Raising Concerns at Work in the Swedish Healthcare and Educational Sectors: Silence and Voice Climates

This article is based on an interview study of workers in Sweden’s healthcare and educational sectors who raise or withhold concerns about conditions at the workplace. Five groups of climates regarding concerns raised internally were found: A high voice climate, a moderate voice climate, a restrained voice climate, a resignation silence climate, and a silence and fear climate. A democratic leadership orientation lays the foundation for a voice climate based on trust and open dialogue. Silence and fear climates are created by an autocratic and retaliative leadership orientation. Also, a laissez-faire leadership can have a silencing effect on the workplace climate. Autocratic leadership and a hierarchical view on communication are silencers both regarding concerns raised internally and externally. If organisations are exposed to competition, loyalty towards the organisations tends to be strengthened in order to protect the brand, thus having a silencing effect on raising concerns externally. As a consequence, teacher practise self-censorship in order to conceal problems at their school, thereby protecting the school’s reputation on the market.


Introduction
Workers are often met with negative reactions from managers when raising concerns about conditions at work. However, raising concerns seems to be more difficult in some workplaces than others, even in the same organisation (Detert & Trevino, 2010), which suggests that organisational factors such as climate and leadership probably affect why workers chose voice or silence. Literature on voice and silence have sometimes taken separate lanes, but I agree with Morrison (2014, p 177) that they should be integrated. "…silence is failure to voice, and voice is a choice (deliberate or otherwise) to not remain silent." Morrison (2014: 174) defines employee voice and silence as follows: "…I am defining employee voice as informal and discretionary communication by an employee of ideas, suggestions, concerns, information about problems, or opinions about work-related issues to persons who might be able to take appropriate action, with the intent to bring about improvement or change…silence refers to not speaking up when one has a suggestion, concern, information about a problem, or a divergent point of view that could be useful or relevant to share…".
The focus of this paper is the healthcare and educational sectors in Sweden. Patients, pupils and students are often in a dependent role in relation to staff and their managers. Staff must therefore have possibility to report concerns when the wellbeing and rights of patients, pupils and students are violated.
In Sweden employees in the public sector have freedom of expression protected by the constitution. They have the right, within certain limits, to raise concerns to external recipients, for instance media, about what is going on in their organisations. However, the freedom of expression can sometimes collide with the loyalty principle. An employee should, according to the employment contract, be loyal to his/her employer. It means that the employee should not harm the employer and put the employer's interest before their own. However, for public employees' freedom of expression outweigh loyalty to the employer. Also, employees both in the private and public sector has a right to raise concerns internally about actions taken by the employer and about conditions in the workplace -indeed, this is a necessary condition for efficient trade union operation at workplaces (Fransson, 2013). In spite of legal protection, workers not seldom experience difficulties in raising concerns, both internally and externally.
The term whistleblowing is often used to refer to workers who report concerns over problems in an organisation. Whistleblowing covers some, but not all complaints from workers that this study focuses on. An often-used definition of whistleblowing is 'the disclosure by organization members (former or current) of illegal, immoral or illegitimate practices under the control of their employers, to persons or organizations that may be able to effect action.' (Near & Miceli, 1995: 508).
Not all raised concerns at workplaces can be regarded as whistleblowing. Often, workers raise concerns, that are not so severe in character, and are intended to improve the efficiency of the organisation, or protect the wellbeing of customers or workers. Some of the concerns are channelled through trade unions, especially regarding working conditions and negotiations about major changes in the organisation. There is an extensive institutional framework in Sweden supporting union activity. Around 70% of the workforce in Sweden are members of a union (Kjellberg, 2017) and most workplaces have union representatives.

Whistleblowing, Voice, Silence and the Organisation -Focusing the Brand, Climate and Leadership
This article is building on research on whistleblowing, voice and silence. Literature on whistleblowing often mention Nader (1972) as the starting point for studies in this area. The concept of voice is usually attributed to the work of Hirschman (1970). Silence is a later concept, introduced by Morrison and Milliken (2000). According to whistleblowing research, whistleblowers often experience retaliation from managers. Miceli, Near and Dworkin (Miceli et al., 2008: 11) define 'retaliation against whistle-blowers to be undesirable action taken against a whistle-blower -and in direct response to the whistleblowing.' Undesirable action, according to the authors, includes hostile action but also the omission of positive action, such as a pay rise. Also research on voice and silence similarly report that employees often are afraid to speak up (E.g. Chamberlin et al., 2017).
However, people who raise concerns are not always met with negative reactions. Sometimes concerns receive positive responses and are even welcomed by managers. Previous research has to a large extent tried to explain these differences by focusing individual differences of the people who blow the whistle, as well as situational factors affecting the whistleblowing process, national legislation and whistleblowing policies in organisations. Previous studies have also highlighted the different phases of the whistleblowing process (Miceli et al., 2008;Olsen, 2014). There is, however, a lack of studies that aim to explain why organisations differ when responding to concerns from workers (Lewis et al., 2014;Miceli et al., 2008;Vandekerckhove et al., 2014).
Raising concerns internally is in line with a prosocial view of whistleblowing, namely whistleblowing as an act that aims to protect people, the society and/or the organisation from detrimental activities (Cassematis & Wortley, 2013;Near & Miceli, 1996). There are advantages, from a management perspective, in promoting internal whistleblowing. The organisation culture might benefit, because workers feel that wrongdoing is taken seriously and is corrected. Also, workers do not need to notify external recipients, for instance media, about problems, thereby jeopardising the reputation of the organisation (Miceli et al., 2008). According to a study of the Swedish police force (Holgersson, 2019), the police management strive to maintain a positive image of the organisation and problems are concealed in order to protect its reputation. It is more important to cover up problems than to disclose information externally that can damage the positive image. Furthermore, employees perceive it to be very likely that they would be retaliated against if they blow the whistle externally, especially about issues that can damage the reputation of the organisation. However, previous whistleblowing research covering protection of the brand has not distinguished between organisations exposed to competition, and those who are not. The police force is an example of an authority, thus it is not exposed to competition. How has the introduction of competition in the public sector (e g. Hood & Dixon, 2016) affected workers willingness to disclose wrongdoing externally?
Some authors have suggested that culture or climate has a stronger influence than other factors when workers make decisions about raising concerns (Blenkinsopp et al., 2019). There is, however, confusion as to how to understand and define culture and climate. Some authors do not define the concepts at all, or if there are definitions, these often differ widely between authors (Furnham, 2005). Following the overview by Furnham (2005) culture is rooted more deeply in the history of an organisation or a group, and is composed of shared values and assumptions that are not so easy to change. Climate, on the other hand, tends to refer to more observable phenomena that are relatively temporary and easier to influence by for instance managers, which is why I am focusing on workplace climate in this paper. The organisational structure, culture and policies influence an organisation's climate as do external forces such as the wider economy, politics and technology. According to Furnham one of the most influential factors on workplace climate is leadership style and behaviour. Morrison and Milliken (2000, p. 714) introduced the concept climate of silence. They define a climate of silence "as one characterized by two shared beliefs: (1) speaking up about problems in the organization is not worth the effort, and (2) voicing one's opinions and concerns is dangerous." How then, is a climate of silence developed? Miceli et al. (2008) refer to several studies where a climate that tolerates harassments result in that the harassments continuous. Furthermore, if workers perceive the climate as threatening and raising concerns are met with reprisals external whistleblowing tend to increase (Miceli et al., 2008). Fear of reprisals is also a concern for non-reporting workers of wrongdoing (Cassematis & Wortley, 2013;Milliken et al., 2003;Morrison, 2014;Morrison & Milliken, 2000). Milliken et al. (2003) conclude from an interview study that a culture not open to upward communication is a silencing factor (See also Francis, 2015). Other silencing factors according to Morrison's (2014) review on worker voice and silence are abusive leadership (managers who insult or blame workers), and a hierarchical organisation. Also supervisor undermining, which is not openly hostile behaviour but actions that hinder a workers to work successfully together can have a silencing effect. Examples are belittling suggestions from workers or ignoring concerns (Frazier & Bowler, 2015).
On the other hand, raising concerns internally is more likely if workers feel that concerns will be listened to, it is normal practice to discuss ethical issues, actions are taken to solve the raised issues, and workers are not met with reprisals (Francis, 2015;Kaptein, 2011;Near & Miceli, 1996;Olsen, 2014;Rothschild & Miethe, 1999;Tangirala & Ramunajam, 2012;Trevino & Weaver, 2001). Keenan (1988) concludes from a survey study of first-level managers that a supportive and low defensive climate positively affect willingness to blow the whistle. Morrison et al., (2011) point out that voice climates, where people feel it is safe to speak up, has some similarity to psychological safety (Edmondson, 1999). However, voice climate is more specific as it focuses on safety to speak up about concerns. In their survey study Morrison et al. (2011) reported that group voice climate highly correlated with voice behaviour. Some of the groups had climates of silence, others had much more supportive voice climates.
Developing an ethical culture in the organisation has been successful in some cases (Benson & Ross, 1998;Hassink et al., 2007;Miceli & Near, 1992). However, as Trevino and Weaver (2001) point out, in order for ethical programmes to work, it is not enough for managers to talk about ethics, they need to react to ethical problems and find solutions to them. Trevino and Weaver (2001) point out that the effect of such programmes is reinforced if workers find that they are treated in a fair and respectful way.
According to Vandekerckhove et al. (2014) research has failed to focus on the recipients of concerns, in most cases managers. Several studies conclude that managers strongly influence employees' decision to speak up or not, depending on employee perception of supportive or inhibiting behaviour from managers (Detert & Trevino, 2010;Frazier & Bowler, 2015;Morrison et al., 2011). An overview of research in healthcare found similar patterns (Blenkinsopp et al., 2019). Managers may interpret voiced concerns as a threat to themselves, if it is seen as critique of the manager's performance or ability (Fast et al., 2014).
From a large survey study in Australia, the results show that workers who have a high level of trust in management are more likely to raise concerns internally (Donkin et al., 2008). See also overviews on voice and silence for similar findings (Chamberlin et al., 2017;Morrison, 2014).
A scenario study, using university students in India, indicates that ethical leadership predicts whistleblowing. An ethical leader is trusted, makes fair and balanced decisions, listens to workers, sets an example in acting ethically, and discusses ethics with workers (Bahl & Dadhich, 2011). Similar findings regarding ethical leadership (Avey et al., 2012) and high leader member exchange (Botero & Van Dyne, 2009) are reported from the voice literature. Authentic leadership, which partly builds upon ethical leadership, has also been positively related to internal whistleblowing in a Chinese survey study. Authentic leadership is based on principles such as listening to other opinions, sharing information and ideas with workers, and a strong moral awareness guiding the leader's way to manage (Liu et al., 2015).
One of the first models in leadership research builds upon three different leadership styles -autocratic, democratic and laissez-faire (Bass & Bass, 2008). The model was developed by Lewin, Lippit and White (Lippit, 1940). I find that this model is well suited to analyse this interview study as it fits well with the findings. The autocratic leader takes all decisions and workers are expected to follow them without argument. The autocratic leader is associated with classical organisation models such as Taylorism, hierarchical and/ or bureaucratic organisations. These models have had, and still have a substantial impact on how to organise organisations (e g. Morgan, 2006). According to this traditional bureaucratic view on organising, communication should take place in a controlled vertical way, following the chain of command (e g. Lewis, 2011;Morgan, 2006;Morrison & Milliken, 2000;Near & Miceli, 1995, 1996. As Near and Miceli (1995: 701) put it, with reference to Max Weber: "When the authority structure of a bureaucracy is challenged, the supporting structure of the entire organization may be eroded…" Questioning decisions is seen as a violation of the authority of managers (Morrison & Milliken, 2000;Near & Miceli, 1996following Weinstein, 1979. The democratic leader is associated with organic organisations and/or learning organisations. Organic organisations (Burns & Stalker, 1961) and learning organisations promote innovation and questioning taken for granted assumptions in the organisation (e g. Argyris, 2004). The democratic leader strives to take group decisions and to have good relations with workers. He or she shares information and asks for ideas and opinions from others. Critical viewpoints are asked for and are seen as necessary in order to develop the workplace and the organisation. The democratic leader also actively manages group processes, for example deals with conflicts and bullying. The laissez-faire leader is inactive and leaves everything to the workers. This leader has often low self-esteem, does not set any goals and does not give support to workers and does not help groups to take decisions (Bass & Bass, 2008).
An autocratic and a democratic paradigm often clash in organisations, as exemplified by an interview study by Glazer and Glazer (1989). The 64 whistleblowers talked about a strong hierarchy and an undemocratic culture. Expressing their own opinions and making complaints was seen as disloyal behaviour by managers and was met with retaliation. Two different traditions are colliding here: the workers acting within a democratic tradition where open discussion is natural, and the autocratic tradition, where managers demand obedience (Glazer & Glazer, 1989).
To conclude, there seems to be agreement on some factors that promote a voice climate: relations at the workplace based on respect, trust and openness, managers listening to concerns and taking action. On the contrary, retaliation and harassment from managers tend to cause silence or fear climates (Morrison, 2014).
There are however a number of deficiencies in previous studies. Previous research has to a large extent assumed that there is either a silence or a voice climate (e g. Morrison, 2014). A study by Morrison et al. (2011) on group climate concluded that voice climates differed a lot. Some were very negative and some were very positive, but the authors did not provide a framework to make sense of the different climates.
There is, to my knowledge, only a few attempts to develop typologies of different silence/voice categories. Wang and Hsieh (2013) Van Dyne et al. (2003) presents three categories of silence and three categories of voice based on an actor's motives. (1) Acquiscent silence is characterised by resignation. Employees either find speaking up is pointless as they perceive it not likely to change anything, or, they do not think they have the capability to influence conditions. (2) Defensive silence is characterised by fear. It is a conscious decision in order to protect the self from threatening situations. Defensive silence can also be based on fear of being seen as responsible to a problem. (3) Prosocial silence is built upon Organisational citizenship behaviour (OCB), thus keeping silent about facts or opinions that can be detrimental for other people or the organisation. The employee for instance decides not to reveal information that can be seen as confidential cause problems for the organisation. (1) Prosocial voice, like prosocial silence, is based on benefitting others or the organisation. According to the authors, examples are providing solutions to problems, suggestions for alternative strategies or ideas for change.
(2) Defensive voice is self-protective and based on fear, for example blaming other people or shifting attention to external factors. (3) Acquiscent voice finally, is voice based on resignation, for instance "go with the flow" because it might be the easiest way, or based on experience that previous attempts to come up with alternative solutions have been ignored. Liang et al. (2012) differ between prohibitive and promotive voice. Prohibitive voice is raising concerns concerns about practises or behaviour that are harmful for the organisation. Promotive voice is providing new ideas and solutions to problems or how to do things better in the future. My study incorporates both these variants.
Furthermore, participation in decision-making has not been taken into account in studies about whistleblowing and voice/silence climates. The research group Industrial Democracy in Europe (Heller, 1998) has developed a scale of participation in decision-making, the Influence Power Continuum, which can be used as an analytical framework: 1. I am not involved at all. 2. I am informed about the matter beforehand. 3. I can give my opinion. 4. My opinion is taken into account. 5. I take part with equal weight. 6. I decide on my own. (Heller, 1998: 159). Also, we do not know enough about how voice and silence climates are produced. Why do managers' responses differ so widely, and what makes managers more or less receptive to voice (Morrison, 2014)? The influence of leadership styles on silence and voice climates and behaviour have not been sufficiently studied (Morrison et al., 2011). There are a few studies, such as ethical leadership (Bahl & Dadhich, 2011) and authentic leadership (Liu et al., 2015). However, these studies, and most other studies have used surveys or scenarios as research methods. Respondents have been presented with hypothetical questions, or scenarios, such as how they would react if they observed wrongdoing. It is however, highly doubtful if the respondents' answers correspond to how they would actually react in reality (Olsen, 2014;Vadera et al., 2009). There is also a lack of qualitative studies that might shed light on the complexities regarding raising concern at the workplace (Loyens, 2013). Furthermore, few studies have focused on real examples of workers raising concerns. Two exceptions are Loyens (2013) and Milliken et al. (2003), who asked respondents to describe actual events when concerns were raised and responses from managers.
There is a lack of studies in the educational sector regarding voice and whistleblowing. Regarding the healthcare sector, there are few examples in the mainstream literature of studies in this sector. According to Blenkinsopp et al. (2019), who have conducted an overview of whistleblowing research in the healthcare sector that covered 55 studies. These studies seldom referred to the mainstream literature and seldom built on previous research and theory building.
The focus of this study is the healthcare and educational sectors in Sweden. The questions this article aims to highlight are: How can different voice and silence climates be categorised? How does leadership practice affect voice and silence workplace climates? How has the introduction of competition between public and private organisations affected the climate for raising concerns externally?

Method
Because of the lack of qualitative studies in this field, this interview study has an exploratory design. A pilot study has been conducted in order to get an overview of the research field in the private and the public sectors, and to help formulate the interview questions for the main study. In the pilot study, five individual and eight group interviews have been conducted, mainly with union representatives in the Gothenburg area of Sweden. In all, 37 people have taken part in the pilot study. In the main study, 89 individual (62 women, 27 men) semi structured interviews have been conducted with employees in the healthcare and educational sectors. The respondents have answered questions concerning 90 workplaces (some respondents have talked about more than one workplace). 50 interviews with employees with a union position and 39 interviews with employees who did not have a union position have been conducted. The interviews lasted between 30 and 70 min. The respondents decided themselves where they wanted to be interviewed, and the interviews took place at University of Gothenburg, at union premises, in cafes and a few at workplaces. Out of 90 workplaces, 19 are in the private sector. Workplace categories as follows:

Selection
The respondents have been found partly by contacting unions and partly through own contacts with people working in the two sectors. The selections have also been made by using the snowball method, namely by asking respondents if they can ask other people working in the two sectors if they want to be interviewed. The number of people who have declined to be interviewed was 13 in the healthcare sector and 17 in the educational sector. The most common reason given was lack of time. It is not known exactly how many people said no in the "snowball process", but it might be at least about the same number of people, namely 30. It has been especially hard to get in contact with employees working in the private sector. They have also often declined to be interviewed.

Interview Design
The focus of the study was broad initially. The aim was to study organisational factors affecting worker's possibility to raise concerns about conditions at the workplace. Through the research process the focus of the study has narrowed down to the present one. The research design is in line with authors who have suggested a mixed methods design where quantitative and qualitative approaches complement each other (Holgersson, 2019; Vadera et al., 2009). The interview questions are semi structured. Some questions have open answers and some have standardised answers. The study is mainly qualitative, but it has, to some extent, a mixed method design (Bryman, 2011).
In line with interview studies by Loyens (2013) and Milliken et al. (2003) the respondents have been asked to describe actual episodes when they or their colleagues have raised concerns, the reactions of their manager, and what actions, if any, their manager has taken to solve problems highlighted by the concern. The respondents have also been asked if these episodes are in line with how concerns are usually handled by their manager.
The other interview themes are as follows: Has the workload changed over time at the workplace? Do you have time to speak to your colleagues? Is the organisation exposed to competition? Do workers have individualised pay levels? Questions have also been asked about if workers hold back concerns (internally and externally), and if so why? Are there any policies for raising concerns internally and externally? Are critical opinions asked for and is it an obligation or a free choice whether to raise concerns? Has it become easier or harder to raise concerns? If so, why?
Questions have also been asked whether the following characteristics (of the person raising concerns) have any effect on the manager's response: ethnicity, gender, age, type of employment, qualifications and position.
Other questions, regarding climate, with standardised answers are: If workers trust/mistrust managers? If workers are treated in a respectful way or not? How managers communicate (giving orders/dialogue)?
The climate at the workplace (degree of openness, trust and support between colleagues)?

Interpretation and Analysis
The analysis and interpretation of the empirical material has mainly been qualitative. The interviews have been transcribed and the data has been coded into categories in an inductive process (Bryman, 2011) and has for instance resulted in the five groups (A-E) based on how managers react to and deal with concerns raised by workers. For instance, in group A (voice climate), the manager listens to raised concerns from workers and, if possible, takes action to solve problems. Decisions tend to be made through dialogue and workers are involved early in the decision-making process. The main voice promoting factor in group A is a democratic leadership orientation, based on the theory described above. The standardised questions, regarding climate, have been analysed using mean values in order to further analyse differences between the five groups. The quantitative questions have complemented the qualitative main part of the study (Bryman, 2011). For instance, group A has the highest mean values for trust, treated in a respectful way, and supportive climate. The democratic leadership orientation seems to lay the ground for a voice climate based on open dialogue, according to the analysis.

Raising Concerns Internally
In 41 of 90 workplaces, workers report that managers ask for opinions about conditions at the workplace in some way. How raised concerns are reacted upon, and to what extent workers can influence differs widely between the workplaces. From the qualitative analysis of the interviews, five groups of workplaces were identified; A, B, C, D and E. There is no evident pattern of subsectors, or private and public workplaces in the five groups. Group A is most amenable to raised concerns and employee influence. Group A, with the high voice climate, is the only group that reach level 5 (I take part with equal weight) in the Influence Power Continuum (Heller, 1998:159). The difficulty of raising concerns, and gain influence in decision-making processes, increases step by step, and is most difficult in group E, the silence and fear climate. In group D and E workers tend to be on level 2 (I am informed about the matter beforehand) of the Influence Power Continuum (Heller, 1998: 159). In 60 out of 90 workplaces, it is more or less possible to raise concerns, and influence decision-making, where A is the easiest and C is the most difficult. In 30 workplaces workers experience severe difficulties in raising concerns and influence decision making, where D is difficult and group E is most difficult (Author, 2015). The results are summarized in Table 1. The respondents from the healthcare and educational sectors are roughly equally distributed in the groupings A to E. The same holds for gender and age of the manager, and number of workers at the workplace. First the voice and silence groups A-E are presented.
The result and analysis are sandwiched but clearly separated. Group A, High voice climate. Number of workplaces: 17 (18%). Workers do not experience any difficulties raising concerns in these workplaces. Their manager is asking for opinions and justified concerns from workers, and it is normal procedure to raise concerns. Managers involve workers in decision-making processes through open discussions about how to solve problems at the workplace. Hence, the workers are actively involved in the decision-making process. 48hs has an academic position at a university and is also a union representative: We did not receive information about how to clean up the building (from a toxic substance), which we told the prefect at a meeting. At one clean up there was a leakage (the toxic substance) in the stairwell. He (the prefect) called in the people responsible for the clean-up. We also received more information about possible risks from the incident. So, there were remedies…concerns are usually received in a positive manner (by the prefect).
40sj, a nurse at a hospital and union representative: Concerns have been raised about stress. What can we do to get a more stress-free work environment? We have raised these concerns at staff meetings to the manager, who listened, and a project was initiated with a very good results…We aim to work with improvements all the time.
Group B, Moderate voice climate. Number of workplaces: 31 (34%). Managers usually accept raised concerns in general, but they do not actively ask for it, as they do in group A. Managers listen to concerns, but some managers can take a defensive position. In most cases managers deal with problems highlighted by raised concerns if the budget allows it, and if it is a decision that can be taken at their level in the hierarchy. Workers are however involved later in the decisionmaking process compared with group A, and they need to struggle more to initiate changes to problems at the workplace.
A nurse (89pr) working at a healthcare centre: There is not much time to talk with each other and at workplace meetings. If we express opinions the manager takes notes sometimes and says that we will deal with it next time. At the previous development day, we did not get an opportunity to discuss anything. Before the last development day, we demanded the creation of a working group, to decide the content of the development day. She (the manager) accepted this, but if we would not have said something. She wants to run her race (take the decisions herself).
A nurse (17sj) and a union representative in a hospital: We have a constant discussion about an inferior work environment. The department is overcrowded. The manager listens and realises that this is the case. The hospital is going to build a new department, until then they do not do anything. However, we do what we can, it works rather well (at the department)…I would like to say, they (the managers) do not really ask for concerns…There is a cultural history, an hierarchical system… Group C, Restrained voice climate. Number of workplaces: 12 (13%). Managers react negatively when concerns are raised, as opposed to group B. They can for instance yell at workers or make patronising remarks. Managers do not ask for workers' opinions, but some actions are taken to solve problems that workers highlight. The negative reactions from managers make some workers reluctant to raise concerns and others choose to use their union representative as a channel.
27 fs, a pre-school teacher and union representative at a private kindergarten raised concerns about the recruitment process of a new teacher: They (the management) wanted to conduct the interviews themselves. We were not even allowed to see the applications that were sent in. Interviewer: What was the reaction from management? The HR person reacted very negatively, got angry, I got a finger pointed at my face. We wanted to have more influence. In the end, we were allowed to take part in the interviews. It took a long time though, a lot of discussions. 33 g, a teacher and union representative at a public primary school, in a meeting about the psychosocial work environment with the headmaster: She (the headmaster) was very offended, face turned scarlet…She could not under-stand… that people had been insulted (by her)…After a while the headmaster under-stood… (started to change her behaviour).
Group D, Resignation silence climate. Number of workplaces: 12 (13%). The manager often takes a defensive position and can be annoyed, or sad when concerns are put forward. Concerns are in most cases ignored, as opposed to group C. It is therefore very hard for workers to influence the decision-making processes, and many workers give up raising concerns as they are ignored. 38 g, teacher and union representative at a public primary school: The headmaster argues that our staff common room can be used when parents are visiting. We do not agree, the common room is ours… Our arguments are not listened to really, so we feel quite despaired at the moment.
A union representative from a local authority pre-school (26 fs): The computers are not working properly. We have mentioned this regularly during meetings… The headmaster says there is no money. She does not realise we use the computers as much as we do… Interviewer: Where there any consequences for you and your colleagues?
The person who raises concerns can be seen as a troublemaker. The manager (changed) is however fairly polite, but to raise concerns is hard in the long run (as it is not acted upon). Then you stop coming up with suggestions. You stop caring about things.
Group E, Silence and fear climate. Number of workplaces: 18 (20%). Managers react strongly against workers who raise concerns. The most important difference compared with group D is that in group E, workers are harassed when they raise concerns, for instance they are yelled at, receive intimidating remarks, do not receive the same pay rises as their colleagues, or their work tasks are changed in a negative way. In group E workplaces, relations between managers and workers are difficult and harsh, which leads to a hostile climate. According to union representatives, negotiations with management are characterised as hostile confrontations. Lying and hidden agendas are part of the game. The union representatives often need to ask for help from their union in negotiations with the employer. Sometimes workers need to contact the occupational health care or occupational health authorities because of an unacceptable working environment. It is very hard for workers to have influence in the workplace. Many workers are afraid to raise concerns in these workplaces. 87sj, a nurse and union representative in public healthcare: She (the manager) had made a suggestion how the primary healthcare should work with suicidal patients. I asked if it (her suggestion) was evidence based. I knew it was not. I mentioned we had discussed this issue at the local cooperation group (union and employer discussions). Then she got furious at me. This is something that we should deal with here (the manager said). I felt very uncomfortable. Other people have also found her to be intimidating in some situations…Summoned to meeting with the manager… you need to go to the psychologist to be diagnosed. Several people have told me this, some of them have refused…It is a way to make a problem psychologic (the individual's fault) rather than an organisational problem. This creates a widespread fear (among the staff). 32 g, a teacher at a primary school and union representative: Five years ago, the climate was more open. Now all concerns are raised by union rep-resentatives…. Workers who have concerns do not dare to raise them themselves… The management is quite quick to summon people to talks…They can be quite sharp.
Analysis: The five group climates to some extent relate to categories presented by Van Dyne et al. (2003). In my study the focus is on prosocial voice, raised concerns that benefit others or the organisation. Resignation silence climate resembles Acquiscent silence, although the latter resignation also can be a result of lack of low self-efficacy. In my study individual behaviour is not taken into account. The silence and fear climate relates to defensive silence, which is based on self-protection based on fear.

Workplace Climate and the Influence of Leadership Orientation
Result: The workers in group A, with high voice climates, have the highest trust in their manager. Workers' trust in their manager decreases step by step from group A to E (See Table 1). Workers in group A also think they are being treated in a fairly respectful way. In group E, with the silence and fear climate, workers report that their managers treat them in a fairly disrespectful way (See Table 1). The questions related to climate at the workplace follow more or less the same pattern. Workers in group A experience the climate as supportive, open, status free and people do not have any problems with acknowledging that other suggestions and interpretations may be better than their own (See Table 1).
Analysis: The result is in line with previous studies. Voice climates benefit from trustful relations to managers (Bahl & Dadhich, 2011;Donkin, Smith, & Brown, 2008), where workers are treated in a respectful way (Trevino & Weaver, 2001), and a climate that is supportive, open, status free where people do not have any problems with acknowledging that other suggestions and interpretations may be better than their own (Detert & Burris, 2007;Detert & Trevino, 2010;Liu et al., 2015;Morrison, 2014).
Result: The managers in group A ask for opinions and actively involve workers in decision-making. The managers carry through changes in order to solve the problems.
A teacher (68 g) at a secondary school: We have complained about insufficient ventilation, too warm in some places and much too cold in other places. We have mentioned this during workplace meetings and through our union representatives. The headmasters have reacted positively. They have called in people to check the ventilation and temperature. Actions have been taken to improve the ventilation and so on, so there have been improvements.
Interviewer: Where there any consequences for you or your colleagues?
The concerns have been taken seriously. It is normal to raise concerns about conditions at the workplace. Managers listen and find solutions to problems when possible.
Some of the managers in group A actively influence group processes. For example, they deal with conflicts between workers, which probably will be beneficial for people working efficiently together. An example, 29gy, the union representative in a public upper secondary school, group A: The former area manager, she was afraid of conflicts, reacted to everything as an attack on her personally. The new (area manager), bloody hell she is tough, deals with conflicts. We appreciate that. At the same time, she can listen (to us).
Analysis: Asking for opinions, listening, actively involving workers in decision-making, and actively influencing group processes indicate a democratic leadership orientation (Bass & Bass, 2008) This is also similar to the ethical leader who discusses ethics with workers (Bahl & Dadhich, 2011) and the authentic leader who has a strong moral awareness (Liu et al., 2015). The democratic leadership is probably contributing to a high voice climate based on open discussions at the workplace as indicated in group A, which is in line with Bahl andDadhich (2011), Liu et al. (2015), and Morrison (2014), who conclude that employee voice is positively related to leaders who are open to input and discussions, are fair and supportive. Kaptein (2011) uses the concept of discussability which means it is normal practice to openly discuss problems and ethical issues at the workplace. If there is high discussability it is more likely that workers will raise concerns with their managers. Open discussion climate is also often mentioned in connection with the concepts learning organisation (e g. Argyris, 2004) and innovative climate (e g. Ekvall, 1990). An open, prestige less climate where people respect each other is beneficial for development and innovation.
Result: In group E the opposite situation prevails. The managers do not take raised concerns into account and employees are met with reprisals. As a result, a climate of silence and fear is created where only a few people, often union representatives, dare to speak up.
A union representative at a secondary school (3 g) in group E: There have been irregular lay-offs. When we address the employer about this, we are told lies. I think some lay-offs have been illegal…Unfortunately, the laid-off workers are often not union members. Interviewer: Who is raising concerns about conditions at the workplace? Union representatives mostly. They do not dare…There are a few who dare, but not many. Raising concerns can result in very low pay increases compared with other workers. Everybody knows that and it makes people very careful. The headmaster can give people a good talking-to when there are no witnesses. It is very underhand.
Also, in group D decisions are taken solely by managers and raised concerns from workers tend to be ignored. Workers in group D (the resignation silence climate) are however not retaliated against as in group E: A private elderly care home, an assisting nurse (86ä) She (the manager) presents ready-made proposals (in meetings). We can comment but the decisions, most of the time, are taken by herself…We on the night shift mentioned at a staff meeting that we wanted lifts in the corridors, so we did not need to run around and look for them. The manager: Good that you told me. Nothing happened. I have even asked for a protocol from this meeting, but I did not receive it. It has been swept under the carpet. This is a typical example… Analysis: In group E managers seem to have an autocratic (Bass & Bass, 2008) and retaliative leadership orientation, as they do not take raised concerns into account and employees are met with reprisals. This creates a climate of silence and fear (Cassematis & Wortley, 2013;Milliken et al., 2003;Morrison, 2014). Also, in group D an autocratic leadership orientation seems to prevail, as decisions are taken solely by managers and raised concerns from workers tend to be ignored. The difference however, is that workers in group D are not retaliated against as in group E. The D manager behaviour thus resembles what Frazier and Bowler (2015) label supervisor undermining. The manager is not hostile but his/her response to concerns are for instance ignored, workers opinions can be ridiculed, and problems are not dealt with.
Result: In group B and D there are some examples of managers who do not deal with conflicts and bullying, which result in an oppressive climate between colleagues, which in turn can hinder people from having an open discussion and expressing concerns. An example from home care service, a nurse and union representative (81ä) in group D: Not many dare to raise concerns. People are afraid to be seen as whiners. The manager devalues critique as well, like, you know that don't you?... Some people are very quiet. One person is quiet because she is afraid to be bullied by another person (by a colleague. She also felt devalued by the manager and a lack of support from the manager).
Analysis: This kind of leadership has been labelled laissez-faire leadership (Bass & Bass, 2008). The laissez-faire leader is inactive, leaves everything to the workers, does not give support to workers and does not help groups to take decisions. For instance, the laissez-faire leader does not deal with bullying tendencies, which in turn can result in an oppressive and silencing climate.
In group B and C, the leadership orientation is more difficult to categorise. Another research design might shed light on the leadership orientations in these groups. From what respondents are saying about reactions to their concerns, the managers in groups D, E, and partly in C seem to have an orientation towards autocratic leadership. Raised concerns are, in this autocratic tradition, seen as a threat to managers' position and authority. Several studies have presented a similar result as in group E. Whistleblowers are met with retaliation, even if the alleged concern is well-founded (e g. Miceli et al., 2008;Milliken et al., 2003;Morrison, 2014).

Hierarchical Communication
Result: Not following the chain of command can lead to negative reactions from managers and this occur in all five groups (see Table 1). In six workplaces, managers get upset if workers raise a concern directly to a manager on a higher level in the hierarchy.
A nurse and union representative at a big hospital, 19sj in group B: At one occasion we employees wanted to make a change (at the workplace). We contacted the department manager, above our unit manager… Our (unit) manager was very offended.
At 9 other workplaces respondents report that managers do not accept workers contacting their union before raising concerns to the manager. 24 g, a teacher and union representative at a private secondary school in group E: The department manager came to the staff common room and started yelling at the people who happened to be present at the time. The staff had contacted the union, who initiated a health and safety check (due to a perceived work environment problem). She said, why do you not contact me first? Why do you go to a third party?
Analysis: Managers probably see communication to managers further up in the hierarchy, and to other recipients such as union representatives, as a break from the chain of command principle in line with the traditional hierarchical organisation model (Lewis, 2011;Morgan, 2006;Morrison & Milliken, 2000;Near & Miceli, 1995, 1996. According to this traditional view, workers should have an undivided loyalty to the organisation and their superiors (Walters, 1975

Raising Concerns to External Recipients
As mentioned in the introduction, workers in the public sector have freedom of expression protected by the Swedish constitution. They have the right, within certain limits, to raise concerns about issues in their organisation to external recipients, for instance the media. In spite of freedom of expression, respondents report that managers have a negative attitude towards raising concerns externally at 31 (42%) out of the 73 public workplaces. These respondents say that managers, either verbally or in writing, have expressed the view that workers should not, or must not, raise concerns externally. At 16 of the 31 public sector workplaces managers have made it clear that only management should communicate with media. At eight of the 31 public workplaces, managers have told workers not to raise concerns about the workplace to parents or to politicians. At six of the 31 public workplaces and at four private workplaces, managers have a negative attitude towards workers contacting their union (see also above -raising concerns internally), which could be a violation of the legal right of association. At the other 42 (58%) public workplaces, there are no policies, or the respondents have not heard managers expressing negative views on workers communicating concerns about the workplace to external recipients (Author, 2015).

Climates for Raising Concerns Externally
As for concerns raised internally, there is no evident pattern of subsectors, or private and public workplaces regarding raising concerns externally. There are four main silencing factors, according to the analysis, regarding climate to raise concerns about the workplace externally. The first two factors are the same as in the internal section. 1) An autocratic leadership tradition, a characteristic feature mainly in group D and E. 2) Managers having a traditional hierarchical view on communication, whereby communication should take place in a controlled vertical way, following the chain of command.
3) The idea that workers should be loyal to the organisation by not raising concerns to external recipients as this could damage the reputation of the organisation. 4) Organisations exposed to competition. 1 Both managers and workers are aware of the importance of protecting the brand in order to keep their customers.

Autocratic Leadership Orientation and Hierarchical Communication
Result: The most frequently used external recipient of concerns is trade unions. Workers contacting their trade union is both an internal communication, especially when contacting the union representative at the workplace, and an external communication, as the trade union is an organisation that has its central body outside the workplace. Union members and their representatives can contact more central union bodies for help with legal or other advice, negotiation aid, and so on. Especially in group E, where the managers have an autocratic and retaliative orientation, union representatives need to contact and receive assistance from the central bodies of their union. The managers react to union contacts in negative ways, as the two quotes below highlight. 15 fs (group E), a union representative from a private kindergarten: The climate is so hostile, so we have to take help from the union association. It is difficult to be a union representative in such a workplace…The staff are burned, worn out because of a conflict (with the manager). It has escalated so much we have had to hand in a 66A (report about a serious work environment problems) to the Work Environment Authority. We have a work environment that is totally unacceptable… She (the manager) has made statements about our collective incompetence. 69 g, a teacher from a public secondary school in group E: The interviewer: Are there people who do not raise concerns and if so why? Yes, they are afraid of reprisals… A guy had an agreement about his work assignments with the headmaster. The headmaster suddenly made his work less favourable. The teacher would not accept it and contacted the union. The teacher then got summoned to the headmaster and got scolded because he had contacted the union.
Contacting external parties breaks the chain of command. 3 g, the teacher and union representative at the secondary school in group E: When we found out we could not get money for schoolbooks, we wrote a letter to the parents, there is a lack of money. The headmaster got a copy of the letter, also the borough mayor. They got very upset; we had not followed the line (of command) You should go to the headmaster… We thought we had freedom of expression in this country.
According to the autocratic leadership (Bass & Bass, 2008) tradition subordinates should obey the manager and follow instructions. Expressing opposing views or contacting parties, that might have different views and interests than the manager can be seen as a threat to the position of the manager, and to the organisation as a hole, as the manager is part of the authority structure of the organisation (Glazer & Glazer, 1989;Morrison & Milliken, 2000;Near & Miceli, 1995). External communication also breaks the chain of command (Lewis, 2011;Morgan, 2006;Near & Miceli, 1995, 1996. Contacting the union is probably not so much a threat to the reputation and the brand of the organisation, because the union is not likely to go public with the matter, exposing the specific organisation. However, there is always a "risk" that any external communication can have wider dissemination.

Loyalty towards the Organisation
Result: Raising concerns to external recipients can be seen by management as a breach of organisational loyalty, because the reputation of the organisation might be damaged if negative information about the organisation is disclosed to the public. The hospital in the quote below is not exposed to competition. Contacting media is of course also departing from the principle of chain of command. 88sj, a nurse at a big hospital in group E: We should not talk to media, the communication department should do that. Also, we should not talk about our workplace in a derogatory way, for instance on Facebook.
The interviewer: Managers have said that? No, it was a woman from the communication department who informed us… Analysis: Negative exposure in media can be seen by management as a breach of organisational loyalty, because the reputation of the organisation might be damaged if negative information about the organisation is disclosed to the public (Holgersson, 2019;Miceli et al., 2008).

Organisations Exposed to Competition -Protecting the Brand
Result: Problems are concealed both in private and public organisations that are exposed to competition, in order to protect the brand of the organisation. At 35 workplaces, managers have told workers that only management should communicate with the media or politicians, or that workers should not discuss concerns with parents of school children. 30 of these 35 workplaces are exposed to competition, for instance schools.
A teacher, 67 g (group B), at a public primary school: We have something called the (name of the school) spirit…you should be a good representative for the school, an ambassador. The interviewer: Who says that?
The headmasters… Because of loyalty to the school, you do not tell the parents what you think…A question on a parent meeting: Do all pupils really get the support they need (when the classes are so big)? Then you would like to say, no they do not. It is terrible. There are pupils that do not get the support they should have.
Also, teachers use self-censorship to conceal problems at their schools because the schools are competing on a market and good reputation is therefore important in order to attract customers. If the school or care centre receive bad publicity, customers might choose other alternatives. At five public and three private schools, workers report that they practise self-censorship in order to protect the brand of the school. 31gy (group E), a teacher and union representative at a public secondary school regarding teachers have been forced by headmasters to grade courses that have not been running, and to raise grades for pupils: Teachers have raised this concern to the headmaster. The union has not dared to raise these concerns (through the formal partnership agreement), since it will be in the minutes then (public information)…it will have a negative impact on the number of pupils applying to the school. We could lose our jobs.
Furthermore, there are managers who do not report to the police about criminal offences at the workplace in order to avoid public exposure. At four private workplaces, violence and threats of violence towards workers, and assault on and thefts from care receivers were not reported to the police by management.
84ä (group E), a nurse assistant and union representative at a private elderly care home told of an assault by a group leader against a caretaker: We (the staff) feel abandoned. The management do not listen to us. I have sent an e-mail to the department manager to book a meeting. I have not received an answer after four months.
Analysis: Contacting any external recipients about a concern at the workplace is a breach of the line of command (Lewis, 2011) and a threat to the autocratic and hierarchical structure of the organisation (Glazer & Glazer, 1989). It can also damage the reputation of the organisation, thereby counteracting loyalty to the organisation (Holgersson, 2019;Miceli et al., 2008). 4) If the organisation is exposed to competition, as highlighted in the last section, also workers chose not to disclose information that might jeopardise the brand. In part this self-censorship relates partly to prosocial silence (Van Dyne et al., (2003) as it is protecting the organisation. However, it does not protect the interest of their customers/ users and the society. Negative publicity might also lead to workers losing their jobs, which is a self-oriented motivation, not a prosocial motivation. It seems that loyalty towards the organisation tends to be strengthened if the organisation is exposed to competition (Author, 2015).

Concluding Discussion
There are several theoretical contributions from this study. The qualitative analysis showed that concerns from workers are treated by management in different ways. The analysis resulted in five groups of climates regarding concerns raised internally, which is one of the contributions from the study. The workplaces in group A have a high voice climate, where managers ask for concerns from workers, concerns are listened to and remedies are carried out to solve issues, if possible. Workers are involved in decision-making processes at the workplace, reaching level 5 (I take part with equal weight) in the Influence Power Continuum (Heller, 1998:159). Participation in decision-making has not, to my knowledge, been taken into account in previous studies about whistleblowing and voice/silence at work. The workplaces in group B can be said to have a moderate voice climate. Concerns are reluctantly accepted and listened to defensively by managers, who sometimes show irritation. Most of the time, however, managers deal with the problems raised by workers if it is possible. Workers are involved later in the decision-making process compared with group A. In group C, workplaces have a restrained voice climate. Concerns are met with negative reactions from managers such as yelling and making intimidating remarks, which make some workers afraid to raise concerns. Some actions are however taken by management to deal with the issues, thus workers have some influence in their workplaces. Workplaces in group D are characterised by a resignation silence climate. Managers take a defensive position, in some cases get annoyed or sad if concerns are raised by workers. The difference compared to group C is that concerns are often ignored. It is therefore very difficult for workers to influence the decision-making process. As a result, a lot of workers stop raising concerns because they are not listened to. In group E, there is a silence and fear climate. As in group D, concerns are met with negative responses but, the difference compared with D, is that workers are harassed if they raise concerns, for instance they are yelled at, insulted, or their work tasks are changed in a negative way. The climate between workers and management is very hostile and workers are therefore afraid of raising concerns. In group D and E workers tend to merely reach level 2 (I am informed about the matter beforehand) of the Influence Power Continuum (Heller, 1998: 159).
A number of voice and silence factors are creating these different climates. A democratic leadership orientation (Bass & Bass, 2008) found in group A lays the foundation for a high voice climate in the workplace based on trust, good relations, open dialogue between colleagues and with the manager (e g. Ekvall, 1990;Morrison, 2014). Research about whistleblowing and silence/voice has not taken democratic leadership into account before. The democratic leader (Bass & Bass, 2008), listens to concerns, carries through changes, deals with group processes such as conflicts and bullying, as highlighted by respondents in group A. The democratic leadership is therefore an important explanation to how high voice climates are created. The active shaping, by the manager, of an open climate based on trustful and respectful relation lays the ground for high voice climates. The respondents in group A have reported high level of trust and that they are treated in a respectful way by their managers. These findings are in line with previous studies. Workers who have a high level of trust in their manager are more likely to raise concerns internally according to Donkin, Smith, and Brown (2008). Also positively related to internal whistleblowing are managers who listen and discuss issues with workers (Bahl & Dadhich, 2011;Liu et al., 2015;Morrison, 2014;Tangirala & Ramunajam, 2012) and treating workers in respectful way (Trevino & Weaver, 2001). These findings are also in line with prosocial organisational behaviour (Cassematis & Wortley, 2013). Furthermore, the democratic leader is associated with organic organisations (Burns & Stalker, 1961) and learning organisations (e g. Argyris, 2004) that promote innovation and questioning taken for granted assumptions in the organisation.
A number of silencing factors have emerged in the study, most of them confirming previous studies. The study points to an autocratic leadership orientation in group D and E. In group E managers are also retaliating against workers who raise concerns, which is a difference compared to group D. Autocratic leaders are associated with bureaucratic organisations (e g. Morgan, 2006), and see criticism as an attack on, and a threat to their position (Glazer & Glazer, 1989;Near & Miceli, 1996). The autocratic manager expects workers to follow orders without questioning, thus, to have undivided loyalty to the manager. A lot of people are afraid to raise concerns in group E and to some extent in group D. The autocratic leadership orientation in group D is contributing to a resignation silence climate. The autocratic and retaliative leadership orientation in group E, what Morrison (2014) refer to as abusive leadership, creates a climate of silence and fear. There are also signs of a laissez-faire leadership orientation in group B and D. There are indications that also laissez-faire leadership, another contribution from this study, can have a silencing effect on the workplace climate, as the manager does not deal with bullying and conflicts between workers. Some workers are afraid of raising concerns if they have colleagues with a bullying tendency, as the passive laissez-faire manager leaves everything to workers to manage themselves. These findings are in line with previous studies that have concluded that a climate that tolerates harassment allows the harassment to continue (Miceli et al., 2008).
A hierarchical view on communication is a silence factor. A traditional hierarchical structure and culture are deeply rooted in the history of many organisations. According to this traditional bureaucratic view, communication should take place in a controlled vertical way, following the chain of command (e g. Lewis, 2011;Morgan, 2006;Near & Miceli, 1995, 1996, Morrison & Milliken, 2000. A breach from this path causes negative reactions if for instance a manager is "bypassed" by workers who raise concerns to a manager higher up in the hierarchy. Negative reactions from managers or politicians, when breaking the chain of command, occur in all five climate groups. The culture in most large and old organisations is often based on the traditional hierarchy. However, as Detert and Trevino (2010) point out, some managers in the organisation can have a more open view of communication and aim to create a voice climate in their unit, in contrast to the predominant hierarchical culture in the rest of the organisation. This pattern can be seen in group A with the high voice climate.
The findings on climates for raising concerns externally have confirmed previous studies about the first three out of the four silencing factors found in this study. The first two factors are the same as the climate for internal concerns. 1) An autocratic leadership orientation (Bass & Bass, 2008). 2) Managers having a traditional hierarchical view on communication, where communication should follow the chain of command (e g. Lewis, 2011;Morgan, 2006). 3) The idea that workers should be loyal to the organisation in order not to do anything to damage the reputation of the organisation (e g. Miceli et al., 2008). 4) If the organisation is exposed to competition. Both managers and workers are aware of the importance of protecting the brand (Author, 2015).
A maximum silence climate for external concerns is present when these four silencers are combined in group E. External disclosure of information tends to trigger very negative reactions from managers in group E, who have an autocratic and retaliative leadership orientation. Although they do not explicitly mention autocratic leadership, it is in line with Glazer and Glazer (1989) who describe the manager as part of the authority structure of the organisation. Questioning decisions and raising concerns are seen as breaking the hierarchical order where managers lead and workers are expected merely to follow instructions (Glazer & Glazer, 1989). Disclosing concerns externally to for instance authorities, media, politicians, trade unions, or pupils' parents can be seen by managers as a breach of the organisation principle of chain of command (e g. Lewis, 2011;Near & Miceli, 1995, 1996. There are many examples of a negative attitude from managers, across the five groups, for workers to raise concerns with external recipients, partly because it breaks the chain of command. Furthermore, raising concerns externally can damage the reputation of the organisation (Holgersson, 2019;Miceli et al., 2008). If the organisation is exposed to competition, the external disclosure of negative information can result in losing customers such as patients or pupils. Many managers have therefore told workers not to disclose negative information to media, politicians or parents of schoolchildren. Some workers, who work in organisations exposed to competition, even use self-censorship when communicating with people outside the organisation in order to protect the brand. The self-censorship can be seen as prosocial silence (Van Dyne et al., 2003) towards the organisation. Thus, when organisations are exposed to competition the organisational loyalty tends to be strengthened and this combination has a silencing effect on raising concerns externally, which is a contribution from this study.

Limitations of the Study and Suggestions for Future Research
Only one person per workplace has been interviewed and people might not always remember the events as they happened. Other workers at the workplace might have other views about the subject. Furthermore, the interpretation about leadership orientation is made from an employee perspective, which does not take into account how managers perceive situations.
The sample of workplaces cannot be described as representative of the two sectors. First of all, the number of workplaces is far too small, secondly in this mainly qualitative study, the aim is not to provide a result that represents the two sectors, rather it is to build new theory on how concerns are received in different ways and how these differences in responses can be explained. However, the workplaces have been selected with the aim to include workplaces from the main subsectors in the two sectors.
Future research could use the five climates as a framework to analyse voice and silence climates. Furthermore, studies could test the suggested theory presented in this paper of the link between democratic leadership and high voice climate, between autocratic leadership and resignation silence climate, and retaliative autocratic leadership and silence and fear climate. As for concerns raised to recipients outside the organisation, future studies could test the theory of the four silencing factors suggested to have a reinforcing effect if combined -(1) autocratic leadership, (2) managers having an hierarchical view on communication (to follow the line of command), (3) organisational loyalty, strengthened if the (4) organisation is exposed to competition.
We also need to know more about the organisational factors that promote and inhibit raising concerns. More qualitative studies are needed. These qualitative studies could take a leadership perspective: how managers perceive raised concerns from workers and their reasons for how they react. New organisation models (Author, 2018), such as New Public Management and Lean Management, tend to inhibit employee voice, increase demands on workers and lead to intensification of work. Furthermore, outsourcing with several layers of suppliers, the decreasing number of trade union members, increasing use of temporary workers, and contract workers in the new gig economy also need to be studied with qualitative methods. How do these new ways of organising work affect workers' possibility to raise concerns and influence decision-making processes in organisations? Other examples are union representatives' experiences and differences between branches.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.