Rally-around-the-organizational-flag?Internalcommunication inaprofessionalorganization duringtheCovid-19 pandemic crisis

Purpose – This study aims, first, to explore and analyze if and how organizational members ’ professions or occupations influence perceptions of internal crisis communication. The second, related, aim is to discuss the roleofinternalcommunicationincreatingastrongorganizationalidentityduringaprolongedcrisissuchastheCovid-19pandemic. Design/methodology/approach – Thisstudyismainlyconceptualbutusesquantitativedatafromasurvey conducted in a health-care organization in late 2020 to illustrate the theoretical reasoning. Findings – The results show that the administrative groups perceive factors in the internal crisis communicationmorefavorablythantheprofessionalgroups.Thestudysuggeststhatorganizationalmembersperceiveinternalcrisiscommunicationdifferentlydependingonwhichintra-organizationalgrouptheybelongto.Thisfurtherpointstotheabsenceofa “ rally-around-the-flag ” effect and highlights the importance of working proactively with professionals and in internal crisis communication. Originality/value – This study highlights the role of professionals in crisis communication, which is an aspect that so far has been ignored. The internal professionalization processes and an intriguing power struggle between professions have obvious consequences for crisis communication. As shown in the overview of earlier research on internal communication, leadership and professional organizations, the prerequisites for creating an increased organizational unity among coworkers are challenging. The idea that a crisis may, as in certain political situations in society, create a “ rally-around-the-flag ” effect is still relevant, even if the case study is an example of how this did not happen.


Introduction
In the field of crisis communication, researchers have for a long time discussed and analyzed various aspects that may influence the practice and effects of communication in a crisis situation.Examples of such aspects are risk society (Beck, 1992), context (Coombs, 2016), complexity (Gilpin andMurphy, 2006, 2008), groupthink (Sellnow and Seeger, 2013), reputation (Benoit, 1997;Ware and Linkugel, 1973), philosophy of science (Falkheimer and Heide, 2010) and national culture (Zhao, 2017).However, one aspect that has been, more or less, ignored in crisis communication research is professions.One exception is a recent article by Deverell (2021) that focuses on the role of communication professionals in crisis communication.In many organizations, there are struggles and tensions between an administrative elite with managers and specialists on the one hand, and coworkers belonging to strong professions on the other hand.Researchers such as Abbott (2010) emphasize that professions exist since occupations have an ability to acquire control of specific work areas.Hence, in almost all larger organizations, there is a battle between various professions in order to reach social approval to define and control their work independently of others (cf.Pieczka and L'Etang, 2001).The process of professionalization is materialized by communicative strategies and rhetoric to increase status, salary conditions and power positions (Evans, 2008).This implies that interprofessional competition defines the status and power of professions (Abbott, 1986;Alvesson, 2013).One form of organization that stands out when it comes to struggles between professions is the one that Mintzberg (1980) calls the professional organization.Examples of such organizations are hospitals, universities and pharmaceutical companies.Professional organizations are characterized by so-called professionals, i.e. highly trained specialists with considerable work autonomy which means that both the formal and informal power can be found in the operating core (Alvesson, 2004;Mintzberg, 1999).Professionals belong to an occupation that has a license from specific education and training, such as physicians and lawyers and they are regarded as strong professions.In other words, the professionals are in a situation of strong power in relation to the administrative elite and to members of semi-professions such as economists, communication officers and HR strategists.
But what happens with the struggle between professions in an organization during a crisis.Could it be that a crisis like the current global Covid-19 pandemic reduces barriers between different professionals and opens the way for enhanced unity and increased organizational identity?Will a crisis lead to a sudden common pride and bridge the gap between the struggling professions?Is there an internal rally-around-the-flag effect due to a crisis such as the pandemic?The expression "rally-around-the-flag-effect" was coined by political scientist Mueller (1970) and describes instant (and often short-term) popular support for the national government or political leaders during an international crisis or war.This crisis effect has been the focus of many studies in political science and international relations since then and has later been nuanced when empirical research showed that this effect is more complex than first anticipated, differing depending on several groups and individual variables (e.g.Johansson et al., 2021).Still, is it a relevant assumption that this effect can also be found within organizations?If so, internal strategic communication would obviously be a crucial driver of such an effect.Internal communication has a vital role in creating trust and organizational commitment in "normal" situations, but even more so in turbulent situations (David, 2011;Heide and Simonsson, 2020).
The aim of the study is to explore and analyze if and how organizational members' professions or occupations influence perceptions of internal crisis communication.The second, related, aim is to discuss the role of internal communication in creating a strong organizational identity during a prolonged crisis such as the Covid-19 pandemic.Although the article is mainly conceptual, we use data from a quantitative study in a health-care organization to investigate the influence of organizational members' professions or occupations on their perceptions of internal crisis communication.The article contributes to research on how different professional groups perceive internal communications and the possible consequences this has for strategy and execution, especially during a crisis.
As mentioned previously, different contextual and background variables, factors or aspects for crisis communication perceptions and outcomes have been the focus of earlier research, but there is a lack of research focusing on the role of professions.The empirical case is based on a quantitative survey in a large health-care organization in Sweden.The survey was conducted in late 2020 when the organization was in an emergency situation due to the intensity of Covid-19.The survey was still answered by 292 respondents.
After this introduction follows an overview of theories and conceptual issues concerning internal communication, crisis leadership and communication and professional organizations.The next sections include a presentation of the method and the results from the empirical study, ending with a concluding discussion.

Internal crisis communication and crisis leadership in professional organizations
The global Covid-19 pandemic is more complex than traditional crises, and it may last for several years (cf.Ansell et al., 2010).For most people, the pandemic is probably interpreted both as a risk and a crisis.Boin et al. (2020) conclude that the pandemic is an example of a newer type of crisis, the "creeping crisis": A creeping crisis is a threat to widely shared societal values or life-sustaining systems that evolves over time and space, is foreshadowed by precursor events, subject to varying degrees of political and/or societal attention, and impartially or insufficiently addressed by authorities (p.122).
A creeping crisis evolves slowly and exists for a very long time.This crisis is not formally recognized until rather late and the development in time is blurred.Suddenly, there are disruptions and then the crisis slows down again.The creeping crisis is developed as a societal crisis concept, but it is relevant also for internal crisis communication.Even if Covid-19 has not implied a crisis for all organizations, it has often brought several organizational changes in practices, such as remote work or temporary lock-downs, and affected internal organizational processes (Boin, 2019).Translating the creeping crisis into an organizational level means that this crisis threatens the overall organizational value system (the culture) and is problematic to manage due to maximal uncertainty since it goes back and forth with no clear end and consequences that are hard to understand.The characteristics of a creeping crisis are not newuncertainty and ambiguity are typical features of contemporary late modernity in both society and organizationsbut the pandemic makes these features even clearer.
The massive complexity of the Covid-19 pandemic raises new demands for leadership.Tourish (2020) even argues that the pandemic has also uncovered a crisis in leadership theory and practice.He claims that mainstream leadership theories, e.g.transformative, authentic leadership, overemphasize the role of leaders as persons and encourage leaders to concentrate even more power in their own hands.Likewise, followers or coworkers expect leaders to make decisions and to know how to handle a situation, especially in turbulent times.According to Tourish (2020), this kind of leadership might work in less complex crisis situations, but we need another kind of leadership to cope with the radical uncertainty we have been facing in the Covid-19 crisis.Along similar lines, several other researchers have called for a more nuanced and multifaceted notion of crisis leadership (Heide and Simonsson, 2019;Muffet-Willett and Kruse, 2009;Snowden and Boone, 2007).Grint (2005Grint ( , 2020) ) makes the distinction between critical, tame and wicked problems to better understand the role of leadership in different crisis situations.Critical problems are acute, intense situations where there is a need for a commander that makes fast decisions and coerces employees, citizens and other "followers" to take certain actions in order to avoid catastrophes.Tame problems are often less acute, but more complicated than critical ones.Even so, tame problems are often problems that we have experienced before and they can be solved by implementing a standard operating procedure, i.e. this is the domain management.One example of a tame problem is how to organize vaccination plans for a nation's population.Wicked problems are highly complex problems without a clear beginning and end, no given solution and when one solution is implemented, new problems are often created (Grint, 2005).Wicked problems require leadershipthe leader's role is to ask the right questions rather than give the right answers.The complexity of wicked problems also calls for leaders who can build and foster collaboration.In situations when we face wicked problems, leaders need to mobilize the community and reinforce the collective struggle (Grint, 2020).The Covid-19 crisis is certainly a wicked problemthere is no clear end to it and different "solutions" tend to counteract each other.At the same time, the pandemic has also involved both critical and tame problems, which have called for a variety of leadership roles.However, on the whole, the pandemic is a wicked problem which calls for a less top-down oriented and a more collaborative, participative leadership approachfocusing on listening and asking questions rather than providing answers and instructions as well as building trustful relationships and facilitating interaction between different groups and functions that need to collaborate to solve the crisis.Muffet-Willett and Kruse (2009) argue that leadership approaches focusing on leaders as persons are especially inadequate in organizations where employees have power and are expected to make and carry out independent decisions in a crisisi.e. a context which is highly valid in a professional organization.
Increasingly complex crises and the need for more collaborative leadership are also related to a broader, process-oriented perspective on internal crisis communication.Traditionally, internal crisis communication has, in practice, been a question of information transmission of messages from the top management to employees via various channels such as the intranet or by using line managers as an intermediary channel (Heide and Simonsson, 2021).Providing employees with quick and accurate instructions during the acute phase is often important to coordinate actions and prevent harm.However, reducing internal communication to a "message service" is an all too narrow understanding of the practice, which will turn internal crisis communication into a solely reactive and rather short-term process.Recent research (Heide and Simonsson, 2021;Mazzei, 2010Mazzei, , 2014;;Vigsø and Strandberg, 2016) rather emphasizes that internal crisis communication has a much broader and more complex role, namely, to facilitate sensemaking of ambiguous situations and enabling coworkers to take an active communication role, both internally and externally.
As crises are often unexpected situations we have never experienced before, they often imply ambiguity, confusion and feelings of disorientation (Maitlis and Sonenshein, 2010).Crises can thus easily lead to a collapse in our understanding of reality and tend to trigger an active and intense search for meaning.It might be that organization members need more information to reduce uncertainty but quite often they also need dialogue and conversations with managers and colleagues to guide their interpretations and reduce ambiguity (Weick, 1995).
Coworkers have traditionally been perceived as receivers of management communication in times of crisis, and it has been more or less neglected that they also have an active communication role in a crisis.Coworkers rather than managers are those who work at the frontline and who have direct contact with a great number of stakeholders during a crisis.How they respond to questions and the image they present of the management and the organization will certainly influence the trust and reputation of the organization (David, 2011;Frandsen and Johansen, 2011).Thus, it seems reasonable to consider coworkers as senders and active communicators in the sense that they have information and experiences that may help managers to make informed decisions.Horizontal communicationwhen coworkers within and between different units communicate with each other-is also vital.In line with the above-mentioned emphasis on dialogue to reduce ambiguity, Vigsø and Strandberg (2016) conclude that "interpretation is not created when the information is presented, but in the interaction afterwards among those whom the management considered to be the 'receivers' of information" (p.95).
Hence, the traditional focus of internal crisis communication on providing instructions through effective channels needs to be expanded to build employee trust and loyalty, especially during a prolonged crisis such as the Covid-19 pandemic.In a recent study on the role of internal communication during Covid-19, Einwiller et al. (2021) found that an informational communication strategy (focusing on instructions, effects of the crisis on the organization and employees' work situation) influences acceptance of management decisions.
But in order to enhance employees' organizational commitment, a relational communication strategyfocusing on dialogue, participation and expressing appreciationis required.Einwiller, Ruppel and Stranzl (2021) further argue that appreciation needs to come early in the crisis and that the most valued recognition comes from top management.In line with the emphasis on collaborative leadership and relational communication, previous research has also underlined the importance of listening to build trust and engagement among employees (Macnamara, 2021;Neill and Bowen, 2021).To realize the potential of employee engagement, Ruck et al. (2017) argue that it is important to both encourage and facilitate employee voice and to build senior managers' listening capacity.
Professional organizations and professionals Mintzberg (1980), one of the most influential thinkers within the strategy, differentiates between five different forms of organization structures: entrepreneurial organization, machine organization (bureaucracy), professional organization, divisional organization and the innovative organization ("adhocracy").In this study, we focus on a professional organization that is interesting to study because it is a fairly common organization type, not least when it comes to public organizations.It is also interesting from a strategic communication perspective that there are many tensions between different professional groups within the organization.Typical for a professional organization is a large amount of knowledge workers, such as lawyers, teachers, nurses, physicians, IT specialists or researchers.They are sometimes called gold collar workers since the professionals have a high status and are well-paid (Alvesson, 2004).This form of organization is sometimes also called knowledge-intensive organizations while the majority of the activities are based on intellectual skills and the professionals have typically a longer academic education (Alvesson, 2004).The professional organization is relatively similar to the machine organization, which is characterized by standardization, centralized decision-making and tasks that are structured by functional departments.The professional organization is also highly bureaucratic, but there is also one key difference between these two organization structures.In professional organizations, professionals who are highly trained and educated demand independency in control of their work.Accordingly, decision-making is decentralized as an effect of the high level of specialization.Yet another characteristic is complexity, as there are many rules and procedures in a professional organization.The professional organization has the same possibility of efficiency as the machine organization, but the professionals have more autonomy and power compared to coworkers in a machine bureaucracy.Consequently, the professionals have considerable power in professional organizations which weakens the executive's possibility to control and influence.This makes, in turn, these organizations difficult to change.
The structure of a professional organization produces a built-in power tension between the administrative elite and the professionals.Earlier research (e.g.Alvesson, 2001;Starbuck, 1992) confirms that this tension is especially common in professional organizations such as hospitals and universities, where the professionals have their own agenda, goals, identity and specific knowledge.Even if this tension and power struggle is especially evident in professional organizations, they do also occur in other organizations (Pieczka and L'Etang, 2001).The goal of such a power struggle for professionals is to obtain social approval which will make it possible for them to independently define and control their work.The power dimension is an important characteristic of professions (Abbott, 2010), and competition between professions defines the status and power of a profession (Abbott, 1986).In organizations, the power game between professions and the administrative elite emerges through communicative strategies and rhetoric with the aim of gaining increased status and power positions (Evans, 2008).While there is a constant strive among professionals to strengthen their occupational power in an organization, there is also a counterforce in organizational professionalism that is driven by the administrative elite who wish to increase Internal communication their ability to control, motivate and regulate the professionals.This form of organizational professionalism is based on organizational and administrative logics rather than on occupational logics (Evetts, 2009(Evetts, , 2013)), and it is possible to argue that this form of professionalization is even more common in professional organizations where management may feel the need to strengthen their legitimacy and power.Hence, organizational professionalism is often initiated through the fear that professionals have too much power and control of the organization (Alvehus et al., 2020).

Method
As stated in the introduction, the primary aim of the study is to explore and analyze if and how organizational members' professions or occupations influence perceptions of internal crisis communication.We use data from a quantitative study in a health-care organization to exemplify the influence of organizational members' occupations on their perceptions of internal crisis communication.For the purpose of investigating the relationship between occupations and the perception of internal crisis communication, a quantitative survey was considered appropriate because it provided an efficient and cost-effective way to collect data.

Sample and procedure
The sample comprised organizational members working in a large health-care organization in Sweden.The sample selection aimed to cover the different occupations existing within the organization and included the following seven occupations: physicians, nurses/midwives, assistant nurses, analysts, administrators, managers and others.Simple random sampling was used to obtain a representative sample.
The data were collected through a quantitative online survey conducted by a private research company with which the authors are affiliated.The authors were involved in and monitored the whole data collection process to ensure compliance with research ethical issues related to outsourcing, such as anonymity and confidentiality (Allen and Roberts, 2010).The survey was conducted between November and December 2020.It was sent out to 1,315 organizational members, of which 294 completed the survey, resulting in a response rate of 22.4%.The response rate for each targeted occupation was: physicians (n 5 52, 17.7%), nurses/midwives (n 5 35, 11.9%), assistant nurses (n 5 33, 11.2%), analysts (n 5 33, 11.2%), administrators (n 5 51, 17.3%), managers (n 5 56, 19%) and others (n 5 34, 1.6%).As the workload of the physicians, nurses and assistant nurses was exceptionally heavy during the time of the study due to the "second wave" of the Covid-19 pandemic, the final response rate was considered sufficient under the circumstances.

Measures
Given the lack of previous studies investigating strategic crisis communication during the Covid-19 pandemic at the time of the study, the survey was developed from scratch.However, previous research on crisis communication was consulted, and to the extent possible, the items were inspired by existing research.The final survey consisted of 30 items and covered different topics related to internal crisis communication such as crisis communication channels and information dissemination, immediate supervisor crisis communication and top management crisis communication.Items were measured on a five-point Likert scale ranging from strongly disagree to strongly agree.
Since no existing scales were used, an exploratory factor analysis (EFA) was initially conducted on the total set of items within each topic to identify underlying patterns in the data and for data reduction purposes.The EFA resulted in four factors which were also tested for internal consistency to ensure that the patterns identified in the EFA were suitable for generating summative scales.The final scales were Perception of immediate supervisor crisis communication (α 5 0.93) consisting of five items, Perception of top management crisis communication (α 5 0.91) consisting of four items, Perception of crisis preparedness (α 5 88) and Perception of internal crisis communication (α 5 0.81).The items for each scale are shown in Table 1.To facilitate interpretation in the results section, each summative scale was divided by the number of items included in the scale so that each scale had the same five-point (1-5) range as the individual items.The top management has been visible in the internal channels during the Covid-19 pandemic The top management has listened to us coworkers, i.e. taken our concerns seriously The top management has shown coworkers appreciation during the Covid-19 pandemic The top management's way of communicating during the crisis has been trustworthy Perception of crisis preparedness (α 5 88) We had a lively discussion in my organization about security and crisis issues even before the Covid-19 pandemic Our organization was well prepared to communicate internally in a crisis situation Our organization was well prepared to communicate with customers, citizens and other external groups in a crisis situation In our organization, it already existed a good preparedness for crises even before the Covid-19Covid-19 pandemic There was already a strong trust between management, managers and coworkers Perception of internal crisis communication (α 5 0.81)

Finding up-to-date information about what applies in our own organization has been easy
There have been a lot of rumors circulating in our organization during the Covid-19Covid-19 pandemic The internal communication has been difficult to understand I have trusted the information that has been available in our internal channels The internal communication has presented a fair picture of reality The internal information dissemination has been too slow I have had a sufficient mandate during the crisis to be able to quickly solve problems in my daily work The internal communication has contributed to a sense of pride in me Table 1.

Scales Internal communication
Analytical procedure Data analysis was conducted using SPSS Statistics 26.0 for Mac.A p < 0.05 significance level was used for all statistical tests performed.To explore the influence of organizational occupation on organizational members' perception of (1) the immediate supervisor crisis communication, (2) the top management crisis communication, (3) the crisis preparedness and (4) the internal crisis communication, a one-way analysis of variance was performed to compare the variance between the different groups with the variance within the groups, and a follow-up post-hoc test was performed to identify which of the groups differed.Levene's test for equality of variance was consulted to determine if the data met the assumptions of homogeneity of variance necessary for performing an analysis of variance.

Descriptive statistics
Before data analysis, the data were assessed for normality and checked for missing values.A Kolmogorov-Smirnov test indicated that the dependent variables were not normally distributed: the immediate supervisor crisis communication D (231) 5 0.82, p 5 0.000, the top management crisis communication D (210) 5 0.96, p 5 0.000, the crisis preparedness D (188) 5 0.97, p 5 0.000 and the internal crisis communication D (182) 5 0.96, p 5 0.000.However, as the analysis of variance is very robust to violations of the underlying assumptions when the sample size is reasonably large (ca 30 þ) (Pallant, 2013;Reinard, 2006), no further actions were taken.
The dependent variable occupation contained no missing values.Missing data for the four dependent variables ranged between 21.4% and 38.1% and were therefore substantial enough to warrant action (Hair et al., 2014).Expectation-maximization algorithm was used to diagnose the pattern of missing data.The result of Little's MCAR test was not significant (x 2 5 32.51, p 5 0.25), indicating that the missing data were missing completely at random (MCAR).In the following analysis, pairwise deletion was performed to ensure only valid data were included.

Results
Four one-way between-group analysis of variance was conducted to explore the influence of occupation on organizational members' perception of (1) the immediate supervisor crisis communication, (2) the top management crisis communication, (3) the crisis preparedness and (4) the internal crisis communication.Respondents were divided into seven groups depending on their occupation (physicians, nurses/midwives, assistant nurses, analysts, administrators, managers and others).
For the first analysis exploring the influence of occupation on organizational members' perception of the immediate supervisor crisis communication, the data violated the assumptions of homogeneity of variances (p < 0.004), so a Welch's ANOVA was used instead.There was a statistically significant difference between the groups as determined by Welch's ANOVA (F (6, 78.25) 5 4.3, p 5 0.001).The effect size, calculated using eta squared, was 0.08, indicating a medium effect size.Post-hoc comparisons using the Games-Howell Post-hoc test indicated that the mean score for managers (M 5 4.51, SD 5 0.58) was statistically significantly different from physicians (M 5 3.81, SD 5 1.23).No other statistically significant differences were identified between the organizational roles in relation to their perception of the immediate supervisor crisis communication.
For the second analysis exploring the influence of occupation on organizational members' perception of the top management crisis communication, the data violated the assumptions of homogeneity of variances (p < 0.010), so Welch's ANOVA was used instead.There was a statistically significant difference between the groups as determined by Welch's ANOVA (F (6, 72.17) 5 9.9, p 5 0.000).The effect size, calculated using eta squared, was 0.22, indicating a large effect size.Post-hoc comparisons using the Games-Howell post-hoc test indicated that the mean score for managers (M 5 3.95, SD 5 0.863) was statistically significantly different from physicians (M 5 2.62, SD 5 1.11), nurses/midwives (M 5 2.87, SD 5 1.05), assistant nurses (M 5 2.69, SD 5 1.33) and analysts (M 5 2.98, SD 5 0.821).Furthermore, the Games-Howell post-hoc test indicated that the mean score for administrators (M 5 3.69, SD 5 0.873) was statistically significantly different from physicians (M 5 2.62, SD 5 1.11) and nurses/midwives (M 5 2.87, SD 5 1.05).No other statistically significant differences were identified between the occupations in relation to their perception of the top management crisis communication.
For the third analysis exploring the influence of occupation on organizational members' perception of crisis preparedness, the data violated the assumptions of homogeneity of variances (p < 0.006), so Welch's ANOVA was used instead.There was a statistically significant difference between the groups as determined by Welch's ANOVA (F (6, 66.05) 5 3.9, p 5 0.002).The effect size, calculated using eta squared, was 0.10, indicating a medium effect size.Post-hoc comparisons using the Games-Howell post-hoc test indicated that the mean score for managers (M 5 3.71, SD 5 0.737) was statistically significantly different from physicians (M 5 2.80, SD 5 1.02).No other statistically significant differences were identified between the occupations in relation to their perception of crisis preparedness.
For the fourth analysis exploring the influence of occupation on organizational members' perception of the internal crisis communication, the data violated the assumptions of homogeneity of variances (p < 0.047), so Welch's ANOVA was used instead.There was a statistically significant difference between the groups as determined by Welch's ANOVA (F (6, 58.59) 5 7.5, p 5 0.000).The effect size, calculated using eta squared, was 0.18, indicating a large effect size.Post-hoc comparisons using the Games-Howell post-hoc test indicated that the mean score for managers (M 5 4.08, SD 5 0.502) was statistically significantly different from physicians (M 5 3.26, SD 5 0.760).Furthermore, the Games-Howell post-hoc test indicated that the mean score for administrators (M 5 3.69, SD 5 0.873) was statistically significantly different from physicians (M 5 3.26, SD 5 0.760).No other statistically significant differences were identified between the occupations in relation to their perception of internal crisis communication.

Analysis
One of the scales measured perceptions of crisis preparednesswhich includes items about the organization's preparations for external and internal crisis communication but also items concerning internal trust as well as the existence of discussions on security and crisis issues before the crisis.The results from the survey revealed that managers are more positive than physicians with regard to how things worked before the pandemic.The results from the other scales measuring perceptions of leadership and internal communication during the crisis indicate that this gap remained.
On a general level, the respondents display a rather positive perception of the immediate manager during the crisis.In a previous survey study of internal communication in eleven organizations in Sweden, we found that employees tend to be quite satisfied with their immediate manager (Heide et al., 2019).The results from the study at hand, conducted in a situation when the organization was under great pressure, show that employees are also relatively satisfied with their immediate manager during the Covid-19 crisis.However, the results also demonstrate variations linked to different professional rolesphysicians are less satisfied with their immediate manager in comparison to the group of respondents being in a management position themselves.This is not especially surprisingif you are a manager yourself you have a better understanding of the challenges of being a manager and you may identify more with your own manager.Furthermore, as discussed above, employees in professional organizations tend to demand and have more autonomy and power in comparison to other organizations.Physicians, who are at the top of the professional status hierarchy in the organization, tend to identify more with peers and, hence, value information and support from peer colleagues rather than their immediate manager.
Turning to the perceptions of top management and its communication, the results indicate that the respondents in general are less satisfied with the top management than the immediate supervisor.On the one hand, this may be expected, as senior management has much fewer possibilities to have direct contact and build a relationship with the employees in a large organization.On the other hand, a crisis is a situation where the top management is often more in the limelight and receives more attention than in "regular" situations.A crisis is certainly challenging for top management, but just as for political leaders acting on a societal level, it may also be an opportunity to strengthen trust and build unity.But, as mentioned, the top management communication gets a rather low value (compared to other results).Again, we also find that respondents being managers themselves are most positive not only in comparison to physicians but also to assistant nurses, nurses/midwives and analysts.Also, administrators are more positive to top management than physicians and nurses/midwives.Thus, in line with previous research, there is a clear gap between the professional groups and staff belonging to the administrative elite.This study does not show whether these gaps have increased or diminished during the pandemic, but the results suggest that the crisis situation has not led to a strong and joint rally around the top management.
The scale measuring top management includes items concerning employees' view of the top management's visibility, listening, expression of appreciation and trustworthiness.All of these can be defined as relational communication, which Einwiller et al. (2021) in a study during the Covid-19 crisis found vital for strengthening employees' emotional bond and engagement.The rather low levels of appreciation of top management found in this study indicate that senior managers have not focused on this kind of communicationat least not sufficiently enough to build employee support among the professional staff who are the most important trust builders as they meet and communicate with stakeholders on a daily basis.We have argued above that different types of crises or different phases of a crisis require different kinds of crisis leadership.As the organization in this study is a health-care organization, there have certainly been several acute phases during the prolonged time period of the pandemic where a more authoritative leadership and informational communication approach has been needed.However, the long durability of the crisis in combination with the high degree of complexity and ambiguity means that managers also need to focus on practicing collaborative, participatory leadership.Given the demonstrated gaps between the administrative elite and the professional staff, it seems that top management should have focused more on both building relationships and trust among all employee groups but also on facilitating collaboration and communication between different groups of employees.
The results of the study showed that the administrative staff (managers and administrators) perceived the internal crisis communication more favorably than physicians who, as previously mentioned, are at the top of the professional status hierarchy in the organization.This finding is in line with previous research on professional organizations highlighting that health-care organizations such as hospitals are driven by medical logics emanating from the norms and values of the medical staff (Heide and Simonsson, 2015).In their study, Heide and Simonsson found that the dominance of the medical logics, that the medical staff adheres to, over the managerial logics, that managers and administrators adhere to, tends to cause a separation, and at times even conflict, between these professional groups.While it is beyond this study to draw conclusions regarding a potential conflict influencing how the groups in our study respond, the results indicate a clear difference between how the management and administrative staff and the professional core staff have perceived the internal communication during the crisis.
The identified difference between how the management and administrative staff and the professional staff group perceived the internal crisis communication is perhaps not that surprising given that professional groups tend to identify with and focus more on their work rather than issues related to the organization as a whole (Heide and Simonsson, 2015).In their recent study, Ecklebe and L€ offler (2021) identified several antecedents to what employees' perceive as high-quality communication, namely, frequency of communication, information dissemination, participative communication and rejection of secrecy.Although it is reasonable to assume that several of these antecedents might have increased the professional staff groups' satisfaction with the internal crisis communication in the organization, the difference in how the administrative staff and the professional staff groups perceive the internal crisis communication suggests that the divide between the staff adhering to managerialism and the staff adhering to professionalism is an additional factor that could be relevant to consider when assessing the staff's perceived quality of the internal crisis communication in knowledge-intensive organizations and professional organizations with strong professional groups.

Concluding discussion
This study highlights the role of professions in crisis communication, which is an aspect that so far has been ignored.The internal professionalization processes and an intriguing power struggle between professions have obvious consequences for crisis communication.As shown in the overview of earlier research on internal communication, leadership and professional organizations, the prerequisites for creating an increased organizational unity among coworkers are challenging.The idea that a crisis may, as in certain political situations in society, create a "rally-around-the-flag" effect is still relevant, even if the study is an example of how this did not happen.The surveyanswered during an emergency periodshows that there is a gap between management and administration on one side, and strong groups of professions on the other.This can be explained by the two parallel professionalization processesfor the professions on the one hand and the administrative elite on the otherthat are simultaneously active in many professional organizations.The result of these intertwined processes is a power game, with a clear goal to increase status and position in the organization, which is rarely productive and valuable for the organization.The professional group at the top of the professional status hierarchy in the organization stand out since they are the most negative toward top management and do not feel aligned with the overall organization.To a certain extent, this can be explained by the fact that this professional group have another logic that does not necessarily comply with the managerial logic and that professionals tend to identify more with the profession rather than the organization.But the results also indicate why this is the case: this group, in this particular professional organization, does not feel appreciated and listened to by management.Without drawing too great conclusions from a limited survey, it still indicates that internal communication focusing on listening and relationship building may be crucial for avoiding an increased decoupling between management, administration and the professional groups.Thus, the study indicates that internal communication during a prolonged crisis has a much broader and more complex role that goes beyond a simple message service.
It may be that the rally-around-the-organizational-flag effect occurs initially during a crisis, which means that the internal support of and trust in senior management and the administrative elite increases.However, as many leaders around the world have experienced, the effect will not be prolonged (e.g.Johansson et al., 2021).It seems important, and especially so in professional organizations, to develop an open organizational climate that nurtures and enhances internal trust and organizational unity.

Theoretical implications and suggestions for future research
We have in this article argued that internal crisis communication has a broader role than just distributing fast and correct informationit is also crucial for building trustful relationships between managers and coworkers and between staff groups.Managers play a vital role in internal communication, and scholars (e.g.Muffet-Willet and Kruse, 2009) have called for less top-down oriented and more collaborative crisis leaders in relation to increasingly complex crises.However, we still do not know very much about the actual communication practices of such leadership and how it works in different kinds of organizations with different kinds of occupational groups and cultures.Another important theoretical implication is that the power aspect should be part of more studies on corporate communication and strategic communication.In too many cases, power is ignored or taken for granted which is a pity.Power is everywhere and influences us all heavily.Many communication problems for organizations have their base in those power games that exist in the organization.Further, there is a close relationship between power and communication, e.g.Foucault (1980) and (Habermas, 1984), which is another reason for adding a power perspective to future studies.

Practical implications
When it comes to practical implications, there is a need to develop a broader and more complex view of internal crisis communication.As has been shown in this study, different professional groups perceive the internal communication in different ways, and it is thus important not to view employees as a homogenous mass but to differ between groups with different roles and needs (cf.Heide and Simonsson, 2019).Internal crisis communication tends to primarily be seen as a matter of disseminating correct and rapid information during the acute phase of the crisis (Heide and Simonsson, 2019).It is primarily a matter of giving instructions in order to prevent further harm and to keep the business going.This is of course relevant, but there is also a need to see to other communication goals such as strengthening trust, loyalty and engagement, especially during prolonged crises such as the pandemic.
Communication practitioners may not be able to solve fundamental struggles and power games between professionals and managers in a professional organization.However, they may cooperate with other functions in the organization such as HR, top management and the quality department to increase the internal trust more powerfully.One first central step is to put internal legitimacy on the management agenda.All too often, senior management focuses only on external legitimacy, while the internal dimension is ignored.Another step is to stop focusing on managers and leaders as persons and rather emphasize leadership as a relational, mutual process.A relation-oriented approach to leadership recognizes that all employees are an asset that is valuable in handling complex organizational situations.Leaders are "not quite as rational, deliberate, and intentional as they claim to be" (Coutu, 2003, p. 89) and they need input in the form of advice, experience, information, opinions, etc. to make wise decisions.As previously discussed, a crisis like the pandemic is a wicked problem that calls for collaborative, participative listening.An important part of such leadership is active listening, which can be extended to overall organizational strategic listening (cf.Lewis, 2019;Macnamara, 2016).Strangely enough, listening is an area that is more or less ignored by communication practitioners, even though listening is an intrinsic part of communication.Organizations, in general, put much more emphasis on speaking and trying to convince stakeholders and coworkers that senior management is right rather than listening to them to receive various understandings.Naturally, it is not easy to implement strategic listening in organizations, but by taking small steps, making small gains and being persistent, it will be possible to reach a listening mindset in the organization (cf.Weick, 1984).Communication practitioners have an important task in being responsible for the implementation and facilitation of strategic listening.Listening has also an important function for communication practitioners, as it may entail important organizational values such as avoiding or lessening the effects of a crisis, learning, engagement, relationship building, etc.

Limitations
The empirical material used in this article has several limitations.It is based on a rather small sample in one health-care organization, and the study is to be seen as explorative and mainly used to illustrate the theoretical arguments.However, the results from this study indicate quite clearly that the administrative elite and professional groups have different perceptions of leadership and internal communication in a crisis situation.Tensions between different occupational groups need further research in order to gain a better understanding of how to build internal trust and organizational commitment in times of crisis.Apparently, the results from the study cannot be generalized to professional organizations in general, and it would be interesting to have more studies focusing on professional organizations in turbulent times, for example, universities, hospitals and law firms.