Vol. 9 No. 4, July 2004
The aim of the paper is to provide guidance on conducting participant and non-participant observation studies of information behaviour. Examines lessons learned during non-participant observation of hospital pharmacists, and participant observation with dependent older people living in their own homes. Describes the methods used in both studies, and discusses the ethical issues involved in gaining access to the subjects. In the hospital setting, professional affiliation between the researcher and the subjects (six pharmacists) made access easier to obtain. In the home care setting, access to subjects (seven clients) for participant observation (as a care worker) was more difficult, as was withdrawal from the field study. In both studies, the observation element was triangulated with survey data. Both studies indicated the fundamental need for trust between the observer and the research subjects. In some situations, professional relations offer instant access and trust, whereas in closed and sensitive situations such as social care, time is required to build up trust. With participant observation, that trust should not be damaged by withdrawal of the researcher from the research setting.
Participant and non-participant observation have been used to study the information needs of various groups of people, often in work settings. This paper compares experience gained in two different settings in the health and social care sector. The first study setting was in a hospital, working with a particular group of staff, pharmacists, and non-participant observation formed a part of a small-scale information needs study. The second was a qualitative, in-depth study of older adults where participant observation played a major part of research on the health and social information needs of dependent adults living at home. For many information science researchers, observation techniques seem to offer a more objective and direct view of information behaviour, as interviews may need to employ indirect means of assessing information needs and information use. Unlike the participant observation study, which had the benefits of considerable planning, duration and simultaneously delivered a service, the non-participant observation study had limited amounts of time and preparation. Nonetheless, the latter has valuable application for those conducting research within their working environment over a short time. It demonstrates the collection of rich and directly observed data for relatively low costs. Both studies share insights into the practicalities of observational fieldwork in both familiar and unfamiliar surroundings. The aim of the paper is to provide guidance for researchers considering the use of observation in information needs research, and to indicate the benefits, as well as some of the problems that may be encountered.
Comparatively few information needs studies have made use of observation techniques, and of these, most have used non-participant observation techniques. One of the earliest studies is that of Line (1971) who used observation along with questionnaires and interviews to study users needs of social scientists. Price (1984) used non-participant observation to study the operation of advice centres in Lambeth. Wilson and Streatfield (1977) used observation in Project INISS to study information behaviour in social services departments. Project INISS used structured observation, in selected departments, and the researcher spent a week with the research subject. Unusually for observation studies, inter-observer reliability could be considered in this study, and the researchers concluded that using a standardised recording format, and coding manual (with frequent discussion of problem issues), helped to increase reliability (see also, Wilson and Streatfield, 1980). Pettigrew (1999a, 1999b) used non-participant observation in an ethnographic approach to examining the role of community health nurses as 'agents of information'. Observations of nurses at work with senior citizens were followed by interviews with the nurses and the senior citizens (separately). Barker and Polson (1999) used observation initially, to determine suitable questionnaire categories. Eager and Oppenheim (1996) undertook a small-scale study of academics, and found that the research subjects found the presence of the observer rather unsettling, particularly when working at their desks for long periods, reflecting comments made by Wilson and Streatfield (1977) that in meetings the presence of the observer was less obtrusive than in periods of desk-based work. Lomax et al., (1999) used observation and interviews to complement data from a questionnaire survey of medical oncologists, and Byström (1997) used observation to complement the task diaries and interviews in a study of municipal administrators.
Observation was used as part of an information needs study of hospital pharmacists (Lewis, 2000). Other parts of the study involved a questionnaire and interview survey of hospital pharmacists. The aim of the observation was to gain a better understanding of the environment in which hospital pharmacists operate and their routine information needs. The study was based on an earlier observation study of pharmacists at work (Savage, 1996).
A purposive sampling approach was used, to select pharmacists performing different roles within the hospital, at different career stages, and with different levels of experience. In total, six pharmacists were observed, for part of their working day. Four of those pharmacists were later interviewed. A further sixty-five hospital pharmacists were sent a questionnaire. The researcher was employed in the hospital library at the time of the study and the research was undertaken for the dissertation element of a postgraduate degree in Health Information Management.
This study, conducted for a doctorate, involved several categories of participants: the clients (primary subjects receiving home care) and (as secondary subjects) the formal care workers and other allied professionals including care managers, trainers in care practice and health practitioners who have direct contact with the primary subjects. A dual, but overt role was assumed as the researcher was employed as a care worker. Although studies on the aged in Wales have been carried out during the 1990s, they were achieved by collecting data through interviews or focus groups (Wenger, et al., 2001; Shah, et al., 1993). A third key study of information needs of the elderly in England and Scotland by Tinker et al. (1993) was also conducted by interview. It appeared that no previous research in this area had adopted an 'insider' approach (de Laine, 2000: 107-108). By becoming a care worker it was hoped that the researcher would be able to see the reality of the everyday world from the position of the actors, and be able to interpret the symbols and meanings underpinning daily social interaction (Lee and Newby, 1989).
A purposive sampling strategy was adopted, with knowledge of the group used to select representative subjects, in this case elders receiving formal home care (Berg, 1995: 178-179). In order that a real-life picture would emerge, in-depth case study and observational approaches were adopted as methods. Kluckhohn (1940, cited in Rock 1979: 187) states that participant observation, 'is the conscious and systematic sharing, in so far as circumstances permit, in the life activities, and on occasions in the interests and affects of a group of persons' Nonetheless, the statement is conditional: 'as far as circumstances permit', and only through shared understanding built up over time can practical knowledge ('knowledge of') be learned by trial and error during daily personal encounters (Smith, 1998). Social interaction is the continuous interplay and interpretation of meaning by individuals in groups. To acquire such knowledge and to identify the personal interrelationships between the actors in a client context would require regular contact time. Health conditions or linguistic barriers may hinder communication, and in those situations, greater reliance is thus placed on symbols including body language, facial expressions, or gestures, as well as the interactions within the setting. In essence, such factors indicated the importance of an observational approach whilst studying the participants. Observation of the interrelationships of the clients and their care workers was complemented by in-depth interviews with the care workers, managers and policymakers. In total, seven clients were observed, and thirty-one care workers interviewed, several were interviewed more than once. Others interviewed were three care managers, one trainer, one director, two informal carers and a range of health and social care professionals (nine in total).
An inductive, grounded theory approach (Glaser and Strauss, 1967) was taken in the analysis of the transcripts and fieldnotes allowing the emergence of categories and themes from the data and the development of theory. Categories were refined and coding reviewed throughout the process for which NVivo software was used.
For the participant observation study there were complications concerning access to primary subjects in the initial phases. Alternative strategies were drawn upon to gain access and recruit additional subjects. Paradoxically, although access might seem the most difficult problem, withdrawal proved even more difficult. This had been a concern, though for different reasons, of the Local Research Ethics Committee (LREC) which reviewed the application. Their concern had been the protection of a client's continued care, although the researcher's team affiliation would guarantee clients' care after the researcher ceased care work.
For the non-participant observation study, there had been no requirement, at the time, to submit the research protocol for ethical approval as patients were not directly involved, and the researcher was a member of library staff at the hospital. Nowadays, ethics committee approval would be required for any research that involved observation of health service staff in their workplace, particularly if they were interacting with patients and other health professionals.
The problem of sustained access to subjects is a well-documented issue in classic ethnographic studies such as those recorded by Hammersley and Atkinson (1983) and de Laine (2000). Where the social care study differs from many other studies is that it transcends the boundaries of the public sphere and enters the domestic setting. Other studies in sensitive areas suggest that a possible mode of entry is to move from the position of an outsider to that of an insider (de Laine, 2000: 107-108), so becoming one who has legitimate access. In anticipation of acquiring ethics approval, and in the knowledge that finding subjects and gaining their acceptance would take time; registration as a self-employed care worker was secured with a well-established private homecare agency. From the outset the management of the agency as the major gatekeeper was made aware of the underlying intentions for the researcher's engagement. At all times it has to be remembered that to clients, co-care workers, and the agency the study was secondary to the researcher's function as a care worker. The agency was located in an area with the greatest density of elders in the population, and also the highest number of ethnic minority elders in Wales.
Whilst the role of care worker grants initial entry to the client's home, it does not guarantee continued access, or acceptance by the client or the existing care worker team. Equally, it does not assure complete and continued membership of the study by actors in each situation, but just one care worker and a single client (out of a population of forty-four subjects) rejected an invitation to participate. In any domestic setting, the care worker and in this case the researcher, must be mindful that in other people's homes they are constantly 'Walking on eggshells' (Apstitis, n.d.). And what were favourable circumstances and behaviours yesterday maybe taboo tomorrow.
For the non-participant observation study, the pharmacists granted permission for the study, on the understanding that they would be observed to obtain a better understanding of their information behaviour. They were not shown the observation sheet used to record, for each identified information need, the type of sources used and what was done with the information, in case this biased their normal behaviour. A variety of situations were encountered, including ward visits, attending meetings, and telephone requests. The researcher concerned in the non-participant observation study was a member of staff, which made permission for the study easier to grant, as the researcher was bound by staff codes of conduct on confidentiality.
For the participant observation study, primary participants were located through the researcher working as a member of care teams in various client situations. Work was obtained with an ethnic minority family and several white majority clients, and over a few months of weekly visits, trust and rapport were established between the researcher, clients, extended family members, and other care workers. It was anticipated that through the ethnic minority family the snowball sampling technique (Plant, 1975 and Mars, 1982 (cited in McNeill, 1990: 39)), could be used to locate contacts to other ethnic minority elders receiving care. However, after a number of visits to the family it transpired that the family was fundamentally atypical to this particular ethnic minority community, in that it accepted 'outsiders' into the home to provide care for the elder members. This phenomenon debunked the pervasive but controversial assumption maintained by local authorities that 'They look after their own, don't they?...' (Alibhai-Brown, 1998: 46 and Murray, et al., 1998). For the proposed research this discovery demanded that a more cross sectional recruitment approach be adopted, incorporating a larger proportion of elders from the mainstream population than was originally envisaged.
For the non-participant observation study, the main concern was to obtain a representation of various roles and settings in which the pharmacists worked. In an organisation the sample selection is far less problematic, although it may be too easy to rely on colleagues who are likely to be compliant.
The initial period of fieldwork for the participant observation took place over a period of several months, and only when it was considered by the researcher that trust and rapport were well established were clients and their families asked to join the study. Nonetheless, after several months of care work with one client, his relative declined the invitation to participate in the study and the researcher withdrew from the particular situation. Simultaneously, all care workers in the various situations were invited to participate in an informal interview. In the first phase of the fieldwork over twenty in-depth semi-structured interviews were conducted with care workers. Similar interviews were also held with all of the agency's care managers, training personnel and a number of allied health professionals. During the second phase of the fieldwork a more open recruitment strategy was executed by means of a leaflet campaign inviting other clients to join the study. Following negotiations with the agency it was decided that a voluntary arrangement with prospective participants would be most practicable. To accord with the strict confidentiality of the agency towards its clients, the flyer was enclosed within a mailshot of other information to elder clients and also to their care workers. Despite one director's pessimistic view of the likely success of recruitment through a mailshot, in practise the exercise proved fruitful in that several elder clients and additional care workers joined the study.
In each of the barriers discussed there is a common theme, which is that of time, namely the time taken to overcome hurdles and progress through each of the stages. Inextricably linked to this is the issue of ensuring that enough has been done. In a situation where information behaviours are spasmodic, how can one know that the current situation has been sufficiently saturated before contemplating the next move? Ethnographic texts suggest that this judgement is not bound by formal scientific rules, but rather that it is more of experience and intuition, i.e., you just know when it is time to move on. However, the management guru Charles Handy (1998: 14) suggests that, 'enough is never enough where your own personal standards are involved'. Inevitably, the temptation is to leave the doors open in order that situations may be re-visited.
In contrast, the non-participant observation study was eased by the collegiality of the work situation. Both the researcher and the pharmacists are 'information professionals' in the sense that information handling is an important part of their working lives, and they have responsibilities for providing accurate information to others. In such situations trust is easier to establish as many values concerning the importance of information management were shared. Nevertheless, it was difficult at first to convince the pharmacists that the information sources of interest were not just limited to books. The approach taken was designed to make the observation as natural as possible, and the researcher chatted to the pharmacist if that seemed an appropriate action, and asked some brief questions about an information seeking event if necessary. Some sensitivity to the situation was necessary, as the dispensary was often far busier and more demanding for the pharmacists than a ward visit.
To many actors in the various fields (client homes), it took a while for the research element to become 'wallpaper'. Nonetheless, some still regarded the researcher's dual role with suspicion, and statements such as 'you're the spy' were recorded. Continuous respect had to be paid to the issue of how observations could be recorded during the researcher's time in the field. As indicated, entrance to the field was acquired through the role of care worker, and it was agreed with the LREC, the agency and clients that no notes would be made at a client's home. Contact sessions with clients often run in sequence, and in practice this means constructing field notes several hours after being in the field. A strategy for remembering details was developed whereby each session was recalled using visual cues such as the actors, particular event(s), the physical context, and other props including mealtimes, taking a client to the bathroom, or putting them to bed and in some circumstances, the time of day facilitated recall of events. Field notes were structured around the guidelines provided by Spradley (1980, cited in Hammersley and Atkinson, 1983: 78). These include the spatial and temporal aspects; the goals, feelings, and activities of the actors; and the physical objects that are present in the situation.
The logistics of keeping the client satisfied, the agency on side, being accepted by 'the girls', and preserving the research function of the operation frequently demand significant emotional and physical reserves, and mental agility. There is a constant trade-off between fulfilling the daily requirements of care worker and satisfying the demands made by the care agency, and simultaneously keeping abreast of the needs of the research schedule.
Although the pharmacists often became accustomed to the shadow observer, their co-workers or visitors found the situation novel. However discrete the observer might wish to be, the temptation was for the co-workers to chat to the observer. While helping in some ways to make the observer part of the 'wallpaper' this makes detailed observation almost impossible.
Another temptation might be to observe more than two subjects at one time. This was tried on one occasion during a quiet time in the pharmacy, but inevitably both subjects started doing something of interest to the study almost simultaneously, which meant that observation of one subject had to be abandoned if details were to be recorded for the other. The observation sheet used in the study (based on Oppenheim, 1992) used a separate sheet for each information seeking event. Experience confirmed that a form needs to be piloted carefully, and that plenty of space needs to be allotted for 'other information'; data that does not fit neatly into the pre-coded categories.
Ideally, the observation periods should have been longer than a day, but time constraints on the study meant that longer periods were impossible. In addition, the environment in which the pharmacists were working was sometimes cramped, making it difficult to follow the pharmacists around discreetly, without bumping into them.
In the participant observation study, the need for information should arise from a deficiency in knowledge or 'knowledge gap' (Chatman and Pendleton, 1995; Ginman, 2000), about a particular aspect of social life. The researcher identified information needs ranging from the regular and temporal, 'what time, you? or, 'you, five?' (information requests from a client to establish what time the duty finishes), to the irregular and more complex - trying to identify an effective unguent to treat an aggressive skin ailment when all others stocked by the local pharmacy had been tried and failed. Information-seeking behaviours are context-sensitive (Chatman, 2000). Care workers often reported information-seeking in response to a client's health condition or social care need for which they had little experience or knowledge. The perceived need is frequently driven by attending a new client with that condition. Thus, to capture instances of information behaviours and needs may require considerable time spent observing primary subjects, or is dependent upon the retrospective identification of an occurrence by other informants. The former is often a matter of chance, in that an observer needs to be in the right place at the right time. Expressed need hinges on either verbal or non-verbal communication and the presence of someone who can interpret and respond to that need. In many circumstances, the researcher witnessed that subjects are more ready to express a need to those they think can do something about the problem. As suggested above, in many of the interviews it emerged that care workers and others did not consciously recognise information behaviours as such. In hindsight, several care workers' responses to a perceived need took the form of an automatic reaction, 'I just do it, I don't think about it'.
Aids to memory during interviewing were used including the critical incident technique described by Flanagan (1954). In the following extract from a transcript the interviewee, a care worker, reflects on events surrounding the delivery of client-related information to the agency, when information behaviour was entangled with questions around professional responsibility.
[Interviewer: Would you pass information back to the agency? If you became aware of a change in a client situation, or there was something you were concerned about, would you call your care manager?] ...Yes, I always do that... Yes, I've done that several times...
Initially, some interviewees appeared reticent to illustrate their responses with illustrations, but when prompted they often revealed rich data. This interviewee was a qualified nurse and shows her cautious, but comprehensive assessment of emergency situations:
...a client... Mrs MA1 had the stroke that... she'd already called the doctor when I arrived one morning..., and obviously, well I think she'd had a mini-stroke, her blood pressure was way up and she just wasn't well, her vision wasn't good, she had a headache
The interviewee also demonstrates her information behaviour in following-up the situation by passing the details to the appropriate people, though later she gave her perceptions of what care managers should do with the information. From over thirty years of experience as a hospital nurse she may have established ideas of the protocols of dealing with information following on from such situations.
..., then another situation was..., a client with muscular dystrophy, his breathing wasn't good, so..., I suggested that he called the doctor, which he did, and then each time I've made the care manager aware... [Interviewer: Have they [care managers] been responsive to that, when you've contacted them?]...Fairly responsive, yes but..., they always seem to think that..., they should pass it on to the family..., which in one way, yes I see their point but..., there has to be someone that makes a decision...Especially if you can't get hold of the family during the day... and as a, as a care worker you feel it's not your responsibility so you put the onus on the care manager...and I think sometimes that they could take things a little bit further.
In a second extract another care worker is prompted to revisit her information-seeking behaviour when identifying a service for a blind client, and what the client did with the information:
Well yes, just the last thing I looked up for him...Was... he wanted me to um look up restaurants, and... caterers...in the Yellow Pages...So that he could find out costs of, it's, it's for a celebration...And the costs of caterers...And so I had to ring up a number of... caterers...And find out what they were offering...
In the following statements the interviewee assesses her client's systematic responses to the information she obtained, but in the final stage of his decision-making process the client utilises a personal information source:
And their costs and he was... happy with the information that he got back... he was happy with having all that information because then he could work through all the information...And once he'd worked through it all he found that there were problems with each...And so he decided on something completely different...Which came from another... of the team...Who just said, why don't you try the Conservative Club? ...Which he hadn't even thought of...And she gave him ...
Without prompting the interviewee analyses the potential problems of commercial caterers and qualifies the client's desire to have a complete mental picture of the possible resources. Her end comment reflects the close knowledge she has of the client's preferences:
And, that seems to suit better than all the other uh ideas that he's had of caterers about you know...bringing stuff, it's, it's complicated, all the information that he asked me to give really complicated everything for him in the end...Because uh he couldn't find an easy solution, this was OK, but that wasn't, and that so, but... for this particular client he loves to have all the information anyway...
The following extracts from the observation field notes illustrate how needs are often tacit, part of the assumed responsibilities of the care assistants (CA) and that they remain unspoken, and unexpressed unless, for example, such needs are not immediately met. DJ1's daily life is driven by routine and ritual, consistent and accurate sequence of events being critical in maintaining a calm environment. These extracts from observation field notes demonstrate unexpressed need or want (but one of many that are very frequent expectations in this particular care setting):
Later on DJ1 wants to eat and at this time of day 6pm he has his main meal, each of the dishes that have been cooked that day are shown to him and then he indicates with his left hand the amount that he wants. During the previous evening he chooses what he wants cooked for the next day. This evening he has stuffed potato, boiled vegetables, chapatti, and lentil soup. The care worker mashes the food is to a pulp using a fork (this is always done on the table in front of DJ1), and mixed with water taken from the kettle, the pureed food is then micro-waved and returned to the table. DJ1 is able then to feed himself using only his left hand and a modified piece of cutlery. Towards the end of his meal he needs some assistance with feeding to gather up the remaining food on the plate. The fingers of his left hand are washed with water poured from a glass that is always placed on the table prior to his meal at this time of day. These needs are not verbally or physically expressed (unless DJ1 has to wait) but he (family members and experienced care assistants) expects that the care assistant will know when and what to do.
This extract illustrates the potential implications of missing the washing ritual:
CA3 moves to take his plate away and I remind her that he will want his left hand rinsed with water from his glass over the plate before she does this. She thanks me for this, as an error such as this can stimulate his unpleasantness.
However, this ritual and several others get out of sequence if there is something else attracting DJ1's attention as in the following extract where an altercation takes place between two family members:
Fortunately, he seems in a reasonable mood though becomes agitated when he is unable to make FJ5 understand what he wants, FJ5 loses her cool and starts to raise her voice, this disturbs FJ1 who comes to the rescue. FJ5 is rude to her sister who leaves the room. CA3 points out to FJ5 that she has to leave at 8pm and we need to get him changed and ready for the night. This we do and he returns to the kitchen, at this point DJ1 normally has hot milk, which I begin to prepare. FJ5 disappears with her laptop to the sitting room. FJ1 is eating her meal and her father wants to have the same, the usual need for milk is forgotten and he embarks on eating a small plate of food. Paradoxically, he does not miss the customary glass of water that he uses to wash his fingers with until I have taken his plate away. The plate is returned to the table, and symbolically he washes the fingers of his left hand as I pour the water over them.
Two of the themes that emerge from comparison of the interviews and the field notes are the importance of anticipation of wants or needs for care, and timing of the response. Providing information, whether to the care manager or to the client, is viewed as part of the process of care, whether social care for the client, keeping them contented, or clinical care needs that the care manager should know about.
Despite the limited number of days that could be allocated to observation of the pharmacists, the study revealed a rich variety of information-seeking behaviour and range of sources used. The ward pharmacists, for example, spent much of their time on the ward communicating with patients and the other health professionals, listening as well as offering advice. In the hospital drug information centre, the pharmacists used their own resources, their 'frequently asked questions', recorded on an internal database, other external databases (Medline, Martindale, Micromedex, British National Formulary, Pharmline, Cochrane Library) as well as the relevant specialist drug information centre. Communication with patients was a key role in the eye centre, with the pharmacists supplying large print patient information leaflets on administration of eye drops, for example, supplementing information provided verbally. Another pharmacist used the services of the professional association (Pharmaceutical Association Library) for information about drugs produced abroad.
Observation helped to identify some of the problem areas for pharmacists, such as:
Careful management of drugs is important and senior pharmacists are involved in benchmarking of prescribing patterns. The questionnaire survey explored the frequency of some of the information problems noted in the observation study, and some of the perceptions pharmacists had about their role in providing information and advice. The questionnaire survey confirmed that pharmacists developed their own routines, but that most would rely on a few key sources in their workplace, ask a colleague for further advice, and would refer the difficult queries to the specialist drug information centre staff. Interviews provided the opportunity to explore perceptions of the role more closely, and four of those who had participated in the observation or questionnaire survey were approached. Interviews illuminated how much of the information gathering and monitoring is done to pre-empt problems, whether errors in prescription or a more expensive drug being prescribed when a cheaper alternative would be equally effective. Establishment of trust is important: 'Pharmacists need to be pro-active and give information to patients and other staff - it is an important part of establishing and maintaining relationship'.
Relationships with colleagues are also important, and the Internet provides a means for pharmacists working in the community to communicate with each other. Even in hospitals, pharmacists need to take time to share knowledge with each other:
'Although we work on the wards alone we do meet up twice a week as a group to discuss things. The first meeting is largely management and financial issues, and the second is to do with clinical issues combined with a sort of teaching session in which we take it in turns to do presentations. It is really helpful to share what we have learnt...or to express new ideas in this group environment'
In participant observation over extended periods of time spent with vulnerable subjects and their families, there is a perpetual need for a delicate balancing act. Establishing the vital relationship of trust with the actors in very personal physical circumstances leads to dependency, in the first instance by the client, who displays visible signs of pleasure when a preferred care worker arrives to start a duty. The care manager responsible for the particular client also becomes dependent upon workers who are accepted by the client and identified by managers as 'able to cope alone' in highly sensitive situations. Competent care workers are frequently called upon to plug gaps in care duties (emergency requests were legion during the course of the fieldwork), and are also relied upon as trainers of new recruits. In turn, novices and those with less confidence in coping with clients are dependent upon those who are accepted in a given client situation. For the researcher these factors lead to difficulty in withdrawing from such situations where, paradoxically, diligent efforts to gain acceptance and access resulted in client and team dependency. Simultaneously, as a care worker the researcher felt bound to demonstrate commitment to the client and the agency, and inevitably the latter showed hostility when withdrawal from a client situation was effected.
One theme which emerges from comparison of both the studies described is that of trust. Trust has been defined as a, 'psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior of another' Rousseau et al., (1998). In her study of trust in a conceptual R&D organisation Sonnenwald (2003) examined affective and cognitive trust. The former relates to 'interpersonal bonds among individuals and institutions', whilst the latter concentrates on 'judgements' about a person's ability to execute a task effectively (Sonnenwald, 2003: 3). Both have relevance to the contexts, which inform this paper. In the non-participant study access to the situation was eased by the fact that the researcher and the pharmacists not only worked for the same organisation, they shared common interests and concerns in information handling and information provision. The finding that the pharmacists themselves viewed the sharing of information as important in establishing and maintaining relationships with colleagues and patients was fortuitous as it no doubt made the research easier to undertake. Wilson and Streatfield (1977) describe the fairly lengthy procedures of familiarisation in Project INISS to ensure that both researchers and research subjects were aware of the interests of each side in the research process. Pettigrew (1999a, 1999b) describes some of the strategies used to decrease the effect of the observer on the participants. Her approach, in combining observation with interviews with both nurses and senior citizens attending the clinic may have had the effect of consolidating relationships, and increasing trust.
In the participant observation study in the social care setting, cognitive and affective trust had to be developed from the outset, with gatekeepers to the research subjects, with the clients' families, the clients themselves, and with the other care workers. Unlike the non-participant observation study, there were few common interests in information provision between the research subjects and the researcher, as many of the care workers, like the nurses in Pettigrew's studies, did not view information in the same discrete way as an information professional might. Information sharing has to be 'grown' and nurtured and trust develops as a result, but for such workers the emphasis is less on the information and more on the process of building relationships. O'Neill (2002: 76) points out that:
'We place and refuse trust...because we can trace specific bits of information and specific undertakings to particular sources on whose veracity and reliability we can run some checks. Well-placed trust grows out of active inquiry rather than blind acceptance'.
Interestingly, some of the behaviour recorded in the non-participant observation studies accord with this process of active inquiry, in which the research subject or their visitors check out the observer with a few gentle enquiries.
For studies of the information behaviour it is clearly an advantage to work with groups or individuals where some values of information management are shared, and there are common professional concerns. The research topic is easier to justify and explain, access should be easier to negotiate and trust should be faster to develop. For situations where information is something that is negotiated, and not a conscious 'product', the research needs to be justified in ways that make sense to those in the situation. Non-participant observation may need to be accompanied with a suite of strategies to ensure that the researcher is increasingly accepted and trusted, and this will mean that careful thought is required on how to respond to enquiries which are 'checking out' the observer. In some research situations, participant observation is the only feasible way of conducting the observation research. This in effect limits such research to doctoral research in most circumstances, given the time that must be allocated to the negotiation of access, and establishment of trust. The main benefit is to obtain a much more realistic view of what happens, although the interpretation is subjective. Paradoxically, as much thought will need to go into the ways of withdrawal from the research situation so that the researcher can leave the research situation, without compromising trust placed in them by the research subjects.
In situations where the information professional is part of a work team, rather than part of a separate 'information service' observation may become part of the job, as empathy and awareness of the work situation are necessary to provide a service to the rest of the team. Skills in observation techniques for conducting studies of information behaviour move from the research setting to practice. The process of building trust, with negotiation and sharing of information, and sensing of unexpressed information needs are research skills that emerge from the studies described in this paper, but these skills are required by information professionals such as clinical librarians. Skills in non-participant or participant observation may therefore deserve greater attention in research methods training and professional development than they may receive at present.
|Find other papers on this subject.||
© the authors, 2004.
Last updated: 21 March, 2004