An exploratory study of critical incidents within public organizations during leadership change

Role of leadership in change and leadership change

Extant literature suggests that leadership is strongly interconnected with change. The type of leadership is vital to the change process (Hussain et al., 2018). The role of leadership in causing effective change has led to the notion that transformational leadership is a critical component of change even though both transactional and transformational leaders are considered complementary (Bass & Riggio, 2006). Transformative leaders are charismatic, motivational, intelligent, and individualized in their approach, while transactional leaders utilize rewards and penalties to inspire individuals to enhance their organizational performance (Bass, 1985). Given that the presence of leaders stimulates change, leaders must take the initiative by defining a vision (Bass, 1985). Leadership style affects a group’s dynamics and the interaction between its members, which in turn influences the level of organizational readiness to change (Wulandari et al., 2020). Leaders’ spirit, insight, knowledge, compassion, ideals, and learning skills are crucial aspects that lead people to embrace change (Paterson & Cary, 2002). Leadership is the mediating force between a changing environment and an organization’s operating system (bureaucracy) (Mintzberg, 1978). In addition, leaders’ behavior makes changing circumstances more effective (Higgs & Rowland, 2005).

According to experts, different styles of leadership are required for different types of change. This study emphasizes the importance of specific leadership skills and characteristics required for successful change and innovation. Organizations are interested in examining and enhancing the leaders’ talent and ability to implement change and innovate at all levels in order to strengthen the change effectiveness skills (Gilley et al., 2008). This distinct change relates to public organizations when a new leader causes a fundamental change. Leadership changes lead to organizational changes, such as change in the developmental vision, priority programs or innovation, selection of government apparatus resources, and various policy adjustments related to public services. This phenomenon alters the status quo—which represents stability—into either antithesis or synthesis, representing a change for worse or better, respectively (van de Ven, 1995). Change can have a comprehensive impact on all aspects of a public organization, or at only a certain point, which is vital to support the organization; thus, change is either incremental or radical (Tushman & Romanelli, 1985). In addition, change periodically affects medium- and long-term organizational capacity, but some are only shock effects at the beginning of the implementation; thus, change can be either sustainable or unsustainable (Meyer et al., 1993). Change initiatives have become common in organizations and may range from large to medium- or small-scale (Weick & Quinn, 1999).

Lewin’s Three-Phase Process was used as the change model in this study. The model is categorized into three stages: unfreezing, moving, and refreezing (Lewin, 1958). The unfreezing of behavior and the establishment of a willingness to change are the first steps in the change process (Burnes & Cooke, 2013). Lewin (1958) described unfreezing as a difficult re-education process. Moving, also known as locomotion, is the second step, which denotes a change in position within a field performed through action research (Lewin, 1947). The final stage, freezing, denotes the permanence of a new situation (Burnes, 2020). In organizational terms, freezing can necessitate changes in culture, norms, policies, and practices (Armenakis et al., 1999). Cummings and Worley (2009) identified five critical leadership activities in the change process: motivating change, defining a vision, developing political support, managing transition, and maintaining momentum. In line with Lewin’s Three-Phase Process, motivating change and creating a vision demonstrate an unfreezing stage, developing political support and managing the transition demonstrate a moving stage of change, and maintaining momentum demonstrates the implementation and refreezing stages of change.

Anxiety

Organizational change raises uncertainty, which is one of the most commonly reported psychological states (Bordia et al., 2004a). Uncertainty has been characterized as an individual’s perceived inability to precisely foresee something (Milliken, 1987) and has been linked to changes in processes and outcomes (Bordia et al., 2004b; DiFonzo & Bordia, 1998). Bordia et al. (2004a) presented a three-factor paradigm of uncertainty experienced by individuals during change: strategic, structural, and job-related uncertainties. Strategic uncertainty refers to uncertainty regarding organization-level issues; structural uncertainty refers to uncertainty arising from changes in the organization’s inner workings; and job-related uncertainty refers to uncertainty related to job security, promotion opportunities, job role changes, and so on (Bordia et al., 2004a). Uncertainty has several negative consequences on an individual’s well-being and satisfaction in an organizational context (Bordia et al., 2004a). This leads to low morale and job satisfaction (Bastien, 1987; Rosenfeld et al., 2004). It is positively related to psychological strain and negatively related to control (Bordia et al., 2004b). Due to its effect on emotional states, uncertainty about future events is inextricably linked to anxiety and worry (Gu et al., 2020). When changes occur very frequently, individuals are likely to feel fatigued by them and experience an increase in anxiety due to their unpredictability (Rafferty & Griffin, 2006).

Barlow (2002) highlighted that according to Kierkegaard, the source of anxiety lies deep within an individual; anxiety is rooted in the fear of nonexistence, non-being, or nothingness, rather than the fear of death. Only by acknowledging and overcoming this dread of becoming nothing—only by facing the prospect of losing one’s identity—can one completely understand what it means to be human. Only through this experience can we make a clear separation between the self and other things or nonbeings. Other well-known theorists and clinicians have proposed comparable bases for assessing diffused and objectless anxiety. According to May (1979) and Barlow (2002), anxiety is the fear triggered by a danger to some value that the individual believes is vital to their life as a personality. The danger could be directed toward one’s physical life (threat of death) or mental well-being (threat of depression; loss of freedom and meaninglessness). Alternatively, the threat could be directed toward a value that one associates with their existence (patriotism, love for another person, “success,” etc.). Anxiety is a mixture of several emotions that can change over time and in different settings (Izard, 1977).

The mechanisms that mediate anxiety in this study are based on the theory of social evaluation. The general criterion of social evaluation is defined as a mental process in which an individual assigns different values (positive, negative) to specific behavioral patterns (e.g., helping and hindering) that occur in social interactions (e.g., problem-solving), associates these behaviors with specific individuals (partnership values), and displays different behaviors (e.g., avoidance or preference) toward others based on their overall value (Abdai & Miklósi, 2016). According to Manstead and Semin (1981), the negative quality of the actor’s subjective public image following a transgression is the source of social emotions resulting from unintended social infractions. The experimental results reveal that actors feel anxious as they believe others are treating them more negatively as a result of witnessing mistakes, although the situation does not affect the actor’s self-image. The fear of negative evaluations felt by actors defines social anxiety (Button et al., 2015). Button et al. (2015) established that the learning biases of social evaluation associated with social anxiety are more strongly related to self-referential information. The anxiety experienced by the actor in the experiment was in line with social anxiety disorder, which is a severe psychiatric disorder characterized by persistent and overwhelming fears of one or more social or performance settings where there is potential for evaluation by others (American Psychiatric Association, 2013). Fear of negative evaluation is a key component in cognitive-behavioral models of social anxiety and has rich empirical support (Heimberg et al., 2010).

Over the past few decades, the study of anxiety has become increasingly integrative. Theories about anxiety have expanded based on behavior, biology, and cognition aspects, and have been integrated into modern emotion science (Barlow, 2002). Most current theories regard emotions as fundamental action tendencies that aim to inspire conduct related to species survival. Preparing for, avoiding, and escaping potentially dangerous, life-threatening events, which are at the heart of fear and anxiety, are examples of these behaviors or action tendencies (Barlow, 2002). Lang and McTeague (2009) regarded behavioral activities or broad reaction dispositions that occur in specific stimulus circumstances as the essence of emotion. These stimulus contexts establish the function and direction of the acts, which include not only stimulus information that inspires action but also relevant response propositions.

Furthermore, understanding anxiety through a functional account of the emotional approach places individuals and events in an inseparable unit, where events gain significance through individual efforts; thus, events and goals are interrelated (Smit & Lazarus, 1990). According to emotion theory, functionalism is concerned with the link between feelings and what a person is trying to do, rather than evolutionary survival value (Campos et al., 1994). Keltner and Gross (1999) explained the functional account of emotion, including why emotions exist and the systems of interrelated components and consequences. This functional approach provides a corridor for identifying the antecedents that mediate anxiety, as described in the social evaluation theory. Furthermore, the system of components is in line with the modern emotion science approach regarding the relevant response propositions that arise from the circumstances of a specific stimulus. However, this functional approach leads to anxiety.

Several studies have examined anxiety in various contexts, particularly health and psychology. However, anxiety related to leadership changes in public organizations has rarely been studied. The low level of research in this field may be due to some very sensitive issues that are influenced by bureaucratic cultures (e.g., high-power distance culture) in developing countries such as Indonesia. There is a phenomenon gap regarding how anxiety arises and is felt, as well as its consequences in the context of leadership changes in public organizations. In fact, research on a functional approach, the antecedents, system of components, and consequences of anxiety remains scant. There is a gap in our understanding of the variable components of anxiety associated with leadership change. Through this exploratory study of critical incidents, we examined anxiety felt by the heads of departments in local governments in response to leadership change. The study provides a framework for comprehending the problem to be solved (cause), system of components (course), and consequences of anxiety related to leadership change in public organizations. Thus, the following research question was examined: “What are the antecedents, reactions, and consequences of anxiety related to leadership changes in a public organization?”

Critical incident technique (CIT) method

Changes caused by the new leader (regent or major elected for the first time) may be radical at any level of public organizational activity. Radical change is implemented completely if large-scale adjustments are made at the beginning of the change (Romanelli & Tushman, 1994). In contrast, some studies have claimed that radical changes must be made gradually and carefully (Pettigrew et al., 1992). However, it has been determined that early change to specific “high-impact” elements is necessary for completing radical transitions (Amis et al., 2004). We believe that changes implemented either on a large scale or gradually are critical incidents that make unusual and significant contributions to human activities. This is very influential in terms of managing change, as various studies have explained that the most successful changes ensure employee involvement (Gopinath & Becker, 2000; Holt & Vardaman, 2013; Kotter & Cohen, 2002; Rafferty et al., 2013; Rafferty & Restubog, 2010).

A significant contribution to human activity was found in the phenomenon of leadership changes in a public organization. It is consistent with the concept of a critical incident, which is defined as contributing significantly, either positively or negatively, to an activity or phenomenon (Grove & Fisk, 1997). As a result, we consider the CIT to be the best tool for studying anxiety experienced by middle managers in relation to leadership change in public organizations. Furthermore, the CIT was selected based on the possible critical occurrences’ need to contextualize particular situations (Ribeiro et al., 2020). In addition, CIT is an effective method for exploratory research on this little-known phenomenon (Bitner et al., 1990).

The CIT is a series of processes for collecting first-hand observations of human behavior in such a way that they can solve issues and generate broad psychological principles (Flanagan, 1954). This study adapted the theoretical model of critical incidents developed by Grace (2007) and Edvardsson (1992) to formulate the CIT method. By collecting data from the informants, we converted Grace’s (2007) self-report survey into in-depth interviews. In-depth interviews were conducted to obtain more detailed responses and help informants feel more comfortable while disclosing information about the change (Nadia et al., 2020). In-depth interviews help researchers to identify the core issues of various stakeholders and prepare questions for subsequent interviews (Malodia et al., 2021).

Data collection and analytical procedures

Data were collected from June 13 to September 28, 2021, through in-depth interviews using a purposive sampling method. This flexible data mining strategy is based on the CIT, which has a flexible set of principles that must be modified and adapted to match current circumstances, as opposed to a single set of strict rules to guide data collection (Flanagan, 1954). The inclusion criteria for informants were (1) heads of departments in the local governments of cities A, B, and C; and (2) they experienced critical incidents of anxiety associated with leadership change. The three cities held simultaneous elections on December 9, 2020, resulting in the election of a new mayor or regent for the first time. We contacted 32 candidates via social media to inform them about the research design. Additionally, we conducted telephone calls to inquire whether they had experienced critical incidents in the form of anxiety related to leadership changes. The researcher had no relationship established prior to the study’s commencement with informants from City B and C. However, we had established relationships with informants from City A in several previous studies. A total of 26 informants met the inclusion criteria and continued with in-person interviews. No one was present apart from the participants and researchers. The number of informants studied follows Flanagan (1954) that there are no strict rules regarding the appropriate sample size for CIT. We purposively selected 26 informants who had experienced anxiety in response to leadership change and uniquely fit the research interests of the study (Miles & Huberman, 1994; Patton, 2002), and we reached data saturation.

All informants’ data were processed anonymously without distorting their scientific meaning. Twenty informants were male (76.92%), and the rest were female (23.18%). The average age was 49.8 years and the majority had entered ranks IV-A and IV-B (administrative ranks of public organization officers in Indonesia). Verbal consent was obtained from the participants to record, process, and display their statements. Consent was obtained at the opening of the interview, where the author asked for the informant’s willingness to be involved in the research. It was documented in the interview transcript. Most participants chose verbal consent due to their concerns about the research topic, which is closely related to sensitive issues and high-power distance culture in Indonesia. Written informed consent was subsequently obtained, and all participants signed the interview transcript on August, 4-10, 2023. The transcript contained additional clauses regarding informed consent, including the verbal consent obtained at the opening of the interview and the protection of participant confidentiality confirming all interview results were recorded anonymously. Ethical approval was obtained from the Postgraduate School of Airlangga University on August 10, 2023. Ethical approval was obtained retrospectively because of the urgency of our study implementation. This study was conducted after the general election which was held simultaneously in Indonesia for the first time in 2020. This first simultaneous election resulted in 63,91% of regents or mayors being elected for the first time in our research areas. The political change in the post-election period, especially within the 6 months after the election, greatly influenced informants and the culture of public organizations in Indonesia. The new regents and mayors determined a new medium-term development program and changed the bureaucratic structure, and various other fundamental changes were observed, which are thought to cause anxiety among members of public organizations. These considerations necessitated that our study began immediately to capture the critical experiences of local government officers during the leadership change. Another consideration was that our study was low risk, and the implementation and results of the study were continuously monitored by the supervisor, and supported by the continuity of the research process interactively with the informants from data collection to the preparation of the final manuscript. Our research procedures, including the consent process, were discussed in an in-depth joint review forum on July 20, 2023, with our supervisors, education staff, and lecturers from the Postgraduate School of Airlangga University. After the reviewing process, the Postgraduate School of Airlangga University stated that our study had fulfilled the ethical principles. This research complied with the principles of study ethics in the Declaration of Helsinki, as stated by the Postgraduate School of Airlangga University, Surabaya, Indonesia (Number: 2446/UN3.SPS/PT.01.06/2023).

The interview adopts a structured format with eight predetermined open questions, consistently asked to all informants. The average duration of the interviews was 47.8 minutes, and audio recordings were made. A total of 16 interviews took place in non-office settings, such as coffee shops, cafes, and restaurants, while the remaining 10 were conducted within the informant’s office. The interview transcripts were presented to the informants from October 11 to December 13, 2021, after the entire interview was completed. Participants were allowed to review and then provide approval for the interview transcripts. Furthermore, informants were offered the opportunity to review and provide input on the research results, presented in journal format, from March 1 to June 30, 2022. Additionally, before publication, verbal consent was reaffirmed in written consent with assistance from the Postgraduate School of Airlangga University from August 4 to 8, 2023.

Data saturation was achieved after the data collection and analysis. To increase the credibility of the research, we used data triangulation through multiple data sources (Bekhet & Zauszniewski, 2012; Stavros & Westberg, 2009). Fusch et al. (2018) noted that triangulation involves the use of multiple external methods to collect and analyze data. Data triangulation came from two judges and three heads of local government departments from cities A, B, and C. Source triangulation enquires about the veracity of the data gathered from multiple sources. The researchers used secondary data from official websites, regional development plans, vision and mission booklets, and social media analytics to complement the interviews. The data were collected from the official websites and vision-mission booklets containing the proposed change programs endorsed by the regents and mayors. The Regional Development Plans encompass the official medium-term development plans for the municipalities during the period 2021-2024. Concurrently, social media contains a compilation of media coverage from various sources, highlighting changes in leadership in cities A, B, and C. The aim of scrutinizing this information is to comprehend the conditions of leadership change in the three cities from diverse perspectives. Consequently, the researcher expected to gain a more nuanced understanding of the information provided by the informants during interviews. For inclusion and exclusion criteria, secondary data is limited to the three city and is related to changes in leadership.

The analysis was conducted in three stages using NVivo Plus 12 for Windows. Open access alternative software besides NVivo is Taguette, ATLAS.ti, QualCoder, RQDA and MAXQDA. First, the changes perceived by the informants and the causes of anxiety associated with leadership changed were determined. Second, the course manifested by the informant in feeling, thinking, and acting immediately in response to the anxiety associated with the leadership change was examined. Third, the consequences of anxiety associated with leadership change were identified. We conducted a content analysis to improve the validity of the study, with particular attention to the concerns regarding sampling, objectivity, reliability, and systematization (Gremler, 2004). In the second and third phases of the analysis, two judges who were experts in the fields of psychology and socio-politics were involved in the determination of the course and consequence themes.

Causes of anxiety

The causes of anxiety related to leadership change were categorized into five themes: political choices, fear of loss, tactical budget expenditure, systems and policies, and cultural change. Figure 2 illustrates the conceptual framework for the causes of anxiety, and Table 1 summarizes the findings of critical incidents that caused anxiety (as percentages).

Figure 2. Conceptual framework for causes of anxiety associated with leader change.

First theme: Political choices

Based on in-depth interviews using the CIT, political choices are a critical cause of anxiety associated with leadership change. Of the 26 participants, 24 (92.3%) revealed that their political choices induced anxiety while facing a new leader. In-depth interviews revealed “different political choices” (42.31%) as the largest critical incident causing anxiety, followed by “supported politically, but lost in the constituency” (30.77%). Furthermore, apart from being different or supporting political choices, being neutral also caused anxiety (19.23%). The act of making a choice in a political event is always anxiety-inducing, as evidenced by the statements of several participants regarding the theme of political choice:

Political choices put participants in a position of uncertainty, especially regarding the most widely acknowledged career paths. As stated by the participants, political choices on the losing side have a significant impact on career mutations and demotions. On the neutral side, they feel marginalized by political supporters, which causes their judgments to be sidelined. This ambiguity induces feelings of anxiety (Tiedens & Linton, 2001; Todd et al., 2015). Differences in political choices relate to the inevitable diversity of opinions, values, beliefs, interpretations, and goals associated with leader change and are driven by political behavior (Fischer et al., 2007; Jafariani et al., 2012; Petracca, 1991). In Indonesia, informants are legally prohibited from making political choices as civil servants in order to support contestants who compete in elections. However, despite this phenomenon being common, the involvement of civil servants in elections has rarely been studied academically owing to its sensitive nature. Although it is against existing norms, this political choice has a positive impact in that political behavior in organizational change is important for achieving effectiveness (Buchanan, 2008). However, like the wheel of life, political choices that cause anxiety cause fundamental changes that determine votes (Ladd & Lenz, 2008). Based on field findings, the informants’ various political choices were allegedly unknown to the new leader. The feelings of anxiety that arose were associated with the fear of the new leader’s view of those who did not support him or who lost the support scheme during the campaign period. This finding supports social evaluation theory, which explains that anxiety arises from the fear of negative evaluation (Button et al., 2015; Heimberg et al., 2010; Manstead & Semin, 1981).

Second theme: Culture change

Culture change caused by leadership change was identified as a source of anxiety by 84.62% of the respondents. A total of 21 informants (80.77%) stated that the emergence of intruders disturbed and tended to harm the work culture of the status quo. The term intruder refers to groups outside the government bureaucracy that comprise a campaign team, such as leaders of sympathizers, party representatives, and even academics. The intruder group is considered disturbing as it fails to understand the ethics of the bureaucracy as a whole, imposes ideas, and tends to lip service. The following excerpts resonate this sentiment:

In addition, the antecedents of anxiety under the cultural change theme were categorized as new learning (61.54%), poor communication (46.15%), and having to work overtime (23.08%). Critical incidents of new learning are closely related to the informants’ comfort zones. They must leave their comfortable areas and learn to support the changes launched by the new leader. At a certain level, due to policy discretion, the change was outside the scope of official duties and functions. This finding is evident from the following statements by several participants:

Critical incidents of poor communication relate to the communication mechanism between the informants and the new leader, either through official letters or direct interaction. Official letters, which require the leader’s decision to comply with the provisions of laws, are hampered as the leader assigns a new secretary or aide who does not have archival experience or experience with official paperwork. Direct interactions are also hampered, because the new leader’s schedule is tight. This is due to the absence of a balanced distribution system for managing official schedules. This finding is evident from the following statements by several participants:

The critical incident of working overtime is closely related to the leader’s activities, which caused the informants to accompany the leader even outside of working hours and holidays. This finding is evident from the following statements by several participants:

Organizational culture is a valuable asset if it supports an organization’s mission, goals, and strategies. Moreover, it plays an essential role in many elements of an organization (Calciolari et al., 2018; Denison & Mishra, 1995; Hartmann, 2006). The organizational culture in Indonesia, although strongly influenced by the existing political system, is unique at each level of city government. Although not studied in-depth academically, this condition is allegedly in line with Schein’s (2004) statement that organizational culture is distinctive, as it is derived from individual experience and history. This strong culture combines internal and external issues while ensuring the organization’s survival in difficult times. Furthermore, the organizational culture in Indonesia is influenced by the high-power distance culture, which is directly suspected to complicate the level of communication during situations of change. Bordia et al. (2004b) suggested that a systematic program of communication during change is warranted to reduce employee uncertainty.

The presence of intruders harms values and beliefs in the bureaucracy rather than new learning; having to work overtime and poor communication change habits. All these dimensions are closely related to organizational culture as a shared meaning for every member of the organization (Robbins & Judge, 2013). These various dimensions follow organizational culture as a set of value systems, beliefs, assumptions, and norms that have been applied, agreed upon, and followed by the organization’s members for a long time as guidance for their behavior in every situation in the organization, and adapting both in internal and external environments (Schein, 2004). Organizational culture refers to a system comprising values, beliefs, and habits that exist in the organization and interact with the formal structure, which results in the organization’s behavioral norms (Suhariadi, 2013).

Third theme: Policy change

Policy change was also identified as a cause of anxiety associated with leadership change (reported by 84.62%). Here, the most crucial critical incident was the unfair recruitment mechanism for the position of structural officers (76.92%), followed by policy changes that are irrelevant to the status quo (61.54%), extending focus on campaign programs and the tendency to ignore basic service programs (53.85%), giving unfair rewards (46.15%), putting extreme pressure (30.77%), and demanding unrealistic targets (23.08%). The unfair recruitment mechanism for structural officers was stated by the informants based on their past experiences, which highlighted that every change in power, especially for those newly elected, would result in massive mutations. This mutation tends to ignore professionalism but is seen as a reward for the campaign team that supported the new leader during the campaign period. This finding is evident from the following statements made by several respondents regarding the theme of systems and policies:

The policy change dimension relates to the implementation of organizational change policies by new leaders. This implementation is important for managing the change process to achieve developmental visions, as proclaimed by the new leader. Policy analysis has long noted the importance of clear and specific policy objectives and coherent thinking about the relationship between the initiatives to be implemented and desired outcomes (Grizzle & Pettijohn, 2002; McFarlane, 1995).

Borrowing from Pierson’s (2000) terminology, a transformation in the policy change dimension can be called a critical juncture. The critical juncture in organizational change, especially in systems and policy changes, is a sudden impetus for change in stable institutional arrangements (Pierson, 2000). According to various informants, the mutation and promotions for public organization officers that had been carried out for years based on the list of ranks and job evaluations were abruptly altered with mass transfers that were influenced by political interests. Thus, politics plays an important role in the overall process of policy change and reform. Politics influences the origin, formulation, and implementation of public policy, especially in the face of significant changes (Reich, 1995). Political events increase employee anxiety because they are often volatile. This causes public organization officers to become emotionally detached and the quality of their work, in turn, declines (Bailey & Raelin, 2015).

Fourth theme: Fear of loss

Fear of loss was identified as another cause of anxiety. The fear of losing a position due to mutation or demotion (61.54%), followed by the threat of losing strategic projects (34.62%), the threat of losing respect (30.77%), and the threat of losing relationships (23.08%) were gleaned from the respondents’ accounts:

Fear of loss is closely related to the threat of losing a position due to mutations or demotion, especially for those who serve in an influential department or have ambitions to become regional secretaries (top bureaucratic positions in local governments). Various studies have shown that loss is a source of resistance to change, such as loss of control (Bordia et al., 2004b; Eby et al., 2000), status (Canning & Found, 2015; Rider & Negro, 2015), comfort and privileges (Msweli-Mbanga & Potwana, 2006), and resources (Dubois et al., 2013; Shin et al., 2012). This dimension also resonates with Chreim’s (2006) statement that employees do not really resist change but reject perceived threats to their sense of autonomy, integrity, ideals, loss of status, salary, and comfort. Subconscious awareness of loss causes uncertainty and anxiety, which illustrates how organizations are experienced as fundamentally emotional places (Vince, 2006).

Fifth theme: Unaccountable financing

Finally, 69.23% of the respondents cited unaccountable financing as a cause of anxiety associated with leadership change. This includes additional expenses that must be incurred personally or the involvement of officials in financing change programs. The following excerpts highlight this theme:

The informants recounted that the new leaders proposed various new programs posed as innovations, which were outside the usual circumstances. These new initiatives were not planned or budgeted in advance, which is contrary to Indonesia’s planning and budgeting system completed one year ago. This phenomenon has caused the implementation of new programs to lead to unaccountable expenditures outside the officially available budget. The informants and their staff had to spend private money to support the program’s financing. They continued to carry out the program, although they lost money and understood that the activity was wrong in terms of planning regulations. However, they could not blame anyone, including the new leader, due to the high-power distance culture. No clear external actor can be held accountable for an increase in anxiety (Lerner & Keltner, 2000; Wagner, 2014). Politically, an economic crisis causes anxiety if responsibility is spread and difficult to determine (Wagner, 2014).

Course of anxiety

The course of anxiety related to leadership change was categorized into three themes: negative affective reactions, cognitive reactions, and behavioral tendencies. The conceptual framework is illustrated in Figure 3, and Table 2 summarizes the findings of critical incidents associated with the course of anxiety (as percentages).

Figure 3. Conceptual framework for course of anxiety associated with leader change.

Negative affective reactions are triggered involuntarily by emotions that lead to an individual response due to anxiety. Feeling like an outcast (reported by 69.23%) was identified as the most cited critical incident under this theme, followed by feeling unhappy due to the policy change (53.85%), negative emotions (50%), and feeling humiliated (dehumanized) (23.08%). The following statements highlight this finding:

Kantor (1923) defined affective reactions as the actual ways in which people react to affective stimuli in manners that are strikingly comparable to emotional, desire, and pain reactions. Affective reactions elicit emotions that can be differentiated based on the specificity of the original stimulus or target object (Pieters et al., 1988). Kantor (1923) categorized affective stimuli into objects, events, conditions, or actions. Various studies have explained that the most proximal cause of affective reactions is events, as explained in affective events theory, which is widely recognized as the seminal explanation of the function of affection that influences employee attitudes and behaviors in the workplace (Weiss & Beal, 2005). The results of Paterson and Cary’s (2002) study on organizational downsizing events, which are strongly associated with anxiety, support this theory. This study also demonstrates the relationship between events that cause affective reactions by establishing relevant judgments and perceptions of organizational downsizing as precursors to anxiety. Organizational downsizing as a form of change is in line with Weiss and Cropanzano’s (1996) statement that change is a potentially affective event; therefore, it is critical to analyze the emotional aspect. Paterson and Cary’s results (2002) were also supported by our findings by considering leader change as an event that triggers various affective reactions to anxiety.

Furthermore, Efendic (2017) showed that a single affective reaction or a combination of several affective reactions (negative and positive) affects decision-making. Experiments by Weiss et al. (1999) demonstrated positive affective reactions, such as joy, when responding to appropriate variations in procedures and distributive justice, and negative affective reactions, such as anger, that arise from biased procedures and distributive justice. Considering change as an affective event, incremental and radical changes result in positive and negative emotions, respectively (Gersick, 1994). In our study, negative affective reactions predominated because of anxiety caused by leadership change. The study found that negative affective reactions often manifest as sadness, anger, and frustration (Brundin et al., 2021; Luby et al., 2009; Pihkala, 2022). This finding supports Kiefer’s (2002) finding that changes in complexity are more likely to elicit negative emotions. This is also in line with previous findings that the majority of qualitative and quantitative studies on the effects of change in followers’ emotions indicate that the emotions elicited by change are primarily negative, such as anger, anxiety, and impatience (Bartunek, 1984).

The second theme is that of cognitive reactions that were triggered by the participants’ views about change after experiencing anxiety. Cognitive reactions were manifested as a result of thinking that change is important, but it must be implemented gradually (80.77%); thinking that changes are not as expected (69.23%); thinking that change is good but not with the approach (61.54%); thinking that change is not inclusive (57.69%); thinking that change is only temporary (show off) (46.15%); and thinking that change will not be attainable, as the target is unrealistic (23.08%). This finding is evident from the following statements:

Cognition is a thought process in which a person first becomes aware of stimuli, evaluates their relevance, and then explores various behavioral responses (Scherer, 1999). Individual beliefs regarding change, whether negative or positive, are referred to as cognitive reactions (Piderit, 2000). Cognitive and affective reactions have a bidirectional relationship in that emotion influences cognition and cognition elicits emotion (Smit and Lazarus, 1990). In the context of change, it elicits cognitive responses (positive, negative, neutral, or mixed), which are mediated by judgments of the fairness, scale, velocity, and timing of change (Smollan, 2006; Smollan et al., 2010). In our study, cognitive reactions were found with mixed evaluation criteria, where the thought of change was positive but was followed by other evaluations such as the use of wrong approach and expectations for gradual change. Piderit (2000) explains this difference regarding ambivalence, both within and among the three dimensions of attitudes toward change, namely emotional, cognitive, and behavioral, as an example of open support. This is followed by covert rejection and thoughts of accepting change, which in turn is followed by emotional resistance. Furthermore, our study found that cognitive reactions to anxiety were dominated by negative evaluations such as non-inclusive, unrealistic, temporary, and unexpected targets. These findings support previous studies on the impact of anxiety on cognitive function, such as causing problems in social and work environments (Robinson et al., 2013). Moreover, Kiefer (2005) explained that employees with strong cognitive reactions are accompanied by strong negative emotions such as fear or anger and are more likely to reject change. However, Piderit (2000) explained that negative cognitive and affective responses are often well-intentioned and lead to an organizational implementation that may be more acceptable and beneficial.

The last theme is that of behavioral tendencies which refers to a controlled tendency or intention to act owing to the stimulation of anxiety associated with change. Putting on a “yes man” face (reported by 72.08%) was identified as the most cited critical incident under this theme, followed by working as business as usual (61.54%), reducing anxiety (46.15%), ignoring some tasks (23.08%), and the emergence of health issues (19.23%). The following accounts highlight this finding:

Behavioral tendencies toward change are formed because of cognitive and emotional responses that may include both positive and negative features (Piderit, 2000). Furthermore, Smollan (2006) explained that cognitive, affective, and behavioral responses to change are influenced by factors within the individual, change manager, and organization. As mentioned previously, the “yes man” face was the most cited behavioral response. This term describes the behavior of accepting and following change with mostly different emotional and cognitive conditions. These findings support the explanation of the ambivalence of cognitive, affective, and behavioral responses related to change (Piderit, 2000; van Harreveld et al., 2015), which is open support followed by emotional and cognitive resistance. This finding is suspected to be closely related to Indonesia’s high-power distance culture. Behavioral tendencies associated with the “yes man” face are closely related to the phenomenon of unequal rights between power holders and non-power holders, but of course, more in-depth research is required to uncover them.

Another interesting finding that is influenced by high-power-distance culture is working as business as usual. This is considered a neutral behavioral tendency related to changes. The findings revealed that two middle managers preferred to carry out routines as usual or avoid conflict. The high-power distance and bureaucratic organizational cultures in Indonesia encourage middle managers to choose their behaviors more deeply. This consideration mechanism was expressed by Piderit (2000), who considered the implications of behavioral choices before a behavior was chosen. Additionally, ignorance of some tasks can be considered a form of resistance to leadership change, which is in line with Oreg’s (2006) study. These three forms of behavioral tendencies are behavioral intentions in the form of a tendency to support or reject change (Cunningham et al., 2002; Madsen et al., 2005; Martin et al., 2005).

Reducing anxiety and the emergence of health issues are considered as a form of neutral reaction related to change from an organizational viewpoint. Such behavior tends to affect middle managers’ personalities, both mentally and physiologically. This finding supports Wisse and Sleebos’s (2016) finding that change is a major stress factor for employees and is mediated by feelings of uncertainty. Furthermore, organizational change has a short-term effect on mental health decline (Bamberger et al., 2012). These findings also support Greenglass and Burke’s (2000) finding that organizational change is correlated with depression and anxiety and is a crucial point for organizational members to manage.

Consequence of anxiety

The consequences of anxiety associated with leadership change were categorized into three sequential themes: in-circle, out-circle, and ambivalent circle participation. “Circle” relates to Indonesian culture, which denotes people close to power as “inside the circle” and vice versa. The conceptual framework related to this section is illustrated in Figure 4, and Table 3 summarizes the findings of critical incidents under the consequence of anxiety (as a percentage).

Figure 4. Conceptual framework for consequences of anxiety associated with leader change.

“In-circle participation” is a category in favor of the leader change. The critical incident related to this category was polishing old programs to be packaged as if they supported change (57.69%). This was followed by implementing the change programs as business as usual (42.31%), implementing the program changes with high engagement (23.08%), mobilizing apparatus resources to support the program change (23.08%), creating innovations to support the change (19.23%), and collaborating with the press to improve the image of supporting the change (15.38 %). The following statements highlight these findings:

“Out-circle participation” refers to the resistance against the leadership change. The critical incidents associated with this theme were low engagement to support the change (42.31%), more concern with work outside the government (23.08%), not implementing change programs (11.54 %), reviewing the weaknesses of the new government, providing information to political opponents (11.54%), and applying transfer to a functional position (7.69%). The following statements highlight these findings:

“Ambivalent circle participation” refers to the participation that supports change, but also disguises resistance. The critical incidents associated with this theme were implementing the change program as business as usual followed by low engagement and commitment to supporting change (23.08%), polishing old programs to be packaged as if to support change followed by low engagement and commitment to supporting change (23.08%), polishing old programs to be packaged as if to support change followed by more concern with work outside the government (11.54%), and implementing the change program as business as usual followed by more concern with work outside the government (11.54%). The following statements highlight these findings:

The results reveal that 50% of the informants showed support for the leadership change or were categorized as in-circle participants, 23.08% of the informants who indicated resistance were categorized as out-circle participants, and 26.92% of the informants were included in the category of ambivalent circle participation.

Bovey and Hede (2001a) found that behaviors in response to change included supportive versus opposing, active versus passive, and covert versus overt behaviors. Based on this, “in-circle participation” tends to be supportive, active, and open. In this study, the consequences of anxiety that support leadership change manifest as explicit behavior through active involvement in change (Cunningham et al., 2002). The most dominant and interesting consequence of this category is the polishing of old programs to be packaged as if to support change. Although it seems tricky, this phenomenon can be considered as the creativity of a middle manager who traditionally describes the key levels of analysis as individual, group, and organizational, with creativity culminating at higher levels as a result of creative output at lower levels (Woodman et al., 1993). Further studies are required to examine the level of creativity as an adaptive process of change. However, Ford (1996) explained that a low level of engagement can also produce creativity, which may not be novel or useful, while high engagement can encourage a person’s ability to produce creative outcomes. Polishing creations can be considered an improvisation of using problem-focused coping strategies and adjusting to change (Amiot et al., 2006) as a self-efficacy to execute the courses of action required to manage prospective situations (Bandura, 1977; van Dam et al., 2008).

We also found that collaboration with the press to improve the image of middle managers is a marker of supporting change. From the perspective of change, the communication process is vital for realizing effective change as an inherent process of change implementation, including the dissemination of the value of change, training, feedback, and interaction (Lewis & Seibold, 1998). The main goal of communication is to reduce employee uncertainty (Bordia et al., 2004b). In this case, the informant used the communication system not only to disseminate the value of change but also to imagine themselves as an actor who supports change.

The extreme resistance observed in this study was highlighted by the informants’ decision to switch to a functional position, meaning leaving their position as the head of a local department in the local government (structural position) and moving to a functional position, which is based on function and expertise and is more independent (e.g., Widyaswara or lecturer). This implies that the informant left the organization. The study found that informants who chose this extreme measure felt strong negative emotions, such as the feeling of dehumanization. This finding supports Kiefer’s (2005) study, which claimed that negative emotions indicate an essential organizational judgment and a behavioral tendency to withdraw both in the short- and long-term. Kiefer (2005) claimed that employees who experience strong cognitive reactions and strong negative emotions are more inclined to reject change. This finding also supports Oreg’s (2006) study, which explains that someone who experiences anxiety, anger, and stress due to change has a greater intention to leave the organization.

Other manifestations of extreme resistance besides withdrawal included disloyalty, neglect (Turnley & Feldman, 1999), and sabotage (Spector & Fox, 2002). We found disloyal resistance and neglect in the decision to not implement the program that was concerned with work outside the organization. This study found that middle managers preferred to run organizational programs that have been planned and budgeted in advance. The tight planning and budgeting regulations in Indonesia with information system-based supervision were cited as reasons for not changing program implementation in the current fiscal year, despite the new leader initiating a change program. We also found disloyalty and neglect associated with working outside public organizations, such as being a consultant or running a business. Finally, resistance in the form of sabotage manifested in the form of finding new government mistakes and providing this information to political opponents. Furthermore, Gaubatz and Ensminger (2017) stated that resisting change can be a “contention” that continues to disrupt and frustrate change efforts.

These four forms of out-circle participation can be categorized as forms of irrational resistance, which refers to the term used by de Jager (2001) as resistance to change for its own sake. Rational resistance is referred to as low involvement in implementing changes or being passive (de Jager, 2001). In this study, we found low involvement in the form of loss of focus, low involvement in task coordination, decreased productivity, and decreased enthusiasm for supporting change. Here, “engagement” refers to a positive, fulfilling, work-related state of mind that is characterized by vitality, dedication, and immersion (Schaufeli & Bakker, 2004) and is categorized as job satisfaction, job involvement, organizational commitment, and empowerment (Macey & Schneider, 2008). Low engagement in the short- and long-term will cause leadership changes to fail, considering that various previous studies have explained that the main determinant of successful change requires employee involvement (Gopinath & Becker, 2000; Rafferty et al., 2013; Rafferty & Restubog, 2010). Furthermore, Neubert and Cady (2001) considered employee engagement in change programs and investigated its association with crucial organizational outcomes (Bordia et al., 2004b).

Finally, ambivalent circle participation was also a form of consequence. This circle denotes behavior that supports change but with covert resistance. This theme signifies ambivalence within and among the three dimensions of attitudes to change, namely emotional, cognitive, and behavioral, as an example of open support followed by covert rejection (Piderit, 2000). Previous studies have shown that an ambivalent approach affects physical behavior (Schneider et al., 2015). van Harreveld et al. (2015) described the consequences of ambivalence on influence, cognition, and behavior, and how these different categories relate to each other and to their source, where affection plays the most central role.