How domanagers promote workplace learning? Learning- oriented leadership in daily work

Purpose – The purpose of this paper is to explore managers’ learning-oriented leadership, and what conditions managers face when working with the promotion of employees’ learning. Design/methodology/approach – Data was collected through semi-structured interviews with managers in three manufacturing firms. Verbatim expressions of the interview participants were analysed using stepwise analytical procedures. Findings – The managers used many kinds of activities to promote learning. Most common were activities related to learning opportunities that arose during daily work. The identified activities ranged from being planned to occurring more spontaneously. Depending on the situation or the learning activity, the managers used different behaviours to promote learning. They supported, educated and confronted employees, and they acted as role models. Factors constraining the implementation of learning-oriented leadership included limited resources, and a lack of commitment from topmanagement, employees or the managers themselves. Research limitations/implications – Future research should study learning-oriented leadership from the employees’ perspective. Practical implications – Managers’ notions about learning and development constitute an important condition for learning-oriented leadership. Therefore, managers need to be trained in how to promote their employees’ learning at work. Originality/value – This study adds to the limited knowledge of how managers carry out a learningoriented leadership in their daily work. The findings contribute knowledge regarding managerial practices of promoting employees’ workplace learning by identifying different activities and behaviours that managers could incorporate into their leadership.


Introduction
Opportunities for learning and the development of employees' knowledge and skills are crucial for improving organisational performance (Ibidunni, 2020) but also for creating sustainable and healthy workplaces that can meet the challenges posed by society and working life (Garavan and McGuire, 2010;Karasek and Theorell, 1990). In the wake of the COVID-19 pandemic, for example, the capacity for learning and innovation has been identified as a crucial component in strategies for organisational change and human resource development (Dirani et al., 2020;Luo and Galasso, 2020;McLean and Jiantreerangkoo, 2020). Furthermore, participation in continuous learning and development activities makes important contributions to the development of favourable working conditions, such as job satisfaction and well-being (Cerasoli et al., 2018;Watson et al., 2018). Learning also generates innovative ideas for business development, increased productivity, greater flexibility in staffing and reduced costs owing to ill health and undesired mobility (Kim and Ployhart, 2014;Noe et al., 2014;Saks and Burke-Smalley, 2014). Opportunities for learning in working life can thus be seen as a central factor for achieving both individual and organisational sustainability (Elg et al., 2015).
However, employee learning does not necessarily arise on its own. It needs to be supported, made visible and disseminated within an organisation. Previous studies have indicated that managers' leadership plays a particularly important role in this regard (Coetzer, 2006;Cohen, 2013;Döös et al., 2015;Ellinger et al., 1999;Ellström and Ellström, 2018;Engeström, 1994). On this topic, there are several studies that have tried, through surveys, to verify the relationship between leadership on the one hand and learning on the other (Loon et al., 2012;Matsuo, 2012). However, some limitations have presented themselves with these previous studies. Firstly, many of them both begin and end by asking if leadership affects employee learning and thus provide limited knowledge concerning how such leadership is performed in daily work. Secondly, in line with Einola and Alvesson (2020), it can be argued that complex organisational phenomena such as leadership and learning are inherently difficult to quantify and study with ready-made measurements. Progress in research on leadership and learning in organisations has been limited by an over-reliance on behaviour description questionnaires that are not well-suited to capturing dynamic behaviours and activities that, for respondents, are not easily observed or remembered (Yukl, 2009). Thirdly, in many of these studies, potentially moderating or mediating contextual factors are not problematised to any great extent (Ellström, 2012), thus leaving out the situational nature of leadership (Hersey and Blanchard, 1969).
To counter these limitations, some studies based on qualitative data have previously been carried out (Coetzer, 2006;Cohen, 2013;Döös et al., 2015;Ellinger et al., 1999;Ellström and Ellström, 2018). While these studies have enhanced our understanding of learningoriented leadership, there is still a need for further research with a focus on what the manager actually does when leading for learning, and how this affects our understanding of leadership and the daily work of managers within organisations. For instance, there is a need to gain deeper insight into what types of activities managers intentionally use to promote their employees' learning, i.e. how they arrange opportunities for learning through formal learning activities such as classroom-based learning (Bednall and Sanders, 2017) and informal learning activities, i.e. through participating in work activities that offer rich learning experiences (Billett and Choy, 2013). Moreover, it is also important to determine how they are acting as managers in regard to these activities, i.e. what types of leadership behaviours they use. Or to be more specific, what manager actually do in their daily work, and under what circumstances and in what contexts this learning-oriented leadership is performed. The purpose of this paper is, thus, to explore managers' learning-oriented leadership, and what conditions they face when working with the promotion of employees' learning.
It is important from the beginning to clarify the scope of the study. When we use the term learning-oriented leadership, we are referring to managerial practices with a more or less conscious focus on the educative task of promoting employee learning at work. Thus, our specific knowledge interest in this paper concerns managers' views on performing this type of educative taskwhich is often quite alien to traditional managerial practices (Hales, 2005) and also how they actually carry out these educative tasks. Furthermore, in line with much research on managerial work (Mintzberg, 2009;Stewart, 1982;Tengblad, 2012), our focus concerns what managers do and not do, and under what conditions they do it. That is, the study focuses on describing and conceptualising managerial practices, rather than on the effectiveness of different practices for promoting employee learning and knowledge. In other words, our knowledge interest concerns what a deliberate learning-oriented leadership could mean in practice in different organisational contexts and under different conditions. This means that we have not included measures of different types of learning or knowledge use as criteria for the effectiveness of different managerial practices.
As far as the study's potential value for the research area is concerned, we have observed a strong bias towards normative theories of effective leadership that put leaders and managers on pedestals, without considering the contextual constraints in different types of organisations. Our focus on managers' daily work may generate knowledge that better reflects what the practice looks like and under what conditions managers will engage in forms of learning-oriented leadership. Such knowledge is assumed to be relevant for developing the learning-oriented leadership skills of managers.

Conceptual framework and previous research
Concept of learning-oriented leadership Learning is used here not as an object of study, but as an analytical tool for conceptualising and defining learning-oriented leadership. Based on previous research on learning-oriented leadership (Wallo, 2008;Wallo et al., 2013), we propose a distinction between performanceoriented and developmental leadership. This distinction has its roots in learning theory, in particular theories of adaptive/reproductive and developmental/innovative learning (Ellström, 2001;Engeström, 1987), but also in management theory and the well-known distinction between March's (1991) notions of organisational learning as knowledge creation and a renewal of organisational practices ("exploration") versus learning as a refinement of current practices ("exploitation").
An important function of learning-oriented leadership is to create favourable conditions for employee learning. Performance-oriented leadership is assumed to focus mainly on the promotion of adaptive learning. That is, a focus on the mastery of given tasks or methods, and on the formation of competencies for handling routine or frequently occurring tasks, with the goal of facilitating efficient and stable task performance. To accomplish this, performance-oriented leadership entails a reduction in performance variation within and between individuals through, for example, the standardisation of work processes and other measures for narrowing employees' scope for discretion in performing their work (Döös et al., 2015).
In contrast, development-oriented leadership entails developmental learning and preparedness to question, analyse and reflect uponas well as, if necessary, changingestablished practices into new solutions or ways of working. One way to promote this type of learning is to create conditions for openness, variation and diversity in thought and action within the organisation by adopting leadership practices (interventions) that aim to widen rather than narrow employees' scope for discretion at work (Döös et al., 2015).
Regardless of whether employee learning is mainly adaptive or developmental, it takes place within a particular physical, social and cultural context (e.g. within an organisation or a workplace). This so-called learning environment (Fuller and Unwin, 2011) influences the Workplace learning learning that it is possible to achieve and can thus be described as an important set of conditions for learning. Examples of such conditions within an organisation's learning environment are the content and nature of the work tasks (variation, degree of difficulty, etc.), the work organisation, management's and supervisors' support for learning, opportunities for feedback on performed tasks and the culture of the workplace (Ellinger, 2005;Skule, 2004). Depending on what the "local" set of conditions for learning looks like in an individual workplace, it is possible to talk about constraining or enabling learning environments (Kock and Ellström, 2011). A major task for learning-oriented leadership is to organise learning environments at the workplace that are conducive to employee learning.

Previous research on learning-oriented leadership
Research on learning-oriented leadership is scanty but has grown over the past decade (Döös et al., 2015;Ellström and Ellström, 2018;Matsuo, 2012). A research task that has attracted considerable interest is to focus on well-known models of leadership, such as transformational leadership (Bass, 1985) or leader-member exchange theory (Graen and Uhl-Bien, 1995), as ways to model and operationalise learning-oriented leadership. These studies are typically carried out as correlational studies based on questionnaires (Amitay et al., 2005;Berson et al., 2015;Beverborg et al., 2015;Loon et al., 2012;Pasamar et al., 2019;Salas-Vallina et al., 2017). While these quantitative studies have focussed on whether an association exists between leadership and learning in organisations, they tend to lack an analysis of when and for whom leadership is important and when it is not important (i.e. boundary conditions). In total, these studies tell us little about how learning-oriented leadership is carried out by managers. However, there are also some studies that are more in line with the practice-oriented approach used in this study. In these studies, the primary focus concerns how managers subjectively understand and actually carry out the educative tasks associated with a learning-oriented leadership climate (Beattie, 2006;Ellinger, 2005;Ellström and Ellström, 2018;Warhurst, 2013). In spite of a number of differences between these studies, not least concerning categorisation, there are some clear conclusions that can be drawn from them, including the critical importance of providing support before, during and after the learning activity; providing resources for learning; placing learning on the agenda; serving as a role model; and creating a learning climate (Beattie, 2006;Ellinger, 2005;Ellström and Ellström, 2018;Warhurst, 2013). Additionally, existing studies emphasise the function of managers in challenging workers and encouraging them to reflect upon their work (Beattie, 2006;Ellinger and Bostrom, 2002).
Based on a study of managers in two industrial companies, Wallo (2008) proposed a model for learning-oriented leadership comprising the types of activities the leader uses and the different roles that the leader assumes. The activities were categorised as planned, partially planned or spontaneous, depending on their different degrees of planning and formalisation. Planned activities are determined in advance and included in the manager's normal tasks and budget. Partially planned activities have a lower degree of formalisation and do not necessarily take place regularly but are rather based on learning needs that the manager or employee continuously identify. The third category is called spontaneous activities and they typically arise ad hoc during the process of carrying out work tasks. Such activities usually have a more informal and improvised character. The leadership roles in Wallo's (2008) model were denoted the supporter, the educator and the confronter. The supporter focuses on fostering relationships with employees to motivate and encourage them to learn. The educator actively arranges learning opportunities and contributes with knowledge and alternative perspectives to facilitate reflection and critical thinking.
The confronter challenges employees and places demands on their learning engagements. Managers can alternate between these roles, depending on the situation. To promote adaptive learning, the focus is on preserving the status quo and supporting, educating or confronting employees to improve upon their already acquired skills and abilities. To promote developmental learning, the focus is instead on using the three roles to introduce new ways of thinking so that employees can take on novel tasks and challenges.

Conceptual framework and research questions
The multiple behaviours identified in previous studies reflect a range of approaches to promoting workplace learning. In this section of the paper, we propose a tentative conceptual framework as a guide for investigations of learning-oriented leadership ( Figure 1).
In the model, we show that managers can essentially promote employees' workplace learning in two ways (Döös et al., 2015): through direct developmental interventions (Beattie et al., 2014) and indirectly through creating conditions within work environments that are favourable to learning (Coetzer et al., 2019;Ellinger, 2005).
In the subsequent sections of this paper, the focus will be on direct interventions, i.e. the upper part of the model, which consist of the activities a manager uses when attempting to promote employee learning and the type of behaviour a manager uses to influence the employees in relation to that activity. Depending on the combinations of activities and behaviours, the leadership will be either performance-oriented or development-oriented (Wallo, 2008), mirroring the two types of learning outcomes previously described, i.e. adaptive and developmental learning (Ellström, 2001). Furthermore, learning-oriented leadership is also situated within specific contextual factors, internal and external, that may either enable or constrain the managers' agency (Archer, 1995;Stewart, 1982).

Workplace learning
Based on the conceptual framework, the following research questions have been formulated to guide our study: RQ1. What activities do managers use with the intent to promote their employees' learning?
RQ2. What leadership behaviours do managers use with the intent to promote their employees' learning?
RQ3. What contextual conditions do managers face when working with the promotion of employees' learning and to what extent are these conditions viewed as enabling or constraining?

Method
The explorative purpose of the study and the effort to capture the contextual nature of leadership necessitated a methodology that was able to capture aspects of leadership that go beyond those that are possible to capture with questionnaires (Yukl, 2009). For that reason, we chose to carry out semi-structured interviews with 33 managers on-site in three manufacturing companies. In addition, we conducted meeting observations and shadowing (Czarniawska, 2007) of managers to learn more about the contextual factors in the companies. This paper reports mainly on the interview material but is informed by other kinds of data. Company Alpha is a global industrial manufacturer of military defence and civil security technology. In total, the company group has approximately 9,000 employees worldwide. Company Bravo manufactures electronic and mechanical products. The company has a total of nearly 1,000 employees in subsidiaries in Europe and Asia. Company Charlie develops, produces and installs automobile adaptation equipment. In total, the company has about 150 employees in a factory in Sweden. These companies were selected to obtain variety in the material, e.g. they differ in size and type of business and also because they share features that make them easier to compare, e.g. being manufacturing companies, having similar management hierarchies and having production systems influenced by the principles of so-called lean production, such as process focus, standardised work tasks, elimination of waste and systematic continuous improvement (Liker, 2004). With respect to the relationship between managers and co-workers in these companies, the adoption of lean production principles shifts the managers' work from managing the details of the work to, ideally, functioning as coaches, facilitators and/or consultants to their work teams (Wallo et al., 2013). Of the managers interviewed, 19 belonged to a higher hierarchical level (top or middle management, e.g. factory managers or department managers), and 14 belonged to a lower hierarchical level (line management, e.g. production leaders). Eight of the managers were women and 25 were men. The highest educational level of 21 of the managers was upper secondary school, while 12 had a university education. The nature of the managers' work in all three companies can be characterised, much like managerial work in general (Mintzberg, 2009;Tengblad, 2012), as varied, complex and replete with reactive problem-solving activities.
The interviews were based on a semi-structured interview guide that included the following areas and themes: personal background in leadership (e.g. education, employment, leadership experience and reasons for becoming a manager); work as a manager (What they do? Why these tasks? How much and with whom?); notions about and experiences of learning and development within organisations; and notions and experiences of the manager's role in relation to employees' learning and development at work.
The questions in the interview guide were formulated based on the purpose and research questions of the study. The ambition was to enable a relaxed conversation that would allow the respondents' stories and experiences to govern the interview process (Kvale, 1996). During the interviews, probing follow-up questions (Miles et al., 2014) such as "could you tell me more about [. . .]?" were used to elicit elaborations of answers, ask for clarifications or to find new angles not included in the guide. The interviews were conducted during the managers' regular working hours, on the companies' premises and in places that ensured privacy. The length of the interviews varied from 1 h and 15 min to 2 h. The interviews were recorded and transcribed verbatim. To build trustworthiness for the study and trust between the interviewer and respondents, written and oral information about the study was presented in advance. All the respondents had a free choice to decide if they wanted to participate or not. The analysis and interpretation of data can be seen as a process that has taken place with varied intensity during the entire study and followed recommendations made by Miles et al. (2014). The entire analysis and interpretation process can be seen as an iterative interaction between data and theory achieved through abduction (Danermark, 2002).
The first phase of analysis began immediately after the initial interviews. This ensured that the data collection instruments were both appropriate for the study setting and honed to pursue emerging issues so that new and potentially better data could be collected. In the second phase of analysis, the researchers read the transcripts several times to obtain a thorough understanding of the content of each interview. In the third phase, the transcripts were imported into the QSR NVivo software program, which allows users to store, organise and manage their data. It also facilitates coding and subsequent data analysis. Here, descriptive nodes, derived from the interview questions, were used to obtain an overall idea of the content of each interview. Examples of descriptive nodes regarding workplace learning were: "Where", "How", "Learning activities" and "Forums for discussion". Further examples of nodes were the managers' "Behaviour", experiences of "Conditions" and "Support" and subordinates' "Career opportunities" and "Job mobility". In the fourth phase, a provisional "starting list" of factors and conditions that might be useful for organising the collected data was created. This list comprised parts of the conceptual framework for learning-oriented leadership (e.g. previous research relevant to managing the data). Miles et al. (2014) refer to this step as creating codes. Specific activities, or actions performed by the managers with the intent to promote learning, were coded under the label "Activities", while the purpose of an activity, and its expected effect, was coded "Why". The specific behaviours that the managers described having used during different learning activities or situations were coded under the label "How". Contextual factors, such as time, production phase, individual interest or goals, which may affect their ability to implement different learning activities were coded under the label "Conditions". In the fifth phase, the findings from the earlier phases of the analysis were discussed to achieve a broader understanding of the data in relation to the research questions and conceptual framework. A cross-case analysis was used to identify common patterns in the three companies (Miles et al., 2014), and these commonalities make up the composition of the findings section of this paper.
In addition, the empirical material was presented in a series of seminars held at the three companies, which aimed to create opportunities for joint analysis of the findings and a critical and reflective learning process for both researchers and practitioners (Ellström et al., 2020). In this step, representatives from the companies validated the findings, i.e. informant validation, but also helped the researchers to develop their analysis by providing alternative perspectives and insights. According to the practitioners, the seminars provided new knowledge that supported leadership development in all companies.
The categories in the empirical material are representations of the most common responses, but answers that indicate alternative notions are also included in the text. Quotes have been used to illustrate the lines of reasoning and to give a rich impression of the aspects or phenomena being discussed.

Findings
The results of the study are presented in three sections relating to the research questions. The focus of the presentation is primarily on the similarities between the three cases.

Activities used to promote learning
The most common activity that the interviewed managers in the case organisations used to promote learning was to conduct regular individual conversation with the employees, e.g. personnel development dialogues or performance appraisal meetings, to discuss their shortterm and long-term needs for learning and development: Well, a lot of it's done in discussions in the development dialogues, at least for the coming year or general interest. And more, if it also comes through in improvement meetings, if they bring up things they would like to learn more about, we'll help them find the right routes for that too. (C6) A widespread notion among the managers was that learning is primarily embedded in daily work practices. In line with this notion, an important activity was to arrange job rotation by assigning new tasks to an employee or making other changes in the employee's work situation. The learning that stems from job rotation is intended to produce more self-reliant employees and help to reduce the workplace vulnerability associated with absent employees, e.g. because of sick-leave, parental leave or holidays: There are a couple of guys who are a little newer than the others and we're currently moving them around so they'll learn. (B5) The managers used similar arguments to describe how offering, or arranging, different types of formal training, such as courses and trainee programmes, were important activities to facilitate learning and strengthen the competence of both individuals and the company. In addition, many managers also talked about designating time in meetings for discussions and reflections aimed at promoting learning. Often, such educational interventions were based on a need for learning that had been observed to arise in connection with daily work. Some of the managers also described how they conducted improvised educational interventions "on-site", for instance by gathering a group of employees together, to both solve and learn from a recurring problem in production.
Other more spontaneous activities that were mentioned as being conducted ad hoc in relation to the process of working included informal conversations during which a manager saw an opportunity to promote learning: Yes, but if somebody says, "I would have enjoyed trying this kind of work sometime," I respond "certainly, then we'll arrange that type of job next time". (C3) Finally, additional activities mentioned by some of the managers were using suggestion boxes to encourage employees to think about development issues, making succession inventories and designing internal career ladders.

Behaviours used to promote learning
The interviews show that the managers used different types of behaviours in connection with the activities described in the previous section. Furthermore, they reported alternating between these behaviours depending on what the employees required. The most common behaviours were supporting and encouraging employees to either improve their pre-existing skills or learn new things, for instance by attending courses or trying out new tasks: Decide what you want, and then I can help you, or we can help you go in for that bit. (A10) The managers said that they spent a lot of time motivating employees, especially those who had the potential for development, but who did not dare, or did not want, to step forward. According to the managers, they needed to "scout" the employees to see who showed the most promise and then gently nudge them into taking on more responsibility. According to the managers, you need to be patient and respect the reality that different individuals have different starting distances. Sometimes the managers also had to reassure employees who had been assigned to new duties or to change departments so that they did not think the manager was trying to get rid of them: And it went very slowly at the beginning, if there was somebody who would be lent out to the next department, then it was like a game show on TV and "I'm voted out", so it went pretty slowly there. (A3) According to the managers, some employees were more eager to take up learning than others, constantly raising their hands to volunteer for learning activities and applying for all the vacant positions in the organisation. Here, the managers talked about helping them to focus their efforts, while concurrently prioritising whose turn it was to attend a course and determining whether an individual's learning need was relevant for the company: I tell them I must prioritisesome people want to learn everything. And that doesn't benefit the company. (C6) A behaviour that was also highlighted as a fundamental part of the managers' leadership was showing commitment to learning issues, partly by discussing learning in daily work and being present for the employees and partly by setting a good example and showing an appreciation of new challenges and a desire to learn and develop. One respondent said that his career was a good example of the opportunities for development in the company: I'm a pretty good example myself, having gone that route. If you want to, you can. (A4) Another way to promote learning was to hold back from giving ready-made answers or instructions. Instead of immediately answering when someone asks how they should solve a problem, the managers challenged employees by asking counter-questions and making them think of a creative solution themselves. One manager explained that this way of facilitating a discussion to find an answer leads to more self-assured employees: So, I ask, "what do you think?" Like that, trying to get them to take the plunge and decide on their own. And it's worked very well because, these days, the person seldom comes to me. (C9) Furthermore, the managers said that sometimes they need to be a little tough on employees by consciously trying to challenge and push them to engage in learning activities, take Workplace learning initiatives and come up with innovative suggestions for improvements. Being demanding could sometimes be perceived as provocative but, according to the managers, it could also help employees to understand the importance of learning and reflection at work: I challenge them a good deal. It's like I said at the beginning, I can come off as hard, annoying, and somebody who chases them. (A10) Another important behaviour was trying to create a shared understanding in a workgroup to increase their knowledge of their work. One respondent explained that this involved trying to get the groups to look ahead, to see past where they are right now. For some of the managers, it was also important to work with standardisation and finding common rules in the groups to get everyone on the same page: Not everyone may agree on everything, but if we agree on it as a group, you still have to respect it and it becomes a form of structure, something to stick to. (A11) In a similar vein, a common response from the managers was that, while it was sometimes important to promote new and creative ideas, for instance in improvement meetings, they also needed to facilitate learning that was directly useful to the company, thereby sometimes having to manage and regulate the learning processes from the top down.

Conditions that may hinder or enable the promotion of learning
The results also revealed several conditions that seem to either hinder or enable the managers' attempts to promote learning. A commonly mentioned factor was a shortage of time to work with learning issues, often caused by having to deal with more pressing matters, such as problem-solving in the production process. In a similar vein, the managers also said that, in a strained financial situation, it was more difficult to find resources for training activities. The respondents pointed out that, in this regard, support from top management was crucial and closely related to how many resources learning activities could use: Well, sometimes it's top management that must provide the company with conditions to be able to work with learning and development. (C7) Another important factor was the employees' commitment to learning activities. In many cases, employees ran their development processes freely, while in others there was almost no such willingness. Similarly, managers also highlighted their own commitment and competencies as influential factors. For instance, a manager's knowledge of the organisation was important because it determined what opportunities for collaboration, job rotation and other learning activities they were able to identify. Finally, there were also practical factors related to the design of the work organisation. For instance, some employees could not be moved because their specific skills were needed in one place or because of certain language requirements in different work areas. In addition, shift work sometimes prevented managers from interacting regularly with all employees. It was also mentioned that the available career opportunities mattered greatly. It was easier to build individuals' skills when they had opportunities to rise through the ranks. In this regard, the results showed that the largest of the three companies, owing to its size and number of hierarchical levels, had better prerequisites for creating internal career ladders.

Discussion
The findings show that the managers used many different planned, partially planned and spontaneous activities (Wallo, 2008) to promote learning. The spontaneous activities included, for example, informal conversations in which the manager saw an opportunity to promote learning, or learning while solving work-related problems (Cohen, 2013;Wallo, 2008). The partially planned activities also originated in a need for learning that arose from practice but involved at least some degree of planning on the part of managers. Examples of such activities included arranging job rotation to allow employees to try out novel work tasks and to learn from more experienced co-workers (Coetzer, 2006;Wallo, 2008). The planned activities were often quite disconnected from the practice itself, typically involving formal training activities, such as courses and educational programmes or recurring individual meetings between managers and employees (Ellström, 2012;Ellström and Ellström, 2018;Wallo, 2008).
The behaviours used by these managers were similar to the managerial roles found in previous empirical studies. The findings show that the managers put a lot of effort into supporting the learning process of the individuals they managed and encouraging them to take steps forward in their careers (Beattie, 2006;Ellinger and Bostrom, 2002;Ellström, 2012;Ellström and Ellström, 2018;Wallo, 2008). The managers also acted as educators, trying to stimulate the intellects of employees, by actively teaching them about work procedures, stimulating critical reflection or refraining from providing answers to encourage them to think for themselves (Amy, 2008;Ellström and Ellström, 2018;Wallo, 2008).
The third type of behaviour was to make demands on employees in terms of their commitment to learning. By confronting employees, for example by asking critical questions, the studied managers hoped to challenge and push the employees out of their comfort zones. This behaviour was used especially in relation to those who showed a negative attitude towards learning and development activities, or those who did not dare to take on new challenges. This also bears a resemblance to the roles found in previous research (Amy, 2008;Beattie, 2006;Ellinger and Bostrom, 2002;Wallo, 2008). In previous research based on transformational leadership, this kind of leadership has typically been considered less favourable for learning (Amitay et al., 2005) and thus has not been extensively discussed in relation to learning and development. However, the managers in our study did not view this role as less important than the supporter or educator roles. These results show that a confrontational leadership style does not need to be understood as innately negative in relation to employees' learning and development. Rather, setting boundaries and demanding reflection may help to initiate learning and development in some situations and contexts (Beattie, 2006;Ellinger and Bostrom, 2002;Wallo, 2008;Warhurst, 2013).
The results also show that the managers acted as role models to support their employees' learning. This behaviour involves leading by example by displaying the willingness to learn by being visible within the organisation and showing a commitment to learning activities. Being a role model has clear similarities with the inspirational aspects in Bass' (1985) transformational leadership and the behaviour that Beattie (2006, p. 108) refers to as "being professional", which involves "behaving in a manner that people respect and wish to emulate".
In terms of the studied managers' ability to perform learning-oriented leadership, the findings suggest that the managers' ways of acting differently depended on how the interaction with employees was enabled and that they were constrained by structural and individual conditions (Beattie, 2006), which may breathe new life into the situational leadership models (Hersey and Blanchard, 1969). Example of factors that can constrain the implementation of learning-oriented leadership include limitations in time and financial resources and lack of commitment from top management, employees or the managers themselves (Ellinger, 2005;Ellström, 2012;Ellström and Ellström, 2018;Wallo, 2008). In light of these results, it is also possible to raise concerns about the universalistic applicability of popular leadership theories, such as transformational leadership (Bass, 1985). Even though there are many similarities between the cases in this study, this does not Workplace learning imply that a learning-oriented form of leadership will look the same in contexts with radically different structural conditions, such as in public-sector organisations. Moreover, learning-oriented leadership may also differ in form depending on what type of learning the manager is trying to promote. The findings provide examples of development-oriented leadership, i.e. promoting innovative thinking, as well as performance-oriented leadership, i.e. improving employees' performance by building on their pre-existing competences (Ellström and Ellström, 2018;Wallo et al., 2013). Although our data does not allow more precise conclusions to be drawn concerning the relative prevalence of one or the other mode of leadership, there are indications that, in practice, many managers were able to successfully handle both modes of leadership. At the same time, the study clearly shows that managers can choose between a repertoire of activities and behaviours in performing a learning-oriented leadership and that these choices can be decisive for the mode of leadership they exercise and thus also for the extent to which they create a constraining or enabling learning environment for their employees. However, as also argued, the managers' choices of activities and behaviours are not in any sense completely "free", but are shaped by a number of demands and constraints (Stewart, 1982) in their work context.

Conclusions
In this paper, we have discussed the meaning and implementation of learning-oriented leadership in daily work, based on previous research and an empirical study of managers in industrial companies. The conclusions we can draw are that the facilitation of employees' learning is an important task for managers and that, while a learning-oriented form of leadership seems to be carried out in their daily work, its implementation can vary greatly depending on several different factors that affect the managers' latitude. Learning-oriented leadership is thus to be understood as situational in the sense that managers can choose how they want to act and what activities they want to use, but their choices are constrained or enabled by broader factors within the organisational context.
We can also conclude that a learning-oriented form of leadership does not necessarily have to be difficult to achieve or require special arrangements; rather, it can be advantageously based mainly on the informal learning that is constantly on-going during the daily work. Although planned activities, such as participation in formal learning arrangements, can provide a welcome addition of new knowledge, the utilisation of partially planned and spontaneous activities can integrate learning-oriented leadership more closely with managers' daily work rather than being seen as yet another new task that they must add to their already full calendars. Finally, we can also argue that different kinds of learning appear to require different kinds of learning-oriented leadership. In relation to promoting adaptive learning, leadership can be directed towards promoting stability and routines within which learning can be found. When it comes to development-oriented learning, leadership is instead about supporting critical reflection, innovative thinking and readiness to reconsider habitual working methods.
When taken together, this study adds to the limited knowledge of how managers carry out learning-oriented leadership in their daily work. It does so by identifying the different activities and behaviours that managers can use to promote employee learning. In this regard, this study complements previous survey-based research by adding more detailed and contextually anchored descriptions of how managers think and act when attempting to exercise a learning-oriented leadership.

Implications for research and practice
An implication of this study for further research is the need to study the important balance between a developmental and a performance focus within organisations. This balance is needed to create opportunities to promote both adaptive and development-oriented learning (Ellström, 2001). However, such a balance does not necessarily arise spontaneously. An important task for future research is, therefore, to investigate what individual factors and organisational conditions could promote the realisation of such a balance. Furthermore, as most studies in the field tend to focus on the managerial perspective, it would be valuable to study learning-oriented leadership from an employee perspective. It is not unlikely that employees will be able to identify additional important behaviours and activities of managers. A focus on the employees' views of learning-oriented leadership may provide insights into how their learning is affected by the different activities and behaviours of the managers.
In terms of practical implications, we have seen that the managers' notions of learning and development constitute an important condition for learning-oriented leadership. With a restricted view of learning comes a restricted repertoire of activities used to facilitate it. Based on this argument, an implication of this study is that managers may need to be trained in how to promote their employees' learning at work. For managers, such training could make them better at identifying the potential for learning in daily work and help them develop abilities to shape the learning environments of their employees. Another practical implication is that an understanding of the conditions, both inside and outside organisations, that may benefit employee learning is crucial for organisations to attract and recruit the best talent. Thus, the results of this study may also be beneficial for HR practitioners working with employee resourcing issues.

Limitations
Finally, one could ask: Are the results presented in this article valid for managers regardless of their line of business? In our opinion, it is not possible to provide detailed descriptions and models that are applicable regardless of the business or the conditions that prevail in a particular case. Moreover, the results presented in this paper should be considered in relation to the limitations of the study, e.g. the use of self-reported data presents potential limitations related to the internal validity of the findings (e.g. discrepancies between what people say they do and what they actually do), the fact that respondents only represented the managerial perspective and that data was only collected in industrial settings. Because the sample of managers was not randomised, the empirical generalisability of the results is limited. However, the fact that the main results are consistent with a number of previous studies arguably strengthens the analytical generalisability of the findings (Firestone, 1993). The findings were also validated by the participants from the studied companies in seminars. In line with Larsson's (2009) discussion about generalisation from contextual similarities, it can be argued that the three companies are not extremes and that we are likely to be able to find other contexts where the external environment and symbolic level are similar. Furthermore, we claim pragmatic validity for the study in the sense that our findings can contribute to concretising the meaning of learning-oriented leadership and can provide a basis for reflection and learning linked to one's own practice.