Introduction

1.1 The pharmaceutical sales industry

The pharmaceutical sales industry has been minimally examined as a case of workplace learning. It is a case of interest in a time of increasing popular reference to the knowledge economy because it is a highly competitive and rapidly changing knowledge-intensive industry. It also employs a geographically distributed workforce. [9] Marsick et al. (2006) indicated that more research is needed to understand the impact of distributed work arrangements on workplace learning. This study describes the ways in which sales agents in the pharmaceutical sales industry learn for work. It specifically documents attributes of (in)formality and other more elemental characteristics of ways of learning for work that participants reported as effective or frequently used. In doing so, it provides a description of learning in specific industry contexts that might be compared to other contexts.

1.2 Attributes of (in)formality

Much research has classified cases of workplace learning as either formal or informal. However, there is no consensus defining these terms. Readers are left with inconsistent, and sometimes conflicting definitions of formal and informal learning, and no clear understanding of how such labelling could support workplace learning. [3] Colley et al. (2003 p. 1) reviewed over 300 titles in the adult learning literature concluding that there is a "complete lack of agreement about what constitutes informal, non-formal and formal learning, or what the boundaries between them might be". They reviewed the approaches used by authors and developed a list of 20 different criteria used to define (in)formality. They applied these criteria against a range of learning situations concluding that aspects of formality and informality exist in most if not all learning situations. They recommended abandoning the dualist vision of formal/informal. Others have proposed limited value in documenting (in)formality in workplace learning ([1] Billet, 2002; [6] Gerber, 2006). Instead, Colley et al. recommended reporting the defining criteria as attributes of (in)formality divided into four clusters:

process;

purpose;

location or setting; and

content.

[3] Colley et al. (2003) provided brief and general descriptions of the properties of each cluster. Process attributes of (in)formality include characteristics of learning associated with the planning, mediation, and pedagogy of learning. Purpose attributes address whether learning is the primary reason for the activity and whose needs the activity is meant to address. Attributes concerning the location and contexts in which learning occurs, are referred to as setting attributes. Content attributes focus on the nature of what is being learned and the outcomes expected. Colley et al. did not explain what they meant by each of their individual attributes, nor did they assign attributes to clusters. They gave indistinct descriptions to guide interpretation and application of attributes. A different researcher may interpret their attributes differently and may assign them to clusters differently. Table I [Figure omitted. See Article Image.] presents Colley et al. 's attributes of (in)formality as I have assigned them based on my interpretation of their descriptions.

2 Method

2.1 Delphi-collaboration: developing a comprehensive list of ways of learning

Two data-collection strategies were used in this study. First, a modified Delphi-style technique was used for developing a comprehensive list of the ways in which Canadian pharmaceutical sales representatives learn for work. [8] Linstone and Turoff (2002) declined to define Delphi as a describable model since its purposes, philosophies and methods vary. They described it as a varying set of ways of structuring communication within a group involving a degree of anonymity, inter-participant feedback, and participant-driven revision. The researcher assumed that some ways of learning engaged in by-agents might not be immediately and explicitly available to participants. It was decided that a multi-stage data-collection strategy which allowed agents to view the responses of other agents might stimulate ideas and facilitate the development of a more comprehensive list. Time between interactions would allow participants an opportunity for exploratory self-observations and reflection. These considerations, the distribution of participants across Canada, and the competitive nature of the industry meant that an e-mail Delphi collaboration would be appropriate.

An e-mail recruitment posting was sent to the Canadian Council for Continuing Pharmaceutical Education's virtual sounding board. In total, 20 participants volunteered their participation and were accepted. They remained mutually anonymous and were tasked with developing and organizing a comprehensive list of ways they learn for work. The collaboration involved three successive stages of e-mail correspondence. The researcher served as a conduit for this process and organized, structured and distributed the participant submissions. Each was asked to submit a detailed list or description of all of the ways they could identify that they use to learn for work and were given at least a week to reflect and respond. The researcher collected their responses, reduced the data by grouping replicated or similar responses, and classified the resulting data by applying classification themes supplied by participants. The resulting summary was sent to all participants with a request to review, comment and make further additions, alterations, and suggestions. At least, a week was given for responses. The process was repeated until a third summary prompted no further revisions. The third summary was approved by all participants.

2.2 Individual interviews: identifying the perceived most frequent and most effective ways

The second data-collection strategy was primarily aimed at identifying the most frequent and most effective ways of learning as perceived by participants. Participants for individual in-depth, guided interviews were convenience sampled. Through word of mouth, sales representatives in Eastern Ontario became aware that research into learning in this industry was being conducted. The first five agents to volunteer their participation were interviewed and were judged not atypical of other representatives based on the researcher's six years of experience in the pharmaceutical industry. One case was excluded because of poor interview recording quality. Interview participants did not participate in the Delphi exercise. The cases analyzed included three male and one female agent from four different companies. The mean experience was six and one half years, with a range of a few months to 12 years. Ages ranged from 32 to 45 years. All participants were married and three had young children. Interviews ranged from 22- to 73-minutes in length and were digitally recorded.

Data describing the work contexts in which learning occurs in this industry were gathered both from the interview participants and from the researcher's personal experiences over six years in the pharmaceutical sales industry. Interview participants were asked to describe their work contexts during their individual interviews. Subsequently, the researcher compiled a description of the work contexts based on interview data and personal experience. The interview participants reviewed and verify the descriptions.

Interview participants were also asked to review the document created by the Delphi collaboration and comment on its accuracy and comprehensiveness. Each participant was asked to discuss their philosophy concerning learning in this industry, the context of the industry, and to provide detailed descriptions concerning the ways of learning for work that they found most effective and most frequently employed. Each was asked to identify and describe in detail one or two ways of learning that they perceive to be most effective and one or two ways that they perceive that they engage in most frequently. Some of the probes included:

- What factors lead to you learning in that way?

- What circumstances lead to you undertaking this activity?

- Who else was involved in this learning and what was their role?

- Describe the location and setting.

- Tell me about the planning you did in advance of this activity.

- How did you feel about the process?

The researcher created a summary of each case and asked each participant to review and comment on its accuracy. One revision was made to one of the summaries and then all were approved by the participants. Interviews were transcribed verbatim and open coded by the researcher. The [3] Colley et al. (2003, p. 28) "attributes of formality" were interpreted and applied to data segments which described ways of learning perceived by participants as most effective or most frequently used. Subsequently, the Colley et al. attributes were extended and reworked into a framing guide (see Section 3.5 and [7] Hunter, 2009) to aid identification of many fundamental characteristics for each of these reported frequent and effective ways of learning.

3 Findings

3.1 Work context

The primary responsibility of sales agents is to promote prescription products to health care professionals (doctors, pharmacists, and nurses) using clinical studies and other data. Since customer practitioners are busy and often sceptical of sales agents, each interview participant reported that it has been difficult to gain access to and the trust of the prescribers. John reported "What differentiates one sales person from another is definitely knowledge." Lewis said:

Your success is dictated by getting access to the doctors [...] The real success of a pharmaceutical [agent] is getting the opportunity to get in front of a customer. And the way you get in front of a customer is by differentiating yourself from somebody else.

Each participant reported that their work activities vary greatly each day and they often spend much time alone and work long hours. Theirs is a knowledge-intense job where new studies, competitors, and practice guidelines are released daily. There is intense competition, pressure to increase sales, and a perception that ongoing learning is critical to business objectives. Reports Lewis: "You're the guy that people will look to as a resource." Steve indicated that he needs to be of value to his customers and he "still need[s] to improve that value constantly."

Participants reported a special role for learning in this industry. Unlike a more traditional role of learning where what is learned is applied directly to the job, agents reported that they often learned knowledge not directly needed to promote their products. Such knowledge may be in a different therapeutic area than the product (s)he promotes: an agent promoting cardiac medicine might learn the criteria for having a diabetes drug covered by a public insurance program. An agent may also choose to learn information concerning medical services not directly related to the drugs that (s)he promotes: An agent promoting oncology drugs might learn which pharmacies have diabetes educators on staff. An agent might also seek out new knowledge unrelated to medicine but related to other interests of the prescriber, such as properties of fine wines, classical music, or local sports teams. Lewis claimed that he differentiates himself:

[...] by doing something different or doing something more than the next person. And what I've always tried to do is to stay ahead of the group as far as knowledge.

John chose to educate himself:

[...] on a lot of the extras beyond the sales skills, beyond your own product skills [...] The right skills to get your job done combined with a lot of background knowledge [...] will give you credibility.

Agents in this industry used learning to develop themselves as resources for their customers in order to earn their trust and respect and gain access to customers in a challenging environment.

3.2 Ways of learning for work

In total, 20 agents from 11 different companies across Canada participated in the Delphi collaboration. Over a two-month period, the researcher received almost 100 e-mails from participants, ranging from short lists and clarifications to very detailed submissions. One agent submitted a 29-item list including self-directed and externally designed approaches; planned and unplanned learning; and collective, collaborative, and individual methods. Another cited 17 items including courses, internet and print resources, and even his children. Another offered a 500-word narrative describing various ways of learning giving special attention to customer-need driven research:

[...] challenging doctors or pharmacists. People who asked me real questions or presented me with scenarios that made me research further for information [...] Doctors can ask about medication studies, rare side-effects, efficacy in specific populations.

Another included a similar experience:

I have a close family relationship with a pharmacist who is also a Long Term Care Consultant for meds in Long Term Care facilities [...] I can discuss various disease states, outside of my world, as well as current and pressing issues that our health care system is facing.

The Delphi-collaborators identified 64 different ways in which they learned for work ([7] Hunter, 2009). They were not tasked with identifying the frequency or effectiveness of the ways of learning; they were only asked to create a comprehensive list. The ways of learning were organized into six categories by the collaborators: corporate-organized activities, externally organized activities, peer-based learning, web-based initiatives, learning on the job, and self-initiated and independent ways of learning. Most corporate-organized activities focused on skills and included: working in tandem with a colleague or a manager (what the industry calls a "work-with"); and soft-skills workshops such as time management. Very few externally organized activities (such as university, college, or industry courses) were reported and were only mentioned by a few participants. In total of 12 agents reported learning through peers despite geographical isolation. These included casual conversations at periodic national meetings, and sharing experiences through e-mail and telephone systems. Seven agents reported web-based learning, for example, through subscriptions to online medical newsgroups, medical education programs, and health agency web sites. All reported learning on the job and self-initiated and independent ways of learning. Some other specific ways of learning included: review of product monographs, speaker slide kits, medical guidelines, and patient health-care groups; self-analysis of clinical trials; having discussions with family and friends; taking chances and making mistakes; and reflecting on customer needs. For further details, see [7] Hunter (2009).

3.3 Perceived most effective and most frequent ways of learning

Interview participants identified six ways of learning that they perceived to be most effective for gaining new knowledge for work: multi-media presentations, customer conversations, medical rounds, remote sharing with a peer network, peer sharing during periodic national meetings, and self-directed review of periodicals.

Multimedia presentations included presentation of information, live or remote, which involved more than one medium or technique for delivery, such as conference presentations involving slides or web-based audio/visual continuing medical education programs for health care professionals. Medical rounds are a form of multimedia presentation which was mentioned specifically as a particularly frequent and effective way to learn, so it is presented here separately. Medical rounds are regular presentations, usually made by senior health care professionals to junior professionals in the hospital setting. These may involve the presentation of new clinical studies, guidelines, or case studies to illustrate a particular aspect of medical practice. Not all agents have access to rounds. Those agents who specialize in hospital settings and whose companies provide unrestricted educational grants to the hospital are often invited to attend rounds.

Agents reported working in geographic isolation, yet peer-facilitated learning was perceived as particularly effective. Peer-facilitated learning was reported in two forms: remote peer sharing through telephone and e-mail, and semi-annual to quarterly sales force meetings. Meetings were reported to last between two and five days. Agents participated in activities whose purposes were not necessarily educational, but formal workshops were occasionally included. Agents reported valuing unstructured and casual times during meetings as opportunities to learn from peers. Alice reported:

The most valuable learning is when I'm sitting down with peers who do exactly my job in other parts of Ontario. Sure I learn by reading my clinical papers, but I can't put out what I've learned to a customer well until I [...] hear somebody else's point of view; hear what they pulled out of the paper; share my point of view; figure it out between us.

Steve reported that "the information and feedback that I would receive from mentors [...] enabled me to be more successful."

The conversations agents have with customers during sales promotion were reported as an effective way to learn. These conversations focus directly on the concerns of patients and care givers and often provide new information to the agent or serve to highlight concerns that a customer has that an agent can subsequently investigate. John said that input from customers was one reason he "looked very successful and I actually came off as kid genius sometimes."

Only three ways of learning were reported as most frequently used. Each of these was also reported as most effective. The ways of learning reported as frequent were: customer conversations, rounds, and self-directed review of medical periodicals.

3.4 Attributes of (in)formality in the perceived effective and frequent ways of learning

The reported effective and frequent ways of learning exhibit a high degree of informality in six of Colley et al. 's seven process attributes, but a high degree of formality was associated with intentionality. In general, agents intended to learn even when objectives were not determined in advance. Agents reported a form of learning that I call intentional-incidental learning: They intentionally put themselves in situations where learning might occur incidental to other activities and keep themselves alert for opportunities. John said:

[Be in] the right place at the right time [...] by fluke, you meet the guy when he wants to vent to somebody. You find out a lot about their world [...] You stumble upon information.

Alice called it "serendipitous accidental learning [...] you don't know you're going to come up with information [...] and you don't know when you are going to use it." She explained that a good agent "is always looking for those opportunities and always putting themselves in positions where they can take learnings away that may not be obvious." The intention to learn is ever present, even when the activities are not primarily learning-focused. These situations often presented themselves during customer interactions, but also while agents waited and listened in waiting rooms, worked display booths, attended to personal matters, participated at charity functions, and interacted in their own social and familial groups.

The level of formality related to content and purpose attributes of the frequent and effective ways of learning was mixed. Purpose attributes tended toward a slightly more informal nature, and content attributes appeared to be slightly more formal. Setting attributes were mostly informal.

3.5 Characteristics of the perceived frequent and effective ways of learning

In order to report characteristics of ways of learning which are more elementary than attributes of (in)formality, it was necessary to reorganize and extend the work of [3] Colley et al. (2003). A framing tool was produced (Table II [Figure omitted. See Article Image.]) to guide the analysis and reporting of such characteristics of learning. Using this guide, each way of learning reported as frequent or effective was analyzed individually ([7] Hunter, 2009) and an extensive list of characteristics of each way was produced. ([7] Hunter, 2009). In total, 47 process, 17 setting, 16 purpose, and 16 content characteristics were identified for the ways of learning knowledge reported as effective or frequent. Only a brief summary of major trends is appropriate here.

Much of the knowledge learned in the ways identified as effective involved high-status (medical or privileged) knowledge but also practical knowledge and tacit understandings. Learning was either a primary or secondary purpose. Agents reported learning for personal reasons or to benefit the customer or employer. Other individuals may or may not be present during learning. When others were present they were often customers or peers and learning was a cooperative and negotiated endeavour. Outcomes, processes, timelines, and settings were often not predetermined and learning took place at various times during the work day and after work hours. Activities were highly learner-centred involving self-determined and self-initiated purposes, processes, and pace, and much learner autonomy. Learners made their own connections between the material learned and application to the world. The most frequent ways of learning were a subset of the most effective ways of learning and shared many of the characteristics listed above. These have also been analyzed individually ([7] Hunter, 2009). Of note, however, is that agents reported learning almost daily from customer interactions and self-directed review of medical journals.

Both customer and peer-facilitated ways of learning emerged as particularly effective. Therefore, the characteristics of these groups of learning situations were documented in detail ([7] Hunter, 2009). A summary is included in Table III [Figure omitted. See Article Image.]. Both peer- and customer-facilitated learning involved multi-directional processes where both the agent and the customer or peer learned. The processes were emergent and negotiated by the participants. Content was also negotiated by the parties and could be directed at meeting the needs of the agents, the customer, or the company. Often, these needs overlapped. Settings were usually friendly, at least somewhat casual, and involved a degree of mutual respect.

4 Discussion

This research provides a comprehensive list of the ways of learning for work used by agents in a previously unexamined competitive, rapidly changing, knowledge -intensive and geographically distributed workforce. It is one of the first to apply the [3] Colley et al. (2003) framework to document the attributes of (in)formality in the ways of learning perceived by workers to be most effective and frequently employed, and in doing so, confirms that most ways of learning have a mixture of formal and informal attributes. It has been argued that distinguishing formal and informal practices "is not helpful" ([1] Billett, 2002, p. 56). We need descriptions of "how people do learn in their workplaces" ([6] Gerber, 2006, p. 35). This study provided such a description and developed and applied a framing tool for guiding the reporting of characteristics of "how people do learn" ([6] Gerber, 2006 p. 35). The findings suggest that the pharmaceutical sales industry, and perhaps industries with similar work contexts, should support peer learning, perhaps by increasing opportunities for peers to work together, or by creating mentorships or remote peer-support networks. Agents also might benefit from opportunities to share strategies for positioning themselves for intentional incidental learning and ways in which industry-nonspecific learning has been used to develop customer relations.

This research presents only a few cases of learners in one industry. Interview participants were not obviously atypical of agents in this industry, but cannot be said to represent any specific proportion of agents. This research may be viewed as a start to a more full understanding of the characteristics of the ways of learning in geographically distributed and knowledge-intense workplaces. The framing tool helps initiate examination of characteristics of ways of learning in different work environments. We now need additional research using and supplementing the framing tool to document the characteristics of perceived frequent and effective learning in other industries. This will aid a more full understanding of how people learn in the contexts of different workplaces. By examining the characteristics of frequent and perceived effective learning in different work contexts, we might be better positioned to support workplace learning in a variety of environments.

Previous research ([2] Boud and Middleton, 2003; [4] Davis and Daley, 2008; [5] Davis and Sumara, 2001; [10] Mittendorff et al. , 2006; [11] Senge, 2006) has shown the importance of learning on the job from co-located peers. This study illustrates that in this knowledge-intense and rapidly changing industry, geographically distributed peers learn from each other and on the job from their customers. In a labour-market trending toward increasingly distributed workforces, this study suggests that we should support peer learning even in distributed work environments.

Much learning in this and other workplaces is applied directly to the job. This study identified another way in which learning can support business objectives: it can be used to develop a worker as a resource to customers in areas outside of the worker's area of specialization. Further research could investigate other non-direct ways in which learning affects business objectives and explore other industries where learning is related to customer-relations development. Industry should consider supporting learning outside of specific job requirements in order to develop its workforce for broadly.

An important finding of this research is the intentional nature of incidental learning for agents in this industry. The intentionality of being alert for unexpected learning while engaged in other activities has not been investigated. Further research should explore the ways in which workers position themselves for unpredicted learning opportunities and the level of intentionality involved in incidental learning in other work environments so that we might better understand how work and industry contexts affect the intentionality for incidental learning. Industry might support peer sharing of strategies and outcomes of intentional incidental learning.

In spite of the blind review within the text of the manuscript, Ms Hunter has had to refer to the online version of the Master's thesis from which this paper is derived. This was to allow interested readers access to more detailed presentation of the extensive data analysis. To maintain the blinding, she has referred to her previous work as "([7] Hunter, 2009)".