Early Career Learning at Work (LiNEA) Project

Methodology and Theoretical Frameworks

Stephen Steadman

SussexUniversitySchool of Education

A paper for the

Symposium on Early Career Professional Learning

AERA Conference, Montreal, April 15th 2005

Research funded by

The Teaching and Learning Research Programme

of the

UK Economic and Social Research Council

Introduction

The Early Career Learning at Work (LiNEA) Project is a longitudinal study of the learning of newly qualified nurses, graduate engineers and trainee chartered accountants during their first three years of employment. The accountants and engineers are formally contracted trainees, for whom employers have developed systems of organised training support. The engineers start with relevant degrees, e.g. in engineering or computer science, and most are seeking the advanced meta-qualification of Chartered Engineer. In contrast, the accountants’ degrees are rarely in relevant subjects and all receive formal out-sourced training to prepare them for the professional examinations of the Institute of Chartered Accountants. The nurses provide another contrast. Their initial training programme allocates 50% of its time to working in practice settings and concludes with a professional qualification. Thereafter, newly qualified nurses are expected to take advantage of a variety of in-service training opportunities to broaden and update their knowledge and skills, but communication failures and work pressures often mean these options are not fully used.

The project has had to devise a methodology to address the problems of accessing hard information on what people need to know at work when most learning at work is informal and therefore unlikely to be readily acknowledged or scarcely remembered without some pertinent prompting. Even formal learning on courses or in workshops, although readily recognised as learning, may be forgotten after a while.

Previous research has shown that important components of what is learnt are often carried as tacit knowledge and skills that are drawn on in working situations to address problems without the user being conscious of their utility. Eraut (2000) has defined the ‘personal knowledge’ that workers use as what individual persons bring to situations that enables them to think, interact and perform. This definition includes non-codified, or unwritten, personal knowledge and a far broader concept of knowledge than that implied in academic performance. For example it includes, not only personalised versions of public, formal, codified knowledge, but also everyday knowledge of people and situations, know-how in the form of skills and practices, memories of episodes and events, self-knowledge, attitudes and emotions. It focuses on the value of knowledge in use. Unless one stops to deliberate, this kind of knowledge is usually available in an integrated form and ready for action. Much of this personal knowledge is, of course, taken for granted and often difficult to explain – even to a researcher.

Thus the first part of this paper describes the methodology chosen by the project to address these, and other associated, problems of accessing what people learn at work. The second part of the paper sets out the theoretical frameworks that have emerged from the project’s work so far. Other papers in this symposium illustrate the frameworks in relation to the three professional groups under study.

Part 1The project’s methodology

A brief overview of the research design

The project has combined visits, observation and interviews, with managers and mentors as well as the new employees themselves, over three years to gather its data. The resulting data set comprises field notes on observations in the workplace, and transcripts of interviews with learners together with transcripts of interviews with their managers, mentors and colleagues. We also have access to a variety of documentary sources such as work logs, training programmes, etc. We are still analysing the fourth and final visits of our three year longitudinal study.

The central research questions have been:

  • What is being learned?
  • How is it being learned?
  • What are the main factors affecting this learning in the workplace?

These questions were originally developed for a previous project on mid-career learning in the workplace (Eraut et al. 1998a and 1998b, and 2000). That research collected its data from two interviews held some months apart. Key changes in the design of the present project have been the addition of a period of one to two days observation prior to interviewing, repetition of the observations and interviews over a far longer period, and the addition of data from interviews with managers, mentors and other work colleagues. These changes in the research design reflect the attention that was given to the methodological problems of eliciting tacit knowledge. For example, the observations not only provide the evidence of the field-notes, they also enable us to use workplace documents and activities as starting points for conversations about embedded knowledge and its acquisition that would otherwise have been impossible. This alone was an important development, and further justification for the incorporation of observation is provided below, but other challenges remained.

The scale of the project, both in its scope and its time span, brought problems. The table below summarises the number of informants from whom data has been collected for the longitudinal study:

Table 1:Numbers of informants

Nursing
/
Engineering
/
Accountancy

First Visit

Main subjects
Managers etc. / 40
36 / 34
92 / 18
14

Second Visit

Main subjects
Managers etc. / 30
2 / 34
38 / 14
2

Third Visit

Main subjects
Managers etc. / 20
2 / 34
40 / 12
2

Fourth Visit

Main subjects
Managers etc. / 13
0 / 24
29 / 10
4

Practical problems are bound to arise when seeking to observe and interview engineers liable to be work station-centred or on site, accountants on client’s premises and client’s time, and nurses who work shifting patterns of shifts. Yet we needed to maintain relationships with organisations and our informants over several years while personnel changed and some of our informants moved to other jobs. The nature of the changing relationships themselves would have an effect on the way observations and interviews were conducted. The focus of the interviews had to change with time, partly because background became filled in, and partly because we needed to collect adequate samples of the learning domain. It was also important to balance the competing needs of consistency in the application of our procedures across the whole project against the need for sensitivity to sector differences. No small part of this has been the challenge of developing a reliable common system of coding and cross-checking for different researchers working in different sectors. In fact, the whole business of data handling, reduction and analysis has been crucial in a project designed to amass very large quantities of qualitative data.

Observations – intentions versus reality

The place of observation in the research design is of key importance because of the known problems in accessing informal and tacit knowledge referred to above. They may be summarised as follows.

  • only knowledge acquired in formal educational/training settings is easily brought to mind, articulated and discussed;
  • tacit, personal knowledge and the skills essential for performance at work tend to be taken for granted and omitted from accounts;
  • most important workplace tasks and problems often require an integrated use of several different kinds of knowledge, and the integration of the components is itself a tacit process.

The visits as planned in the project proposal, were supposed to take the following form. Day one, go along for an extended period of observation, stay out of the way and keep quiet, except at breaks and lunch time, but talk to the managers and mentors as you get the chance to do so. The next morning continue observing, and in the afternoon do the interview with the advantages of being able to refer to observed incidents and processes where the informant has either been learning or appears to have drawn on previously learnt knowledge, skills or understandings.

Despite its practical problems, observation brings many advantages. Some examples are that observation:

  • educates the observer/interviewer about the working context, necessary tasks and priorities, and thus enriches subsequent data gathering;
  • provides ‘clues’ to the use of knowledge that must have been previously learnt.;
  • allows complexity to be appreciated;
  • enables comparisons that suggest questions; and
  • discourages the painting of ‘ideal pictures’ by informants when they know reality has been observed;

A main advantage is that, with observation as a starting point, an interview becomes a discourse of description, rather than a discourse of justification that can so easily result from asking what an informant has had to learn in order to do the job. Moreover, many benefits accrue from combining observation and interview data. The meaning of observation data alone is not always explicit. What should an observer infer from seeing a trainee accountant walk round an organisation, following the path of an invoice through the stages of its handling? But being able to discuss the events with the trainee, ask what was going on and what the trainee found out helps to decide the correct inference and brings out what was being learnt.

The hand-over on a hospital ward to a new shift team is one instance where a rich mix of information has to be handled. The new team has to know about patients, their conditions, treatments given and their effects, necessary monitoring and possible changes, links to doctors, relatives and other agencies as patients are to be moved or discharged, and so on. Ephemeral documentation is used as both a reminder for the hand-over, and a reference point for those being briefed. A hand-over may take up to an hour. No wonder then that newly qualified nurses find these events full of calls on their ability to learn – and to learn quickly if their credibility with colleagues is not to suffer. It is not simply the need to understand the medical content of what is being relayed, often embedded in a stream of acronyms and abbreviated terms – the working jargon of a professional: important implications have to be appreciated and priorities for attention decided. They also have to learn – through their own observations and judgements – how best to conduct a hand-over when it is their turn to do so. The newly qualified find these events stressful, and observing such events helps a researcher to understand why.

Two examples, abbreviated and adapted from field notes, illustrate something of the range of situations observed with engineers.

Example 1.

Recently I was with a civil engineer, working on an upgrading of water main systems. On the first day she insisted that I accompany her to a meeting of the agents who are responsible for digging up the pipes and for the work of renewing or replacing them. She then took me to two places where this was going on, to show me what the pipes look like and what can be done to the insides of these pipes to keep them in use before the most expensive option of excavation and complete replacement has to be faced. Without that observation I would have been unable later on, during the visit, to fully understand what she was telling me about, let alone anything else, even before I could begin opening up the areas of learning that she’s been engaged in.

Example 2.

I was observing a mechanical engineer who was working at her desk while I was sitting, because of restricted space, with my shoulder jammed up against her small bookcase. After a while I spotted a manual on a particular software program that I’d never heard of before. So I asked her about it, not in the formal interview, but at another time of the day. She said, ‘Oh yes, that’s something I’m working on.’ And when we got to the interview itself, it turned out that she had been picked by senior people in the firm to find out how that piece of software could be used to forecast risk assessments and allied costs. The senior management had picked on her because, in their eyes, a young graduate was expected to be well up in IT skills.

In total these examples show how observation educates the observer/interviewer, prompts them to suspect the deployment of previously acquired knowledge, skills or understandings, provides ‘clues’ to things learnt, and reveals learning opportunities that a trainee may fail to mention. In relation to Example 1, the graduate engineer had herself found it essential to make the same field visits in order to understand the implications for the work gangs on the ground of what she was planning back in the office.

Observation thus enriches subsequent data gathering. Many other examples could equally be given to illustrate the kinds of interactions and opportunities that have proved the wisdom of having observations as an integral part of the research methodology. But, as noted above, observation often runs into practical problems that have to be negotiated. In the nursing sector our researchers have found that shift patterns determine how long observations can last. So, although we intended to observe for a day and a half before an interview, the natural period of observation is often not that. In these circumstances we have to go with the flow, and rightly so. We have also found that nurses are so busy, that finding the time to do the interview is a problem in itself. In fact they are so busy that, by the end of their shift, they’ve often forgotten what they were doing at the start of the shift. This is a problem which previous researchers have noted. Being able to refresh a nurse’s memory from the observer’s field notes, or by other means such as the digital photographs that Christine Fessey (2002a, b and c) used in her very rigorous work, is absolutely necessary when this happens.

In the other two sectors rather different problems arise with observations. Trainee accountants can spend significant parts of their working lives on the premises of clients. Although this could have raised significant problems, in practice, by allowing the accountants to negotiate access on our behalf, this has not proved to be the case. Most clients have accepted that it is the trainees who are being observed, not themselves, and we have been careful in our positioning and other actions and conversations on site to make it clear that this is the case.

Nowadays many engineers and accountants, spend prolonged periods in front of computer screens, much more time than we had ever expected when planning this research. So an observation period of up to six hours could be spent looking at the back of someone’s head. Viewing a computer screen from further than three feet away is itself a challenge. So we negotiate with the trainee and manager what goes on in those circumstances and, if it’s not profitable to prolong an observation period, we negotiate it down, or we rearrange how we spend our time. In all sectors, being an observer brings problems of where one should place oneself to be able to see, yet be out of the way and definitely not in the line of sight of the person being observed. This has proved awkward at times but we have not found this an insuperable problem.

Observation of an individual over many hours has many of the characteristics of ‘shadowing’ and there is the danger of becoming an oppressive presence. Presumably, this was felt by one trainee accountant who was sufficiently disconcerted by the first period of observation to wish to drop out of the project. However, subsequent negotiations with the trainee, and their training manager, found a compromise. The trainee agreed to be interviewed on each round of field visits, but not observed. Clearly, we agreed to this arrangement because we did not wish to lose all the data from this individual, and we hoped the decision would eventually be reversed when the trainee could see how the project was progressing. But we also agreed because we were still able to observe the other trainee accountants and use insights from that set of data when interviewing the one person who objects to being observed. Thus the arrangement does not completely undermine our position on the importance of using data from both observations and interviews.

Very few others among our 90 trainees have shown signs of this being a problem. However, we have found that, as nurses move into more senior positions, they are less happy about being observed and less happy about taking time out for an interview, even when the time requested is only 30 minutes. Oddly enough, in one engineering firm, it was someone totally unconnected with the project, and to this day not known to us, who found the presence of our observer unsettling and complained to the management. They quickly resolved the issue by explaining fully what was going on. However, particularly when trainees are working with computers, but generally too, we have found it helps to use tea and coffee and other breaks to raise some of the things we think we may have noticed. This keeps us on track, makes sure we understand what the trainee is working on, and gives opportunities for more general conversation that often identifies topics worth following up in the formal interview.