Workplace and classroom incivility and learning engagement: the moderating role of locus of control

This study aims to examine the relationship between workplace and classroom incivility to learning engagement and the moderating role of internal locus of control in these relationships. An online questionnaire was administered to 432 students from three private universities in Jakarta, Indonesia. The regression analysis results showed that both workplace and classroom incivility has a negative and significant effect on learning engagement. In addition, the direct effect of workplace incivility on learning engagement is moderated by the locus of control. The negative effect of workplace incivility on learning engagement is stronger for students with low levels of internal locus of control than for those students with high levels of internal locus of control. This study provides a better understanding of the internal mechanism condition to reduce the negative effects of incivility experiences that occur in the workplace and classroom among student employees. The implications and limitations are also discussed.


Introduction
During the last two decades, the issue of uncivil behavior has become a growing problem among employees in the workplace and has also emerged as a problem in elementary school to college/university (Moore, 2012). Uncivil behavior started from the emergence of reports in the United States about the decline of civility in U.S. societyeverything from the loss of civility in the workplace to the absence of manners on mass transit (Bjorklund& Rehling, 2009). Dramatically, the issue of incivility also spread in Asia. Tricahyadinata et al. (2020) noted several studies in Asia, including China, Korea, India, the Philippines, Singapore, Malaysia, and Indonesia. This condition illustrates that the issue of incivility has become a global issue and occurs in all sectors throughout the world.
According to Andersson and Pearson (1999), workplace incivility is "a low-intensity deviant behavior with ambiguous intent to harm the target, in violation of workplace norms for mutual respect" (p.475). The term "ambiguous intent" refers to the perpetrator's unclear motive to intentionally or unintentionally harm the victim in the context of violating norms and mutual respect in the workplace. Aside from the workplace context, incivility is also widely studied in the educational world, especially in the classroom and learning process. Segrist et al. (2018) identified classroom incivility such as sending text messages, disturbing the learning process, arriving late, and leaving the class early as uncivil behavior that most often occurs in the classroom. Students who experience uncivil behavior in class tend to decrease learning involvement and personal well-being (Vuolo, 2018). Eka and Chambers (2019) conducted a literature review that provided a broad overview of the incivility related to emotional and physical stress, financial waste, and hindering the educational process. The further impact of the incivility experience can lead to decreased academic achievement (Al-Jubouri et al., 2020;Bai et al., 2019), sleep disorders (Fritz et al., 2019), emotional exhaustion (Welbourne et al., 2020), and well-being (Vuolo, 2018).
Despite previous findings on the impact of workplace and classroom incivility, there are still several gaps that the current study aimto remedy. First, research on workplace and classroom incivility has mainly attempted to explain the impact of uncivil behavior on the context in which the experience occurs. For example, uncivil behavior in the workplace can lead suffering, fatigue, and decreasedemployees job satisfaction (Kim et al., 2013;Rahim & Cosby, 2016;Welbourne et al., 2016), turnover intention (Cortina et al., 2001;Rahim & Cosby, 2016;Tricahyadinata et al., 2020), and decreased work engagement (Beattie & Griffin, 2014;Tricahyadinata et al., 2020). Classroom incivility can reduce learning involvement or total detachment from the learning process (Vuolo, 2018). This study combines workplace and classroom incivility experiences and examines the impact of incivility on learning engagement on student employees; thus, this will be a unique contribution to the field.
Second, the moderation effect of locus of control on various situations and their responses have been studied intensively (Sassi et al., 2015;Xiao et al., 2018;Zhou et al., 2015). Individuals with an internal locus of control tend to face negative events more positively than those with an external locus (Sprung & Jex, 2012;Zhou et al., 2015). Given the progress made in understanding workplace and classroom incivility, there is a need to identify where untested or unexplored elements are about potential moderators of relationships of workplace incivility with its outcomes (Zhou et al., 2015), and whether individual differences affect workplace and classroom incivility, especially for student employees. The present study expands the relationship between incivility and learning engagement by adding an internal locus of control, which has previously been studied as a predictor of academic engagement and learning performance (Albert & Dahling, 2016;Chukwuorji et al., 2018;Yang et al., 2017). As far as we know, no study has examined the moderating effects of locus of control on relationships between workplace and classroom incivility to learning engagement-so this study makes a theoretical contribution to the proposed model (see Fig. 1).
The third contribution is that the present study enriches studies on incivility in non-Western countries, especially in Indonesia and other Asian countries that have relatively the same cultural characteristics. Liu et al. (2009) in the Asian context culture as well as Miner and Smittick (2016) in the U.S. specifically highlighted the collectivitist culture and power distance as differentiators in terms of actors and targets' responses. Cahyadi et al. International Journal for Educational Integrity (2021) 17:4 For example, employees working in countries with a high collectivitist-such as Indonesia-may regard colleagues' abandonment as impolite compared to employees working in an individualistic culture such as the United States. Furthermore, Indonesia is a country with high power distance characteristics Hofstede et al. (2005), so the behavior of leaders who ignore subordinates is likely to be more acceptable than in countries where low power distance is more prevalent. High power distance is also associated with hierarchy, unequal rights between leaders and subordinates, inaccessible superiors Hofstede et al. (2005), and various other forms of neglectful behavior. Such leadership behavior is likely to be responded to as a form of incivility by employees in low power distance countries in Europe and North America. Focusing on the Indonesian educational setting, the present study aims to investigate a novel, preliminary investigation to examine the relationship between workplace and classroom incivility to learning engagement, and the moderating role of the internal locus of control in these relationships. Specifically, workplace incivility at personal resources from the work environment, classroom incivility at learning activities in other environments, and the combination of the two sources of incivility have a negative effect on learning engagement.

Theoretical background and hypothesis development
This study is anchored in the conservation of resources (COR) theory (Hobfoll, 2001), and the job demands-resources model (JD-R) model (Bakker & Demerouti, 2007). According to the COR theory, when an individual has to face actual threats of losing resources, they have a tendency to experience psychological distress (Hobfoll, 2001). In the context of incivility, when employees experience rude and disrespectful behavior from colleagues or superiors at the workplace, they are more likely to have negative emotions such as mood swings, which leads to resource wastage (Robinson et al., 2014). Furthermore, individuals who become victims are likely to experience disturbances in achieving work goals, social relations, and control negative emotion because their resources are widely used to respond and think about the causes and motives of the perpetrators (Hobfoll, 2001;Zhou et al., 2015). Using the perspective of the JD-R model, Dormann et al., (2018) develop a positive and negative learning (PNL) model to examine the effects of demands and resources related to learning engagement and burnout in students. Engagement is a broad concept used in previous studies to define the various kinds of constructs and experiences, such as affection, cognition, and behaviors (Dormann et al., 2018). The main components of learning engagement were identified in two instances: emotional (such as feelings of belonging and respect, interest, and joy) and behavioral components, which comprise participation and commitment to class activities, task implementation, and persistence in learning (Dormann et al. al., 2018;Finn & Zimmer, 2012), including an individual's ability to engage in behavioral, cognitive, emotional, and motivational ways in the ongoing learning process (Ifenthaler et al., 2018). Recent studies have shown that learning engagement is specifically proven to have positive consequences on learning outcomes such as study completion, increased performance, and perceived achievement (Eccles & Wang, 2012;Li et al., 2019;Yoon et al., 2020).
A more coherent engagement concept is obtained from the workplace context (Bakker & Demerouti, 2007;and, Bakker & Xanthopoulou, 2009). Engagement is expressed as "a positive and satisfying work-related state of mind characterized by strength, dedication, and absorption" (Bakker& Xanthopoulou, 2009, p. 1562. In accordance with the workplace, dedication is associated with the act of being enthusiastic about work, finding it useful, meaningful, inspiring, and motivating, while absorption is characterized by being completely immersed in work without being conscious of time and other surrounding factors. Based on this definition, it can be stated that learning engagement is in line with the attitude and behavior of students to "engage" with the process that occurs in school marked by vigor, dedication, and absorption (Bakker & Demerouti, 2007;and, Bakker & Xanthopoulou, 2009).

The relationship between workplace and classroom incivility to learning engagement
Workplace incivility has been identified in various forms be it verbal or non-verbal, such as using derogatory language, making hidden threats, gossiping, ignoring coworkers' requests, speaking harshly, or being disrespectful (Reich & Hershcovis, 2015). In particular, incivility is distinct from other forms of deviant behavior such as aggression, bullying, and abusive supervision (Ferris et al., 2017), which is generally more easily recognized by victims because it is intentional, and the perpetrator has certain motives and goals. Due to the ambiguous nature of incivility (Anderson & Pearson, 1999), organizations sometimes ignore their forms at work and then create a "spiral", spreading widely in organizations (Andersson and Pearson, 1999). This neglect is because incivility is perceived differently from other forms of deviant behaviors; it is considered not a serious thing, and as a result, these behaviors do not get the attention of management (Cahyadi et al., 2020;Rahim & Cosby, 2016;and Schilpzand et al., 2016). In the same line, classroom incivility is "any action that interferes with a harmonious and cooperative learning atmosphere in the classroom" (Feldmann, 2001, p.137). In general, uncivil behavior forms in the educational environment that are often reported, including students' lateness to class or leaving early, use of cell phones during classes (text messaging, letting telephone ring), packing up books before class is over, interrupting lectures by chatting with other classmates, and eating and drinking in the classroom (Bisping et al., 2008;Bjorklund & Rehling, 2009;and Cahyadi et al., 2020). The topic of workplace incivility attracts the attention of many researchers because these behavioral effects are often ignored (Rahim & Cosby, 2016), but have negative impacts on work behavior and can extend to people's personal lives outside their workplace. For example, previous studies found workplace incivility had negative consequences on family satisfaction and could increase sleep disturbances, mental health, emotional disturbances, and distress at home (Blanco-Donoso et al., 2019;Bunk & Magley, 2013;Liu et al., 2020;Marchiondo et al., 2020;and, Tremmel & Sonnentag, 2018).
Subsequently, workplace and classroom incivility have the same negative impact on students' attitudes and behaviors: decreased job satisfaction and work engagement (Kim et al., 2013;Rahim & Cosby, 2016;Tricahyadinata et al., 2020;and, Welbourne et al., 2016), and increased turnover intention (Cortina et al., 2001;Rahim & Cosby, 2016;and, Tricahyadinata et al., 2020). In the same line, classroom incivility is negatively related to academic achievement (Al-Jubouri et al., 2020;and Bai et al., 2019) and well-being (Vuolo, 2018), and also increases sleep disorders and emotional exhaustion (Fritz et al.,2019;and, Welbourne et al., 2020). More recent studies have stated that classroom incivility reduces involvement or leads to total detachment from the learning process (Myers et al., 2016;and Vuolo, 2018). Classroom incivility also has a negative impact on students' academic and intellectual development, thereby leading to reduced energy for critical thinking in class. Uncivil behaviors distract students during class time and undermine the learning environment (Segrist et al., 2018). Therefore, based on this explanation, the following hypotheses were proposed: Hypothesis 1: Workplace incivility is negatively related to learning engagement. Hypothesis 2: Classroom incivility is negatively related to learning engagement.

The relationship between locus of control and learning engagement
Locus of control was first introduced by Rotter's (1966) seminal work on social learning theory (Galvin et al., 2018). In general, locus of control is "the degree of individual belief in controlling events and outcomes in their lives" (Galvin et al., 2018;and, Rotter, 1966) and can be divided into two dichotomies: internal and external. Rotter (1966) underlines that the difference between internal and external is how these individuals make plans for the future and respond to various events that happen to them. For example, because they have the belief that they control the events that occur, internal individuals tend to have high motivation and commitment (Domino et al., 2015;and Zhou et al., 2016) to achieve success, and vice versa; it is not easy to blame others for the failures they have. Conversely, individuals with external tendencies feel they have no control over what happens to them, so they will blame other parties for their successes or failures. Internals are likely to utilize all their resources and think hard to achieve goals. Individuals who have a high internal locus of control have a tendency to make personal reflections of the events that happen to them, while externals are more likely to see themselves having no control over future circumstances so they will act passive to determine the outcomes (Ng et al., 2006;and, Johnson et al., 2015). This difference makes internal individuals more active than external ones when faced with targets to be achieved in the future. The consequences of locus of control on future behavior can be explained through expectancy theory (Yukl & Latham, 1978), where a person's behavior is a function of expectations and valences. As internals associate success with education, it motivates their basic learning behavior. In the context of education, students with a high internal trust in their abilities and accept personal responsibility to achieve learning achievement (Rinn et al., 2014), will consequently be more engaged in learning activities. Conversely, external individuals believe that they have little control over circumstances;therefore, they tend to attribute their successful or failed learning outcomes to external conditions (Yang et al., 2017). Furthermore, recent studies confirmed that locus of control is related to academic engagement (Albert & Dahling, 2016;Chukwuorji et al., 2018;and, Yang et al., 2017). Therefore, based on this explanation, the following hypothesis was proposed: Hypothesis 3: Internal locus of control is positively related to learning engagement.

Moderating effect of internal locus of control
The moderation effect of locus of control on various situations and their responses have been studied intensively (Sassi et al., 2015;Sprung & Jex, 2012;Stiglbauer, 2017;and, Xiao et al., 2018). Individuals with an internal control locus tend to face negative events more positively than those with an external control locus (Sprung & Jex, 2012). Other empirical evidence stated that the workload effect on stress in counterproductive behavior is higher in individuals with an external locus of control (Sassi et al., 2015). Stiglbauer (2017) provides results which show that the role of locus of control in the relationship between time pressure and "work engagement" is more beneficial internally than externally. The moderating role of locus of control was also tested by Xiao et al., (2018) who show that the internal locus of control weakens the relationship between job insecurity and deviant behavior. Based on the COR theory and previous findings, it can be predicted that the internal locus of control moderates the negative relationship between workplace and classroom incivility to learning engagement. This argument is based on differences in internal and external motivation and personal responsibility (Rotter, 1966;Yang et al., 2017) so that the high and low academic engagement will be influenced by locus of control (Albert & Dahling, 2016;Chukwuorji et al., 2018;and Yang et al., 2017). In addition, the internal locus capability reduces the negative effects of various events such as workload, time pressure, and job insecurity (Sassi et al., 2015;Stiglbauer, 2017;and Xiao et al., 2018) so that the study proposes that the relationship of incivility (workplace and classroom) with learning engagement will be stronger for individuals with a low internal locus of control. In other words, individuals with internal locus of control will be better able to reduce the negative impact of workplace and classroom incivility on learning engagement. Therefore, based on this explanation, the following hypotheses were proposed: Hypothesis 4: The internal locus of control moderates the effects of workplace incivility on learning engagement; such effects are stronger in individuals with a low internal locus of control than in those with a high internal locus of control. Hypothesis 5: The internal locus of control moderates the effects of classroom incivility on learning engagement; such effects are stronger in individuals with a low internal locus of control than in those with a high internal locus of control.

Participants and procedure
A total of 600 undergraduate student employees from three private colleges in Jakarta, Indonesia, were invited to participate in the survey. This group of students takes evening classes at their respective colleges. To assist the data dissemination process, threefaculty members at each college were involved in the survey. Furthermore, to bolster the response rate during the online data collection, participants were offered the chance to be one of the ten winners of an IDR 100,000 cash deposit. A total of 432 valid questionnaires were collected (response rate 76%). There were 246 females (57%) and 186 males (43%),and 37% of the participants were "married."The average age was28 years (SD = 4.13), and 268 (62%) had worked in their current organizations for less than 5 years (M = 4.7 and SD = 1.21). Twenty six percent of respondents work in finance and banking sector, manufacturing (23.4%), trade and commerce (21.2%), government institutions (15.2%), education (10%), and restaurants (4.2%).

Measurement
The workplace and classroom incivility was measured using the Cortina et al. (2001) seven-item Workplace Incivility Scale (WIS). Furthermore, the items were modified to reflect the daily incivility that occurs in the workplace and classroom. The items were also rated on a five-point Likert scale of 0 = "never" to 4="most of the time." For example the item: "How often in the past year have you received harsh and disrespectful words from your colleague at __________." The scale was created by averaging the 7 items' responses and a higher score, which indicated greater workplace incivility. Cronbach's alpha was .91 for workplace incivility and 0.90 for classroom incivility. Learning engagement is adapted from the short Utrecht Work Engagement Scale (UWES-9) developed by Schaufeli et al. (2002), with items modified to describe engagement in classroom learning settings. The UWES consists of three subscales: vigor, dedication, and absorption, consisting of three items. For example, vigor ("I am enthusiastic about going to college"), dedication ("I need information regarding the results of my work"), and absorption ("time flies when studying in class"). Each item is rated on a five-point Likert scale from 1 ("never") to 5 ("always"). A higher score reflects a high level of learning engagement. The coefficient alpha for the scale was .89.
Locus of control (LOC) was measured by adapting the 8-item short form of the Work Locus of Control Scale (WLCS) developed by Spector (1988). Participants were asked to indicate their agreement with each of the items on a 5-point Likert Scale, ranging from 1 (strongly disagree) to 5 (strongly agree). An example of the item statement is "promotion atwork is entirely based on my achievements." A higher score indicates an internal locus of control, while a lower score indicates an external locus of control. The alpha coefficient for the scale was .86.
This study comprises four demographic variables: gender, age, marital status, and job tenure. Age and job tenure were measured as continuous variables, while gender and marital status were categorical. Common method Bias, validity, and reliability As the data were drawn from a single source and self-reported by respondents (crosssectional method), the data might be vulnerable to common method variance (CMV) (Podsakoff et al., 2012;Tehseen et al., 2017). We tested for common method bias using a Harman single factor (Podsakoff et al., 2012). The results showed that no single dominant factor of the subscale could explain more than 50% of the total variant. Therefore, it could be stated that CMV is not a serious problem in this study. A confirmatory factor analysis (CFA) on the four constructs of workplace and classroom incivility, learning engagement, and locus of control was performed to measure the internal consistency reliability, convergent validity, and discriminant validity of the constructs in the proposed model (see Table 1). Convergent validity was tested using factor loadings, composite reliability (CR), and average variance extracted (AVE). Evaluation of the measurement model was based on the loading factor value, that is, an item stated to be the forming factor if the value of standardized factor loadings>.50 (Hair et al., 2010). The test results showed that the values of all the items had a loading factor > .50, indicating preliminary evidence for the convergent validity of the measurement model. Composite reliability illustrates the extent to which construct indicators are indicated as part of latent variables, ranging from .90 to .93. The results exceeded the recommended value of .7 (Hair et al., 2010). The average variance extracted (AVE) reflects the total number of variants in the indicator representing latent constructs; in the study it reached the recommended threshold of 0.5, providing support for convergent validity (Fornell & Larcker, 1981;Hair, et al., 2010). Next, we tested discriminant validity as evidence to be able to discriminate between measures of dissimilar constructs (Hubley, 2014). Table 2 shows that the assessment of discriminant validity is provided by the resulting correlation between workplace incivility and classroom incivility of .058, thus supporting the fact that both are not found to be related to each other. Another method is to use the Fornell-Larcker criterion where the AVE root value is compared with the correlation between variables. The results show that the roots of AVE were all greater than the correlation between variables in the model so that discriminant validity was acceptable (Hair et al., 2010). Table 2 shows the mean, standard deviation, and correlation between the constructs. The descriptive analysis results show that workplace and classroom incivility scores are above the average median level on each scale. Similarly, themean score of learning engagement and locus of control were above the median score of 3.62 and 3.67, respectively. Furthermore, the variance inflation factor (VIF), classroom incivility, and locus of control were, respectively, 1,06, 1,02, and 1,08, with learning engagement as the dependent variable. This is far below the conservative threshold value of 10, which is used to determine serious multicollinearity problems (Hair et al., 2010). Therefore, multicollinearity is not a problem in interpreting the subsequent statistical results.

Hypothesis testing
This study used hierarchical regression analyses (HRA) to test Hypotheses 1, 2, and 3, while macros with 5000 bootstrap samples were used to test Hypotheses 4 and 5, as shown in Table 3. The results of the hierarchical regression analysis are shown in Table  3. Four control variables were entered in Step 1, while workplace incivility, classroom incivility, and locus of control were in Step 2. Table 3 shows that all control variables, such as age, gender, job tenure, and marital status, insignificantly affect learning engagement. Workplace incivility was negatively and significantly related to learning engagement (β = −.21, SE = .34, p < .01), thereby supporting Hypothesis 1. Furthermore, the classroom was negatively and significantly related to learning engagement (β = −.13, SE = .04, p < .01), thereby supporting Hypothesis 2. Additionally, in line with Hypotheses 1 and 2, the internal locus of control was positively and significantly related to learning engagement (β = .45, SE = .03, p < .01). Hypotheses 4 and 5 predicted that the internal locus of control has the ability to moderate the relationship between workplace and classroom incivility to learning engagement. Table 3 presents the PROCESS results, which show that the interaction term of workplace incivility and locus of control (int_1) was positively and significantly related to learning engagement. Furthermore, different insignificant results were shown in the interaction of classroom incivility and locus of control (int_2). Therefore, H4 was successfully supported, while H5 was rejected.
The construct bias-corrected in this study is in accordance with the Xie et al. (2018), with 5000 bootstrapped samples used to determine the confidence intervals (CIs) for all the significance tests. Table 4 shows that the direct relationship between workplace incivility and learning engagement was significant for low and high internal locus of the control group (β = −.278; p < .,01) and (β = −.14; p<,01). Workplace incivility has a negative and significant relationship with the low and high internal locus of control, with a 95% bias correction. This means that the internal locus of control in this relationship is negative, where a high level of locus of control reduces the negative effects of workplace incivility on learning engagement.

Discussion
We examined whether workplace and classroom incivility reduces learning engagement and how the internal locus of control moderates these relationships. Five hypotheses were put forward, and four were supported. The empirical results show that workplace and classroom incivility reduced learning engagement, whereas the internal locus of control increased learning engagement. Moreover, employees with higher (vs. lower) Table 3 Results of hierarchical regression analyses and macro process Step 1 Step 2 Process Macro levels of internal locus of control coped with workplace incivility and were more positively engaged in learning. In other words, a high level of locus of control reduces the negative effects of workplace incivility on learning engagement. Unexpectedly, internal locus of control did not play a moderating role in the classroom incivility-learning engagement relationship. The current study extends our understanding of the effect of workplace and classroom incivility experience on learning engagement and suggests that when an individual is exposed to more uncivil experiences over a working day it can affect their personal life. This provides a better understanding of the relationship between the experience of workplace incivility and other outcomes. This study extends our understanding of the effect of classroom incivility experience on learning engagement and suggests that when an individual experiences more classroom incivility, he or she tends to reduce involvement or may even experience total detachment from the learning process (Myers et al., 2016;Vuolo, 2018).
As hypothesized (H1), workplace incivility is negatively linked with learning engagement. In other words, participants experienced higher levels of incivility on days when their work influenced their behavior outside the work environment (e.g., learning engagement). The findings provide support for existing research showing that experiencing incivility during work has a negative effect on employees' non-related work and theirpersonal life, including family satisfaction, sleep disturbances, emotional disturbances, and distress at home (Blanco-Donoso et al., 2019;Liu et al., 2020;Marchiondo et al., 2020;Tremmel & Sonnentag, 2018). Specifically, we succeeded in proving that incivility impacts behavior outside work among student workers. This study provides empirical evidence of the "crossover" and "spillover" effect (Marchiondo et al., 2020) of workplace incivility on learning engagement. Also, as hypothesized (H2), classroom incivility is negatively linked to learning engagement. It can be stated that students who experience uncivil behavior in class tend to decrease learning engagement. This study provides evidence that classroom incivility can reduce involvement or lead to total detachment from the learning process (Myers et al., 2016;Vuolo, 2018). Learning engagement in this study is indicated by vigor, dedication, and absorption (Schaufeli et al., 2002) so that the classroom incivility experience that occurs in the learning process can reduce students' efforts to learn (dedication), make them feel uncomfortable in class (absorption), and weakens learning enthusiasm (vigor). In other words, students with high perceived incivility will tend to avoid learning activities. However, when compared to workplace incivility, classroom incivility has a weaker relationship with learning engagement. This is reasonablebecause the respondents in the sample in this study were part-time students who spent the daytime working and attended classes at night. In other words, the respondents spent more time in a work environment than in a classroom. Hypothesis 3 proves that locus of control has a positive effect on learning engagement. This study supports previous research that indicated individual factors as antecedents of student engagement and learning outcomes, such as locus of control (Albert & Dahling, 2016;Chukwuorji et al., 2018;Yang et al., 2017). We can state that a high level of internal locus of control has a higher tendency for personal responsibility (Rinn et al., 2014) to achieve learning success. This condition then has implications for higher vigor, dedication, and absorption (Schaufeli et al., 2002) in learning activities.
This study further highlights the role of internal locus of control on the effects of workplace and classroom incivility on learning engagement. The combination of these two experiences has a negative effect on learning engagement, but the effect is different based on the level of internal locus of control. Hypotheses 4 and 5 examined the moderating role of locus of control on the relationship of workplace and class incivility to learning engagement. The study results show that locus of control is only proven to moderate the relationship of classroom incivility-learning engagement and doesnothave a significant effecton classroom incivility-learning engagement relationships (see Fig. 2). The significant moderating effects of locus of control have not only provided further evidence for the moderating role of individual differences in COR theory, it also extends our understanding of how people with different personality traits react to workplace and classroom incivility as personal resources in the JD-R modelandthe PNL model (Dormann et al., 2018). The study's findings on the moderating effect of locus of control is consistent with previous studies that suggest that individuals with a low internal locus of control tend to have stronger negative emotional reactions to experienced workplace incivility (Zhou et al., 2015). Individuals with an internal control locus tend to face negative events more positively than those with an external control locus (Sprung & Jex, 2012). In workplace incivility experiences, which are physical and Fig. 2 The relationship between workplace incivility and learning engagement at high and low levels of internal locus of control psychological, this study supports findingsthat internal locus of control tends to respond more positively than external. The moderating effect of locus of control suggests that internals who believe that their achievement and outcomes in academic activities are controlled by their personal actions are able to undermine the negative effect of workplace incivility on learning engagement. Our results showed that workplace incivility is negatively correlated (β = −.21) with learning engagement, contrary to the locus of control construct (β = .45).
This study clearly shows that the effect of negative workplace incivility on learning engagement is much reduced from −.21 (for high level of internal locus of control) to −.14 (for high level internal locus of control). Accordingly, when employees experience unpleasant days at work during the day, the negative effect on learning behavior at night tends to decrease because individuals with high internal locus of control are better able to handle stress (Sprung and Jex, 2012).
Unexpectedly, locus of control does not have a moderating effect on the classroom incivility-learning engagement relationship. Technically, Baron and Kenny (1986) explain that the moderator variable has a role to influence the direction and/or strength of the relationship between the independent variable and the dependent variable. In this context, the relationship between classroom incivility and learning engagement will be influenced by locus of control (strengthened or weakened), represented by interaction variables. In this study, the interaction between classroom incivility and locus of control was not significant (p > .05) so that the level of internal locus of control did not make a difference to the effect of classroom incivility on learning engagement. This study found that the weak effect of classroom incivility on learning engagement was probably due to students not taking uncivil behavior that occurred in the classroom environment too seriously, so that regardless of the level of internal locus of control, their interaction did not significantly affect learning engagement. This condition may further explain why locus of control did not moderate the link between classroom incivility and learning engagement.
This study suggested some practical implicationsas ways of promoting civility that include responding and coping with uncivil behavior effectively in work and campus environments. This method includes conducting open discussions and encouraging respect for others as well as the implementation of effective rules against uncivil behavior: First, more attention should be paid to employees' workplace and classroom incivility. The manager needs to realize that uncivil behavior in the workplace has an adverse effect on life outside of work-especially on learning engagement. Intervention is needed to improve effective communication skills at all levels. Furthermore, social intelligence training is carried out to reduce incivility with increased social skills, which are used to carry out effective, verbal, and non-verbal interactions and communication ethics in accordance with the rules and norms applicable to Indonesian society. Companies also need to reaffirm the organization's basic values for employees' guidance to ensure mutual agreement as a characteristic of organizational culture. Second, the faculty members or administrators need to reaffirm learning discipline policies that regulate student ethics in attitude and behavior in the school environment-especially in the classroom. Therefore, policies in the form of disciplinary rules also need to provide instructors, administrators, Cahyadi et al. International Journal for Educational Integrity (2021)  and students with a means of dealing with incivility acts in the classroom. This study also provides clear reasons for administrators choosing certain class behaviors in order to increase politeness and move students toward a friendly learning environment.

Limitations and future research directions
This research hascertain limitations, irrespective of the contributions and implications. Therefore, further research is needed to expand a broad understanding of the workplace, classroom incivility, and learning engagement by considering the following points. First, the cross-sectional data and self-assessment have limitations in explaining the causality of the relationship between variables. Although this study reported no bias effect through CMV, further research needs toconsider longitudinal or experimental designs that can control the relationship between variables. Second, although this study explores the relationship between incivility experiences in the workplace and classroom, locus of control, and learning engagement, further research is needed to examine the mediating effect of learning engagement on student academic achievement. Finally, incivility triggers in the workplace, such as supervisors, coworkers, and subordinates, and in classrooms with instructors and fellow students need to be explored. Gender factors also need to be considered to differentiate the response and impact of incivility on other attitudes and behaviors (Tricahyadinata et al., 2020;Welbourne et al., 2016).

Conclusion
Incivility has been identified as a factual problem in education though perceived differently based on the culture and social norms of society. This study advances the research on learning engagement by incorporating both workplace and classroom incivility variables, and to some extent, responds to the call for new approaches explaining learning engagement. Specifically, we identified that the internal locus of control moderated the negative effect of workplace incivility on learning engagement, and this effect was stronger in a lower internal locus of control. The current study demonstrated that when an individual experiences more workplace and classroom incivility than usual, he or she is more likely to have low learning engagement. Finally, both managers and school administrators need to make efforts to promote civility, including responding and coping with uncivil behavior effectively in work and campus environments. This method includes conducting open discussions and encouraging respect for others as well as the implementation of effective rules against uncivil behavior.