeed, the midwife's invitation to the client: 'shall we go upstairs then?' appears to specifically preclude him. Having secured this privacy, the importance of maintaining it is evidenced in the midwife's comments about shutting the door in lines three to seven below.

Sound clip     Help

7.5

These observations are comparable with Emerson's study of hospital based, gynaecological examinations in which she suggests that the exclusion of lay persons from the examination room delineates 'the contact of medicine' and the 'contact of intimacy' (Emerson, 1970: p. 81). This point may be further illustrated with reference to examples of talk about the video-taping of the examination. In the first of these, one midwife informs the client of the 'good news' that, as she has an intact perineum, she won't require a perineal examination and will therefore not be having this part of the examination video-taped (S12, home visit). Rather than seeing this as a problem for the method, the shared laughter around this sequence may be seen as supportive of primary analyses which suggest held-in common cultural beliefs that genital examination is unpleasant and best avoided (Ragan, 1990: p. 77).

7.6

Similarly, on another occasion in response to the researcher's request to video tape a hospital consultation, the midwife's utterance expresses surprise that she would want to '... video the whole thing, even the.....' (S10, hospital). In this case the midwife cannot, apparently, bring herself to utter the words associated with the perineal examination. Again, however, this need not be perceived as a problem for the research. In fact, our primary analysis indicates that the absence of formal terminology and its replacement with euphemism and vague references is a consistent feature of the organisation of the examination. This is in contrast to the way that participants talk about and organise the examination of non-sexual body parts. As we have discussed in an earlier paper (Lomax, 1994), the social organisation of examination sequences which require undressing and exposure of the body, in particular genital exposure may involve the relocation of activity to a bedroom, the use of euphemistic language to describe the procedure ('little check up' and 'peak at your stitches') and an absence of eye contact and talk during the procedure itself.

7.7

Thus, examinations, in the way that they are socially organised by participants, constitute and reproduce a taboo around sight, touch and reference to the genital area (Emerson, 1970; Ragan, 1990 and Lomax, 1994). Thus, while midwives comments to the researcher appear to exacerbate the delicacy of the examination, the argument is that this is not contaminating but instructive: participants reactions to the video camera, merely confirmed other methods by which the delicacy of the practice was constructed. Further, the researcher's evolving awareness of various constitutive processes in the development of the taboo and its captured analysable presence on the video text allowed her to become necessarily sensitive not only to the existence of the taboo but also to her role in its 'here and now' creation. Thus, specific arrangements took the form of explaining to participants the way in which the examination would be video-taped and acknowledging the view that physical examinations of that part of the body are unpleasant and to be avoided if possible.

7.8

Thus, rather than interpreting this as a problem for the research we consider instead, the ways in which this additional dimension of the data - that is, the constitutive role of the researcher in the research 'product' and its ready availability as a research resource - may contribute to the understanding of how body exposure is managed in midwifery encounters. We demonstrate that the researcher's and the participants' acknowledgement of the taboo is consistent with our primary analysis which demonstrates that the perineal examination is managed in such a way as to acknowledge the special status of the genital area as private and sexual (Lomax, 1994).

Summary and Conclusion

8.1

The aim here has been to challenge commonly held and long standing views about the methodological value of audio-visual data. It has been seen how some of the existing literature suggests either that the presence of the camera and researcher has minimal or no effect on participants and research or that video methods are valid only if used covertly or in conjunction with other methods which reduce contamination of the research data. Our argument has sought to show not only that these views misunderstand the epistemological status of the research data produced, but also run the risk of blinding themselves to the advantages of videoing as a method.

8.2

From an analysis of our own data of situated researcher-participant interaction we have demonstrated that the video text-as-data is neither a representation of social life as it would have occurred had the researcher not been present, but neither is activity so contaminated by the research process as to make the data invalid. We have shown that the researcher is an active participant helping constitute recorded interaction, in this case of midwifery consultations. Similarly, the very process of recording interaction which is then preservable and available for everyday inference in the future, sets a context to which participants show an orientation. It could be claimed that such sensitivity on the part of participants is unsurprising in the case of genital examinations but our argument is that in any interaction the practise of recording helps organise 'what happens'.

8.3

However, the intention here has not simply been to suggest that the presence of a researcher and the process of videoing influences what occurs in any social interaction. Rather, the aim has been to point to the considerable methodological value presented by this approach as long as an awareness of the status of the data is maintained.

8.4

We have agreed with others that the activities of participants as research subjects and the means of obtaining data cannot be ignored. But, we have shown that by reflexively analysing these activities, it is possible to explore the way in which the research process - ie. the various participants and the means of recording their interaction - produces data. In contrast to the view that these activities are evidence for the problematic nature of video we have shown that such data may, in fact, provide additional insight in to the research question. So, a midwife's attempt to influence when the camera is turned on can, when reflexively analysed, provide insights into how midwives organise and differentiate between different parts of their professional duties, and in so doing, orient to and manage sexual taboos.

8.5

In conclusion, therefore, we accept that we cannot observe the world without being present in it. We accept also that video methods, far from being a means of neutrally reproducing social activity, create and define the event and are therefore fundamentally part of knowledge production. Further, we conclude that the capacity of video to record the occasioned activities of participants as research subjects means that the video camera is the research instrument par excellence, providing as it does an exclusive means of reflexively exploring the socially constitutive nature of the research process.

Acknowledgements

Technical and computing contribution: Paul Gowers, Institute for Health Services Research, University of Luton.

References

ALBRECHT, G. (1985) 'Videotape Safaris: Entering the Field with a Camera', Qualitative Sociology, vol. 8, no. 4, pp. 325 - 344.

ARBORELIUS, E. and TIMPKA, T. (1990) 'General Practitioners' Comments on Video Recorded Consultations as an Aid to Understanding the Doctor-Patient Relationship', Family Practice, vol. 7, no. 2, pp. 84 - 90.

BERGMANN, J (1990) 'On the Local Sensitivity of Conversation' in I. Markova and K. Foppa (editors) The Dynamics of Dialogue. New York: Harvester Wheatsheaf.

BERGSTROM, L.; ROBERTS, J.; SKILLMAN, L.; DEIDEL, J. (1992) '"You'll Feel Me Touching You Sweetie": Vaginal Examinations During the Second Stage of Labour', Birth, no. 19, pp. 10 - 18.

CARPENTER, L and MERKEL, T (1988) 'The Effects of Three Methods of Observation On Couples in Interactional Research', The American Journal of Family Therapy, vol. 16, no. 1.

EMERSON, J.P. (1970) 'Behaviour in Private Places: Sustaining Definitions of Reality in Gynaecological Examinations' in H.P. Dretzel (editor) Recent Sociology No. 2; Patterns of Communications Behaviour. Canada: Collier-Macmillan.

ERIKSON, F. and SCHULTZ, J. (1982) 'The Counsellor as Gatekeeper: Social Interaction in Interviews' in E. Hammel (editor) Language, Thought and Culture: Advances in the Study of Cognition. New York: Academic Press.

EIBL-EIBESFELDT, I. and HASS, H. (1974) 'Film Studies in Human Ethology', Current Anthropology, vol 8, no. 5, pp. 477 - 479.

FELD, S. and WILLIAMS, C. (1975) 'Towards a Researchable Film Language', Studies in the Anthropology of Visual Communication, pp. 25 - 32

FRANKEL, R. (1983) 'The Laying on of Hands: Aspects of the Organisation of Gaze, Touch and Talk in a Medical Encounter' in S. Fisher and A. Todd (editors) The Social Organisation of Doctor-Patient Communication. Washington D.C: The Centre for Applied Linguistics.

GOODWIN, C. (1981) Conversational Organisation: Interaction Between Speakers and Hearers. New York: Academic Press.

GOTTDIENER, M. (1979) Field Research and Video Tape. Sociological Inquiry, vol. 49, no. 4, pp. 59 - 66.

GRIMSHAW, A. (1982) Sound-Image Data Records for Research on Social Interaction: Some Questions Answered. Sociological Methods and Research, vol. 11, no. 2, pp. 121 - 144.

HAMMERSLEY, M. and ATKINSON, P. (1991) Ethnography: Principles in Practice. New York: Routledge.

HANSON, B. (1994) 'The Potential of Videotape Data: Emotional Correlates of Senile Dementia in Families as a Case in Point', Quality and Quantity, vol. 28, pp. 219 - 232.

HEATH, C (1986) Body Movement and Speech in Medical Interaction. Cambridge: University Press.

HEATH, C. and LUFF, P. (1993) 'Explicating Face- to-Face Interaction' in N. Gilbert (editor) Researching Social Life. London: Sage.

HEIDER, K (1976) Ethnographic Film. Austin (editor) Texas: University of Texas Press.

JEFFERSON, G. (1979) 'A technique for inviting laughter and its subsequent acceptance/declination' in G. Psathas (editor) Everyday Language: Studies in Ethnomethodology. New York: Irvington.

LIVINGSTONE, E. (1987) Making Sense of Ethnomethodology. London: Routledge and Kegan Paul.

LOMAX, H (1994) 'The Social Construction of Sexuality in Midwifery Postnatal Examinations: Some Preliminary Findings', British Sociological Association Annual Conference, University of Central Lancashire, 28th-31st March.

LOMAX, H. and ROBINSON, K (1996) 'Asymmetries in Interaction: An analysis of Midwife-Client Talk During the Postnatal Period', Proceedings of 24th Triennial Congress of International Confederation of Midwives. Oslo, Norway.

LUNDEVALL, S.; NJOLSTAD, I. and AARAAS, I (1994) 'Stop the Video and Involve the Observers: An Inter-Reflective Method to Stimulate Doctors' Learning about their Own Consultations with Patients', Medical Teacher, vol. 16, nos. 2/3, pp. 189 - 195.

MALLETT, J. (1990) 'Communication Between Nurses and Post-Anaesthetic Patients', Intensive Care Nursing, vol. 6, pp. 45 - 53.

MALLETT, J. (1993) 'Use of Humour and Laughter in Patient Care', British Journal of Nursing, vol. 2, no. 3, pp. 172 - 175.

MARTIN, E. and MARTIN, P. (1984) 'The Reactions of Patients to a Video Camera in the Consultation Room', Journal Royal College General Practitioners, vol. 34, pp. 607 - 610.

MURPHEY-BLACK, T. (1994) 'Care in the Community During the Postnatal Period' in S. Robinson and A.M. Thomson (editors) Midwives, Research and Childbirth. London, Chapman and Hall.

ROBINSON, K. (1987) The Social Construction Of Health Visiting. Unpublished PhD Thesis, Polytechnic of the South bank

RAGAN, S. (1990) 'Verbal Play and Multiple Goals in the Gynaecological Exam Interaction', Journal of Language and Social Psychology, vol. 9, nos. 1 - 2, pp. 67 - 84.

ROBSON, C. (1993) Real World Research. Oxford: Blackwell.

RIDER, A. (1984) 'Managment: A Manager's Approach to the Organisation of Normal Postnatal Care', Midwives Chronicle and Nursing Notes, October, pp. ii - iv.

SACKS, H. (1995) Lectures on Conversation. Cambridge, Massachusetts: Blackwell.

SCHEGLOFF, E. (1968) 'Sequencing in Conversational Openings', American Anthropologist, vol. 70, pp. 1075 - 95.

STARR, R. (1987) 'Clinical Judgements of Abuse- Proness Based on Parent-Child Interactions', Child Abuse and Neglect, vol. 11, pp. 87 - 92.

TEN-HAVE, P. (1991) 'Talk and Institution: A Reconsideration of the "Asymmetry" of Doctor-Patient Interaction' in D. Boden and D. Zimmerman (editors) Talk and Social Structure: Studies in Ethnomethodological Conversation Analysis. Polity Press: Cambridge.

VIHMAN, M. and GREENLEE, M. (1987) 'Individual Differences in Phonological Development: Ages One and Three Years', Journal of Speech and Hearing Research, vol. 30, pp. 503 - 521.

VUCHINICH, S. (1986) 'On Attenuation in Verbal Family Conflict', Social Psychology Quarterly, vol. 49, no. 4, pp. 281 - 293.

WEIMANN, J. (1981) 'Effects of Laboratory Videotaping Procedures on Selected Conversation Behaviours', Human Communication Research, vol. 7, no. 4, pp. 302 - 311.

Copyright Sociological Research Online, 1998