Emotional demands,emotional labour and occupational outcomes in school principals

Abstract

Most research into emotional labour is focussed on front-line service staff and health professionals, in short-term interactions. Little exists exploring the emotional labour involved in repeated on-going interactions by educational leaders with key stakeholders. This study explored the relationships between emotional demands, three emotional labour facets, burnout, wellbeing and job satisfaction in 1320 full-time school principals. Principals displayed significantly higher scores on emotional demands at work, burnout and job satisfaction, and significantly lower wellbeing scores than the general population. Structural equation modelling revealed that emotional demands predicted the elevated use of all emotional labour strategies. Surface Acting-Hiding emotions had an inverse relationship with burnout, wellbeing and job satisfaction. Surface Acting-Faking emotions had an inverse relationship with job satisfaction. Deep Acting demonstrated no significant associations with outcome variables. The findings of this study extend the current literature on the effects of emotional labour. The study also extends understanding about the separate effects of the facets of emotional labour, which will aid in the development of interventions to reduce high levels of burnout reported by educational leaders.

Keywords

Emotional labour schools leadership principals burnout wellbeing job satisfaction emotional demands

Introduction

School leaders’ accountability and performance demands have continuously increased in Australia and internationally over the last two decades, affecting both leader and school (Earley et al., 2002; Lingard et al., 2013; West et al., 2010). Increasing accountability has the effect of decreasing decision latitude (Lingard et al., 2013) and autonomy (Fink and Brayman, 2006), and leads to “scrutiny stress” (Lasalvia, 2011). This negatively impacts psychosocial and physiological health (Dewa et al., 2009; Kuper and Marmot, 2003), reciprocally affecting job performance and satisfaction (De Nobile and McCormick, 2010; Johnson et al., 2005). When school principals’ wellbeing declines, their ability to significantly impact school functioning, student engagement and whole-school wellbeing also declines (Hallinger and Heck, 1998; Leithwood et al., 2008; Ten Bruggencate et al., 2012).

The considerable job demands experienced by school principals (Riley and Langan-Fox, 2013; Friedman, 2002; Phillips et al., 2007) include such diverse tasks as managing staff, organising budgets and providing strategic organisational focus alongside high stakes testing (Billot, 2003; Clarke, 2006; Lingard et al., 2013). Various studies have shown that these demands can be perceived as stressful, via the influence of multiple variables; environmental, organisational, individual and demographic (Boyland, 2011; De Nobile and McCormick, 2010; Friedman, 2002; Phillips et al., 2007). Further, chronic stress leads to feelings of burnout, affecting job performance, satisfaction and the motivation to remain in the role.

School principals continuously meet multiple stakeholders at different developmental levels: children, adult employees, peers, parents and supervisors/employers; all of whom may sometimes display extremely high levels of emotional arousal. This is emotionally demanding. To keep a school running effectively, they must also be sensitive to the needs of all these groups, balance competing objectives and be able to switch seamlessly between stakeholder interactions, while continuously managing the impression others have of them (Berkovich and Eyal, 2015). This requires a great deal of emotional labour. In the following section we consider two of these important components – emotional demands and emotional labour – and their impact on burnout and job satisfaction.

Emotional demands

School principals are required to present a controlled and calm face to all stakeholders while maintaining a balance between caring and managing (Berkovich and Eyal, 2015; Blackmore, 2010; Day et al., 2001; Eacott and Norris, 2014). However, some of the more stressful challenges faced by principals are those that are accompanied by high emotional demands. Of the five main sources of stress experienced by the principals identified by Gmelch and Swent (1984), two involved high emotional demands; interpersonal relations and intrapersonal conflicts. Likewise, Friedman (2002) reported that interactions with staff and parents affected burnout levels more than role overload, while Poirel et al. (2012) found that interpersonal stress sources were second only to administrative constraints.

School principals experience the full gamut of emotions in their work; responding to their own and others’ emotions is a central part of the role (Beatty, 2000; Berkovich and Eyal, 2015; Blackmore, 1996; Crawford, 2009; James and Vince, 2001; Rajah et al., 2011; Zikhali and Perumal, 2015). For example, principals are expected to appropriately decide when to suppress or amplify negative emotion when confronted with students or teachers who have transgressed rules, or amplify positive emotions; behaving calmly in the face of problems to which they may not know the solution, putting on a fake smile to influence others’ emotions and be positive for parents (Crawford, 2007; Rhodes and Greenway, 2010). The successful management of widespread education reforms that demand compliance (for example, decentralisation, along with increased high stakes testing and accountability) could also be emotionally burdensome (Crawford, 2007; Lingard et al., 2013). Moreover, there is a dominant “display rule” (Zapf, 2002) pressuring school leaders to manage personal emotional responses so as to express their most rational selves (Berkovich and Eyal, 2015). Taking a critical cultural perceptive, these changes to the educational milieu in Australia and similar western countries in recent times, for example, the dominance of a rational masculine discourse (Hofstede, 2001) and the pressures of market reforms (Ball, 2012; Lingard et al., 2013), further justify the focus on this aspect of school leadership. It is, therefore, likely that principals in the current study will report high levels of emotional demands, which in turn will predict poor occupational functioning; surviving rather than thriving in the role.

Emotional labour

Emotional Labour (EL) is generally described by two core processes: surface acting (SA), which encompasses both the inhibition and manufacturing of emotions; and deep acting (DA), which is considered an extension of true emotions, aligning one’s real emotions to the situation (Grandey, 2000, 2015; Hochschild, 1983). SA has two components: hiding or down regulating felt emotions and faking false feelings (Brotheridge and Taylor, 2006; Lee and Brotheridge, 2011, Mann, 1999). Acting is thought to cause emotional dissonance, which can affect wellbeing (Rafaeli and Sutton, 1987; Zapf, 2002). DA involves attentional deployment and cognitive change, while SA is aligned with response modulation (Grandey, 2000; Gross, 1998). Therefore, the outcomes associated with EL vary according to the type of EL performed.

Deep acting

Mixed effects have been found regarding DA. For example, personal accomplishment has demonstrated positive (Brotheridge and Grandey, 2002), negative (Näring et al., 2006) and no significant associations (Lee et al., 2010) with DA. When people reported using DA>SA, it positively predicted job performance (Hülsheger and Schewe, 2011) and truncated emotional exhaustion up to one year later (Philipp and Schüpbach, 2010). In a meta-analysis of EL, DA had no generalisable significant relationships with negative occupational outcomes (Hülsheger et al., 2011) though Grandey et al. (2013) did find an unusual positive association between DA (and SA) and job satisfaction in college students and Taiwanese salespeople.

Surface acting

Grandey (2000) posited a model in which physiological stress theories could account for the outcomes of SA, assuming that the cognitive and motivational demands induce job strain (Demerouti et al., 2002). In support of this, Gross (2002) and Harris (2001) found that emotional suppression increased cardiovascular activation. The negative outcomes of SA may also be attributable to self-perceived inauthenticity or overloading personal resources from persistent modification of emotional expression (Grandey, 2000; Grandey et al., 2012; Hülsheger and Schewe, 2011). Positive associations between SA and poor outcomes such as burnout and declining job satisfaction, psychological and physical health are well established (Brotheridge and Grandey, 2002; Kinman et al., 2011a; Pugliesi, 1999; Totterdell and Holman, 2003). For instance, emotional exhaustion (a facet of burnout) has been associated with SA in physicians (Lee et al., 2010) and teachers (Näring et al., 2012), while the use of SA predicted increases in psychological strain after two months in trainee teachers (Hülsheger et al., 2010). Hiding emotions is correlated with emotional exhaustion and depersonalisation in service sector employees (Bayram et al., 2012), and emotional exhaustion in doctors. However, only faking correlated with depersonalisation in doctors (Lee et al., 2010).

To summarise, EL is used to manage internal emotions and manufacture external expressions of emotions to specifically match organisational norms, expectations and demands (Diefendorff and Gosserand, 2003). It is performed to elicit appropriate responses in clients, customers or other stakeholders to achieve, or comply with, organisational aims. For example, school leaders use EL for “impression management” (Rhodes and Greenway, 2010). Its use in educational leadership is associated with: “macro- and micro-contextual factors” including gender relations or lack of supervisor support; “leadership role factors” such as loneliness; and “mission-related factors” such as ongoing experiences of social injustice (Berkovich and Eyal, 2015). Other caring professions such as nurses (Bartram et al., 2012), healthcare professionals (Grandey et al., 2012; Lee et al., 2010), teachers and childcare workers (Lee and Brotheridge, 2011; Näring et al., 2006; Schutz et al., 2009; Yilmaz et al., 2015), aspirant school leaders (Gallant and Riley, 2013) and the clergy (Kinman et al., 2011a) have all demonstrated differing levels of EL. For the current study we sought to ascertain whether school principals’ levels of emotional demands predicted their use of EL strategies. We hypothesise that emotional demands will be high and will predict an increased use of EL strategies.

Burnout

Burnout is a multi-dimensional “ill-being” concept, defined as a state of emotional exhaustion with follow-on and/or corresponding effects of depersonalisation and a diminished sense of personal accomplishment due to protracted experiences of workplace stress (Federici and Skaalvik, 2012; Lee et al., 2010; Maslach, 2003; Maslach and Jackson, 1986). Occupational stress theories posit that burnout results from the combination of prolonged demands and/or effort and limited or over-used resources (Bakker and Demerouti, 2007; Bakker et al., 2010; Hobfoll, 1989; Karasek, 1979; Lazarus, 1991). Studies of burnout across occupations and countries have demonstrated links with poor job performance, and physical and emotional wellbeing (Maslach et al., 2001). Emotional exhaustion is the first phase of burnout and has been suggested to be the central component of the syndrome (Bakker et al., 2003; Maslach, 2003), and educational leaders are more likely to indicate burnout as emotional exhaustion than either of the other facets (Combs and Edmonson, 2010). Further, significant negative associations have been found between SA and emotional exhaustion in teachers (Yilmaz et al., 2015). Recently, Grandey and Gabriel (2015) suggested that investigating the effects of EL through the lens of ill-being (as burnout) was limited and that whole-person wellbeing ought to be assessed too. We have addressed this concern in the current study.

Leaders across industries report EL and burnout levels similar to those recorded for “people work” occupations, such as nurses and social workers (Brotheridge and Grandey, 2002), and leaders in managerial roles comparatively perform more SA than workers in non-managerial roles (Sloan, 2012). Humphrey et al. (2008) reviewed EL leadership literature, finding that leaders use EL more frequently and with more variation than service workers. They postulated that the successful management of both one’s own and others’ moods could enhance a leader’s effectiveness, while noting negative effects such as emotional exhaustion. Little research into EL in leaders has been performed (Gooty et al., 2010; Haver et al., 2013; Humphrey, 2012). We hypothesise that SA-Hiding and SA-Faking will be positively associated with burnout and negatively associated with wellbeing.

Job satisfaction

Job satisfaction, commonly used as a key indicator of employee wellbeing (Page and Vella-Brodrick, 2009), is defined as the positive emotional regard one holds for one’s job (Schaufeli and Bakker, 2010). Across multiple professions, low job satisfaction is related to declines in physical health and wellbeing (Johnson et al., 2005). Facets of SA have consistent negative relationships with job satisfaction. Faking was found to have a significant negative relationship with job satisfaction in a mixed group of public service workers including managers and teachers (Hsieh et al., 2012). Hiding and Faking combined had negative correlations with job satisfaction in academic and administrative university staff (Pugliesi, 1999), nurses and child-care workers (Seery and Corrigall, 2009), teachers (Kinman et al., 2011b) and the clergy (Kinman et al., 2011a). We hypothesise that SA-Hiding and SA-Faking will be negatively associated with job satisfaction.

The relationship between DA and job satisfaction is less clear. One meta-analysis of the effects of EL revealed consistent moderate negative correlations between job satisfaction and SA (Hülsheger and Schewe, 2011), but no significant relationship between DA and job satisfaction. However, another meta-analysis found the same moderate negative relationships between SA and job satisfaction, but contrarily found a small positive effect of DA on job satisfaction (Wang et al., 2011). We will explore the relationship between DA and job satisfaction in this study with no directional hypothesis.

Empirical investigation

Research aims

We could find no research that specifically measured EL or determined the separate contributions of hiding and faking emotions to burnout, wellbeing or job satisfaction in school principals, prompting the current study. We aimed to identify the extent of emotional demands placed on leaders, assess the types and amounts of EL they reported using, measure the positive and negative impacts of EL strategies, and model these relationships. The hypothesised model is shown in Figure 1 below.

Figure 1.

Hypothesised model. Absence of a line implies no hypothesised significant effect.

Methodology and participants

This study is part of a larger, on-going longitudinal project, which began annual data collection from school principals in 2011. The background to the study, entire question set and progress reports are available at [www.principalhealth.org/au]. A representative sample of full-time school principals (N = 1320; 51.8% female, 48.2% male) was drawn from three parallel governance sectors (Government, Catholic and Independent), across all states and year levels, in Australia. Ages ranged from 26 to 73 years (M = 52.90, SD = 6.78). Years of experience in leadership roles ranged from 0 to 43 years (M = 14.90, SD = 7.28).

Data collection

Data was collected using an online survey. Web-based surveys produce sound data when used on target populations and/or when driven by non-consumer needs (Ganassali, 2008); therefore, no monetary reimbursements were provided for participation. Careful survey design helps ameliorate common method bias (CMB) when independent data sources are not available (Favero and Bullock, 2015). To this end, issues concerning CMB were addressed ex-ante by separating scales of interest within the larger survey (Favero and Bullock, 2015) and by the use of different scale anchors for question sets (Podsakoff et al., 2012). Further, the respondents were aware that honest answers were required to assist with future job design and were provided with detailed individual wellbeing reports upon completion. These motivating factors can diminish the potential effects of CMB (Podsakoff et al., 2012).

Respondents were invited into the study via their specific professional occupational organisations. Upon completion of the survey, a detailed, personalised report was made available to each participant. The total number of participants in the entire dataset was 2084, representing ∼20% of Australia’s school principals. Of these, 539 participants had registered and completed the survey for the first time, and 1545 completed the survey for the second year (75.4% response rate to the recontact invitation). Only data from respondents in full-time employment, as the overall school leader, were used to create a homogenous group of leaders for analysis (N = 1320).

Measures

Four of the scales (emotional demands, quantitative demands, burnout and job satisfaction) were drawn from the revised Copenhagen Psychosocial Questionnaire (COPSOQ-II: Pejtersen et al., 2010b). The COPSOQ-II scales have good reliability and validity (Bjorner and Pejtersen, 2010; Pejtersen et al., 2010b; Rugulies et al., 2010. See Tables 1 and 2 for Cronbach alphas and descriptive statistics).

Table 1.

Descriptive statistics for the independent and dependent variables; population comparisons where available.

Scale	Principals M (SD)	General population M (SD)	Teachers: Primary/Secondary M	Managers M
Emotional demands^a	69.54 (16.35)	40.7 (24.30)	69 / 48	47
Quantitative demands^a	59.44 (19.31)	40.2 (20.5)	48 / 39	49
Job satisfaction^a	74.47 (18.43)	65.30 (18.20)	63 / 64	73
Burnout^a	55.95 (21.22)	34.10 (18.20)	39 / 36	28
Wellbeing^b	79.24 (9.61)	81.04 (12.71)	–	–
SA-Hiding	3.27 (0.69)	–	–	–
SA-Faking	2.79 (0.83)	–	–	–
Deep Acting	2.52 (0.85)	–	–	–

Note: Comparison groups have different numbers. In this study N = 1320.

^aAll the COPSOQ-II samples are from the Danish validation study: Population N = 3517; primary school teachers n = 120; secondary school teachers n = 25; managers n = 107.

^bAustralian Wellbeing sample N = 2731.

Table 2.

Correlations between all variables and scale reliabilities.

	1	2	3	4	5	6	7	8	9	10
1. SA-Faking	.91
2. SA-Hiding	.62**	.91
3. Deep Acting	.50**	.36**	.90
4. Emotional demands	.42**	.44**	.28**	.79
5. Job satisfaction	−.28**	−.31**	−.16**	−.27**	.84
6. Burnout	.35**	.39**	.21**	.48**	−.38**	.91
7. Wellbeing	−.32**	−.38**	−.22**	−.36**	.53**	−.59**	.93
8. Quantitative demands	.28**	.32**	.17**	.38**	−.30**	.46**	−.36**	.79
9. Age	−.08**	−.08**	−.06^*	−.09**	.11**	−.19**	.09**	−.07**
10. Male	−.05	.00	.03	−.09**	−.09**	−.06^*	−.02	−.10**	−.09**
11. Years of experience	−.07**	−.06^*	−.01	−.03	.04	−.09**	.02	−.04	.45**	.23**

Note: Scale alphas are on the cross diagonal.

**p < .01 (2-tailed).

*p < .05 (2-tailed).

Emotional demands were measured with four items: for example, “Is your work emotionally demanding?” Two questions were measured on a five-level Likert scale where the response options ranged from “Always” to “Not at all”. The other two items were also measured on five-level Likert scales with different anchors “To a very large extent” and “To a very small extent”.

Quantitative demands was measured to account for workload, which has previously been seen to affect outcomes. It was measured with four items: for example, “Do you have enough time for your worktasks?” All questions were measured on a five-level Likert scale where the response options ranged from “Always” to “Never/Hardly ever”. One item was reverse scored.

Burnout was measured using four items, which relate directly to the Emotional Exhaustion subscale of the Maslach Burnout Inventory (Maslach et al., 1996). An example item is “How often have you been emotionally exhausted?” All items had the same five Likert-scale response options ranging from “All the time” to “Not at all”.

Job satisfaction was measured using four items, each beginning with the stem “Regarding your work in general. How pleased are you with…”, followed by four options. An example option was “your job as a whole, everything taken into consideration?”. Items had the same four Likert-scale response options ranging from “Very satisfied” (100) to “Very unsatisfied” (0).

Wellbeing was measured using the Assessment of Quality of Life scale (AQoL-8D: Richardson et al., 2013). The AQoL-8D has 35 items and covers eight dimensions such as coping, relationships, self-worth, mental health and happiness. Items have between four and six response levels. The AQoL-8D has good internal consistency and reliability (α = 0.96) (Richardson et al., 2013. See Table 2 for scale reliabilities).

Emotional labour

Three scales (SA-Faking, SA-Hiding; DA) were each measured with three items from the Emotional Labour Scale – Revised which has good reliability (ELS-R: Brotheridge and Lee, 2003; Lee and Brotheridge, 2011). All items had the stem, “On an average day at work, how frequently do you…”, and the five response levels ranged from “Never” to “Always”. An example SA-Hiding question was “Hide my true feelings about a situation”. An example SA-Faking item was “Show emotions that I don’t feel”. An example DA item was “Really try to feel the emotions you have to show as part of your job” (see Table 2 for scale reliabilities).

Results

Preliminary analyses

As all dependent and independent measures were collected via the same survey, a confirmatory factor analysis (CFA) marker test was initially applied to assess the effects of common method bias (Williams et al., 2010). Another COPSOQ-II scale from the survey “Social responsibility” was chosen as a marker variable, as it had the fewest theoretical links to the substantive variables. The scale comprised three questions regarding levels of discrimination (race/religion, age and health) in the workplace (α = 0.83, M = 76.28, SD = 23.63). As per Williams et al. (2010), a set of increasingly constrained CFA models were estimated to see if the marker variable would have relationships with the variables of interest, even though there was no theoretical reason to think it would. This tests how much the data is affected by the common method of administration. Chi-square difference testing established that the marker variable only loaded on three items and that the common method accounted for 0.36% of the variance. Thus, this test suggests that the effect of common method bias in the study is negligible.

Descriptive analyses

The first aim of the study was to identify the extent of emotional demands in school principals. Descriptive statistics and comparisons with general population norms for the COPSOQ-II and wellbeing variables are presented in Table 1. Bonferroni-adjusted independent t-tests were conducted to compare the scores in this study with general population norms from the COPSOQ II. As it can be seen in Table 1, the participants in this study reported significantly higher levels of emotional demands (t₄₈₃₅, 39.86, p < .01), quantitative demands (t₄₈₃₅, 29.53, p < .01) and burnout (t₄₈₃₅, 15.56, p < .01) compared to COPSOQ II norms. However, they also reported higher levels of job satisfaction (t₄₈₃₅, 35.49, p < .01).

Limited comparisons can be made by occupational groups as only group means are available from the COPSOQ II validation study (Det Nationale Forskningscenter for Arbejdsmiljø, 2011; Pejtersen et al., 2010b). Note that the comparison group sizes vary but all completed the same questionnaire. School principals report emotional demands at a level similar to that of primary school teachers, but higher than managers and secondary school teachers. Their quantitative demands and burnout levels are higher than in all comparison groups. They report job-satisfaction levels similar to those of managers and higher than all teachers. Minimally Important Score Differences (MID) have been established for the COPSOQ II scales. MIDs are as follows: emotional demands = 12.9, quantitative demands = 8.6, job satisfaction = 8.0 and burnout = 9.1 (Pejtersen et al., 2010a). For each scale, the mean differences compared to the general working population exceed the recommended MID, indicating that the differences are of practical significance. The wellbeing measure does not have occupationally specific information, but principals demonstrate significantly lower wellbeing than the general population (t₄₀₄₉, 4.55, p < .01).

The second aim of the study was to assess the types and amounts of EL reported by school principals. It can be seen in Table 1 that principals report hiding their emotions more than either faking emotions or deep acting. Deep acting was the least-used strategy. Pearson correlations were calculated between all variables (see Table 2). As it can be seen in Table 2, strong positive correlations were found between SA-Faking and SA-Hiding, and between SA-Faking and DA. SA-Hiding and DA have a moderate positive correlation. Further, there were significant relationships between all three EL variables and each dependent variable. Increases in SA-Faking, SA-Hiding and DA correlated with increases in burnout, and decreases in wellbeing and job satisfaction. The correlations between years of experience and the variables of interest were very small or not significant in the matrix, so only age, gender and quantitative demands were controlled in the subsequent analysis.

Regression analysis

The final aim of the study was to model the predictive relationships between emotional demands and the three EL facets – burnout, wellbeing and job satisfaction. Simultaneous estimation structural equation modelling (SEM) using Mplus version 7.11 was performed to estimate the relationships between emotional demands, EL and outcomes. A maximum likelihood parameter estimator that is robust to violations of normality (MLM estimator) was used (Boomsma, 2000; Muthén and Muthén, 1998–2012). Unstandardised estimates were reported to ensure scale meaning was retained; note that measures were scored on different scales (emotional demands, burnout, job satisfaction and wellbeing: 0–100. EL variables: 1–5).

After estimation, it was seen that the proposed model had good fit (Root Mean Square Error of Approximation (RMSEA) = .04, Comparative Fit Index (CFI) = .97, Tucker Lewis Index (TLI) = .96, χ² = 887.35; df = 248, p <.001, scaling correction factor 1.05, Standardized Root Mean Square Residual (SRMR) = .04) (Hu and Bentler, 1999). The χ² was significant but this can occur in large samples (West et al., 2012). Figure 2 shows the SEM model with significant unstandardised β estimates displayed only.

Figure 2.

The research model; significant unstandardised β estimates displayed only.

As it can be seen in Figure 2, significant effects in the expected directions were found between emotional demands and all other variables, providing support for those hypotheses. The level of emotional demands positively predicted the increased use of all forms of EL, as well as amplified burnout, diminished wellbeing and lowered job satisfaction. Regarding the EL hypotheses, increased levels of SA-Hiding did predict higher burnout, lower wellbeing and lower job satisfaction. However, the SA-Faking hypotheses were only supported with regard to negative impact on job satisfaction. DA had no significant effects on burnout, job satisfaction or wellbeing.

Discussion

“… As a leader I am expected to be everything to everyone … trying to be positive all the time when I just want to scream is hard work!” (Anonymous participant, this study)

This study examined the associations between emotional demands, three dimensions of EL and outcomes – burnout, job satisfaction and wellbeing – in school leaders. While the themes that emerged from the findings will be discussed separately, they co-occur in an everyday context, as leaders simultaneously manage themselves, and collegial and student relationships, in providing strategic direction and management of the whole school.

Extent and effects of emotional demands

The results of this study suggest that school leaders face significantly increased emotional demands compared to the general population, and this is associated with poorer psychosocial health. Despite having many of the attributes that predict wellbeing, such as stable family backgrounds, stable family relationships, approximately double the average income and secure employment (Riley, 2014), school leaders reported poorer wellbeing than the general population. Their reported burnout was higher than either managers in other fields or teachers (Det Nationale Forskningscenter for Arbejdsmiljø, 2011). These findings support current stress theories, predicting that emotional exhaustion results when demands increase beyond available resources (Hobfoll, 1989; Karasek, 1979; Lazarus, 1991; Siegrist, 2001). This is a real cause for concern. The sustainability of current educational systems relies, to a very large degree, on leaders’ remaining in control of the demands, and these findings indicate that they may be nearing their capacity as the requirements of the role continue to increase. Further, as Australia typically implements education policies developed in the UK and USA (Ball, 2012; Lingard, 2010), these findings suggest replication studies in other jurisdictions may uncover similar issues that need to be addressed.

Job satisfaction was reported at similarly high levels to managers in other industries. While high burnout and concomitant high job satisfaction is unusual, it has previously been seen in school principals (Darmody, 2011) and is also found in psychiatrists (Lasalvia, 2011). Both groups have highly charged emotional interactions with multiple “clients” during a typical day, and both regard their work as important. However, this finding suggests that job satisfaction does not appear to be a protective factor for school leaders’ wellbeing, when the level of emotional demand is very high. It may be that emotional demands are the important variable, rather than the amount of emotional labour required to meet the demand, or the level of satisfaction produced by accomplishment.

Amounts and types of EL used by school principals to meet the emotional demands

Our interest was whether this study could add some clarity to the mixed results produced by EL research in teachers’ EL (Näring et al., 2012; Yilmaz et al., 2015). The school principals in this study consistently reported increased EL when emotional demands increased. They were more likely to hide their real emotions, than either fake false ones or deeply act to align their own emotional response to the situation. Perhaps the cognitive effort involved in deep acting is too onerous or effortful given the constancy and duration of interactions in a principal’s day.

Relationships between emotional demands, EL and negative outcomes

We found that all forms of EL increase in conjunction with emotional demands, providing evidence that the same mechanisms driving EL in other work roles also drive EL in school principals. These findings support the job demands-resources (Bakker and Demerouti, 2007) and transactional models (Lazarus, 1991). The negative effects for principals in long-term relationships are the same as those for service industry workers in short-term interactions. It appears that for all workers, as demands increase beyond the capacity of individuals to access appropriate resources, so the symptoms of stress in the form of burnout and poor wellbeing follow. Our concern is that the education sector is facing steadily increasing demands, while resources are diminishing (Lingard et al., 2013) and principals’ wellbeing is being threatened as a result. Given that the education sector employs the largest professional workforce, these findings are potentially costly. The education industry health insurer recently reported that the cost of providing psychological services to their members (N ≈ 130,000) had almost doubled in the last five years (Joyce, 2014). This suggests that replication studies are needed to determine whether, and to what extent, these issues also affect teachers.

This study supports the literature that identifies a crisis in education as the reform agenda promulgates from system to system (Ball, 2012), and provides evidence of potentially serious health consequences for a substantial workforce through increasing emotional demands. It is not drawing too long a bow to suggest that school principals find themselves in the current policy milieu in a similar position to middle-ranking public servants who were the original subjects of the Whitehall I and II studies. The long-term health consequences of increasing demands and diminishing resources, particularly decision latitude, and the emotional correlates of this change are potentially very harmful (Kuper and Marmot, 2003).

Significant relationships in the expected directions between SA-Hiding emotions and all outcomes were found. SA-Faking significantly negatively impacted job satisfaction, but did not affect burnout or wellbeing. Previous studies using SA measures that combine hiding and faking emotions implicitly suggest that they contribute equally to deleterious consequences. This study demonstrates that this is not the case for school principals; the outcomes of hiding and faking are different. Also, we found no significant effects between DA and burnout, wellbeing or job satisfaction. Though DA is reported as a strategy and correlates with occupational outcomes, it does not directly predict outcomes when all other variables are taken into account. These results support the prior findings that deep acting does not have significant relationships with job satisfaction (Hülsheger and Schewe, 2011) and are in opposition to others (Wang et al, 2011). Perhaps the high levels of job satisfaction seen in the sample obfuscate potential relationships regarding DA.

Burnout/wellbeing. Hiding emotions is the only component of SA that predicts emotional exhaustion and wellbeing in school principals. Like physicians (Lee et al., 2010), school leaders report hiding emotions as the most frequently utilised EL strategy, which is associated with worse outcomes than faking. It is also possible that when faking is performed with positive motives, as would be the case with these participants, this reduces the negative effects. The high level of burnout seen in this population and this association with emotional suppression is of particular importance given the negative health consequences that can result from prolonged emotional job stress (Grandey, 2000; Gross 2002; Harris, 2001, Lazarus, 1991).

Job satisfaction. Hülsheger and Schewe (2011) found a strong negative relationship between the effects of SA and job satisfaction across occupations in their meta-analysis of EL research. This study provides more detailed evidence, showing that hiding and faking separately predict reduced job satisfaction. This could be explained via two simultaneous processes. When one fakes an emotion, there is an outward expression as well as an inward suppression. Outward fake expressions of emotions can lead to feelings of inauthenticity in the workplace (Brotheridge and Lee, 2003; Goldberg and Grandey, 2007). The resulting dissonance between the truth of felt emotion and the mask of expressed emotion is obvious (Mann, 1999). The associated lack of authenticity may threaten the sense of self. Perhaps when one’s sense of authentic self is impacted by workplace stressors, then satisfaction with that workplace diminishes. Second, the simultaneous suppression of true emotion (hiding) can lead to a stress response and perhaps associating work with feelings of physical stress also contributes to workplace dissatisfaction.

Limitations and future directions

Limitations of this study include the use of self-report measures, which can be affected by things such as social desirability. General job resources such as social support levels, autonomy or self-efficacy were not held constant in this study. Future studies may wish to include these variables as controls or to use them in a difference model as moderators. The valence and duration of the emotions being hidden or faked was not specified and it is possible that there are differences in outcomes depending, for instance, on whether the emotion being suppressed is positive or negative, or whether the emotionally laborious interaction is ongoing or fleeting. The COPSOQ-II was validated in Denmark and cross-cultural contrasts may play some part in explaining the differences in mean scores (Hofstede, 2001); replication studies will be needed, both in Australia and other countries, to assess this possibility. Finally, this was a cross-sectional study, and though the causative pathways modelled were based on theory and prior findings, future research could assess the model longitudinally to more clearly ascertain the directionality of effects.

Practical implications

The findings in this study extend the understanding of type and amount of emotional labour performance across occupations, by demonstrating that school principals use a range of emotional labour strategies in executing their role. This study explored the direct paths from emotional demands and EL to outcome variables. Hiding emotions predicts more poor outcomes than faking, while DA had no consequences. As discussed earlier, DA is a form of cognitive reappraisal (Gross, 1998). This type of antecedent-focussed emotional regulation occurs before responding and, as such, appears to be the most desirable form of EL. Therefore, one of the major implications of this study is that DA should be the recommended EL strategy for leaders. Gross’ (2002) research demonstrated that while suppression (hiding emotion) reduced the outward expression of emotion successfully, it did not change the internal emotion. Cognitive appraisal changes expressed and felt emotion, and does not activate cardiovascular stress responses (Gross, 1998; Harris, 2001).

However, schools are rife with emotionality, as they deal with people’s most important hopes, fears and dreams on a daily basis: the lives and futures of children. It will never be possible to either always express true emotions or reappraise them, so hiding and faking will continue. However, explicit education to help leaders manage the effects of high emotional demands, planned or reactive hiding and faking, and healthy post-interaction behaviours would provide some inoculation against burnout and improve wellbeing. For instance, mindfulness has been demonstrated to be effective in coping with emotional demands by reducing stress levels following SA, reducing burnout and improving job satisfaction (Hülsheger et al., 2013). What is clear to us following this study is that emotional demands and associated labour are essential aspects of the school principal’s role and, given their potential to interfere with both individual and school functioning, need more research attention.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Start up funds for this project were provided by Monash University. Funding to continue the project has been provided by: Teachers Health Fund; Australian Primary Principals Association; Australian Secondary Principals Association; Catholic Secondary Principals Association; and, The Association of Heads of Independent Schools of Australia.

References

Bakker

Demerouti

(2007) The Job Demands-Resources model: State of the art. Journal of Managerial Psychology 22: 309–328.

Bakker

Demerouti

Taris

, et al. (2003) A multi-group analysis of the job demands-resources model in four home care organizations. International Journal of Stress Management 10: 16–38.

Bakker

van Veldhoven

Xanthopoulou

(2010) Beyond the Demand-Control Model: Thriving on high job demands and resources. Journal of Personnel Psychology 9: 3–16.

Ball

(2012) Global Education Inc.: New Policy Networks and the Neo-Liberal Imaginary. Abingdon: Routledge.

Bartram

Casimir

Djurkovic

, et al. (2012) Do perceived high performance work systems influence the relationship between emotional labour, burnout and intention to leave? A study of Australian nurses. Journal of Advanced Nursing 68: 1567–1578.

Bayram

Aytac

Dursun

(2012) Emotional labor and burnout at work: A study from Turkey. Procedia – Social and Behavioral Sciences 65: 300–305.

Beatty

(2000) The emotions of educational leadership: Breaking the silence. International Journal of Leadership in Education 3: 331–357.

Berkovich

Eyal

(2015) Educational leaders and emotions: An international review of empirical evidence 1992–2012. Review of Educational Research 85: 129–167.

Billot

(2003) The real and the ideal: The role and workload of secondary principals in New Zealand. International Studies in Educational Administration 31: 33–49.

10.

Bjorner

Pejtersen

(2010) Evaluating construct validity of the second version of the Copenhagen Psychosocial Questionnaire through analysis of differential item functioning and differential item effect. Scandinavian Journal of Public Health 38: 90–105.

11.

Blackmore

(1996) Doing ‘emotional labour’ in the education market place: Stories from the field of women in management. Discourse: Studies in the Cultural Politics of Education 17: 337–349.

12.

Blackmore

(2010) Preparing leaders to work with emotions in culturally diverse educational communities. Journal of Educational Administration 48: 642–658.

13.

Boomsma

(2000) Reporting analyses of covariance structures. Structural Equation Modeling: A Multidisciplinary Journal 7: 461–483.

14.

Brotheridge

Grandey

(2002) Emotional labor and burnout: Comparing two perspectives of “people work”. Journal of Vocational Behavior 60: 17–39.

15.

Brotheridge

Lee

(2003) Development and validation of the Emotional Labour Scale. Journal of Occupational and Organizational Psychology 76: 365–379.

16.

Brotheridge

Taylor

(2006) Cultural differences in emotional labor in flight attendants. In: Zerbe

Ashkanasy

Härtel

CEJ

(eds) Individual and Organizational Perspectives on Emotion Management and Display (Research on Emotion in Organizations, Volume 2). Emerald Group Publishing Limited, pp.167–191.

17.

Clarke

(2006) From fragmentation to convergence: Shaping an Australian agenda for quality school leadership. School Leadership & Management 26: 169–182.

18.

Combs

Edmonson

(2010) Professional burnout and job demands among high school principals: A mixed methods study. In: Toward a Broader Understanding of Stress and Coping: Mixed Methods Approaches. Charlotte, NC: IAP Information Age Publishing, pp.437–462.

19.

Crawford

(2007) Rationality and Emotion in primary school leadership; an exploration of key themes. Educational Review 59: 87–98.

20.

Crawford

(2009) Getting to the Heart of Leadership: Emotion and Educational Leadership. Sage Publications.

21.

Darmody

Smyth

(2011) Job Satisfaction and Occupational Stress among Primary School Teachers and School Principals in Ireland. Ireland: The Teaching Council and ESRI.

22.

Day

Harris

Hadfield

(2001) Challenging the orthodoxy of effective school leadership. International Journal of Leadership in Education 4: 39–56.

23.

De Nobile

McCormick

(2010) Occupational stress of Catholic primary school staff: A study of biographical differences. International Journal of Educational Management 24: 492–506.

24.

Demerouti

Bakker

Nachreiner

Ebbinghaus

(2002) From mental strain to burnout. European Journal of Work and Organizational Psychology 11: 423–441.

25.

Det Nationale Forskningscenter for Arbejdsmiljø [National Research Center for the Working Environment] (2011) Psychological working environment, COPSOQ II data. Available at: http://olddata.arbejdsmiljoforskning.dk/Nationale%20Data/3DII.aspx?lang=da, accessed 1 April 2016.

26.

Dewa

Dermer

Chau

, et al. (2009) Examination of factors associated with the mental health status of principals. Work 33: 439–448.

27.

Diefendorff

Gosserand

(2003) Understanding the emotional labor process: A control theory perspective. Journal of Organizational Behavior 24: 945–959.

28.

Eacott

Norris

(2014) Managerial rhetoric, accountability, and school leadership in contemporary Australia. Leadership and Policy in Schools 13: 169–187.

29.

Earley

Evans

Collarbone

, et al. (2002) Establishing the Current State of School Leadership in England. London: Department for Education and Skills, 151 pp.

30.

Favero

Bullock

(2015) How (not) to solve the problem: An evaluation of scholarly responses to common source bias. Journal of Public Administration Research and Theory 25: 285–308.

31.

Federici

Skaalvik

(2012) Principal self-efficacy: Relations with burnout, job satisfaction and motivation to quit. Social Psychology of Education 15(3): 295–320.

32.

Fink

Brayman

(2006) School leadership succession and the challenges of change. Educational Administration Quarterly 42: 62–89.

33.

Friedman

(2002) Burnout in school principals: Role related antecedents. Social Psychology of Education 5: 229–251.

34.

Gallant

Riley

(2013) The Emotional Labour of the Aspirant Leader: Traversing School Politics. In: Newberry

Gallant

Riley

(eds) Emotion and School: Understanding how the Hidden Curriculum Influences Relationships, Leadership, Teaching, and Learning, pp.81–97.

35.

Ganassali

(2008) The influence of the design of web survey questionnaires on the quality of responses. Survey Research Methods 2: 21–32.

36.

Gmelch

Swent

(1984) Management team stressors and their impact on administrators’ health. Journal of Educational Administration 22: 192–205.

37.

Goldberg

Grandey

(2007) Display rules versus display autonomy: Emotion regulation, emotional exhaustion, and task performance in a call center simulation. Journal of Occupational Health Psychology 12: 301–318.

38.

Gooty

Connelly

Griffith

, et al. (2010) Leadership, affect and emotions: A state of the science review. The Leadership Quarterly 21: 979–1004.

39.

Grandey

(2000) Emotion regulation in the workplace: A new way to conceptualize emotional labor. Journal of Occupational Health Psychology 5: 95–110.

40.

Grandey

(2003) When “the show must go on”: Surface acting and deep acting as determinants of emotional exhaustion and peer-rated service delivery. The Academy of Management Journal 46: 86–96.

41.

Grandey

(2015) Smiling for a wage: What emotional labor teaches us about emotion regulation. Psychological Inquiry 26: 54–60.

42.

Grandey

Chi

N-W

Diamond

(2013) Show me the money! Do financial rewards for performance enhance or undermine the satisfaction from emotional labor? Personnel Psychology 66: 569–612.

43.

Grandey

Foo

Groth

, et al. (2012) Free to be you and me: A climate of authenticity alleviates burnout from emotional labor. Journal of Occupational Health Psychology 17: 1–14.

44.

Grandey

Gabriel

(2015) Emotional labor at a crossroads: Where do we go from here? Annual Review of Organizational Psychology and Organizational Behavior 2: 323–349.

45.

Gross

(1998) The emerging field of emotion regulation: An integrative review. Review of General Psychology 2: 271–299.

46.

Gross

(2002) Emotion regulation: Affective, cognitive, and social consequences. Psychophysiology 39: 281–291.

47.

Hallinger

Heck

(1998) Exploring the principal’s contribution to school effectiveness: 1980–1995. School Effectiveness and School Improvement 9: 157–191.

48.

Harris

(2001) Cardiovascular responses of embarrassment and effects of emotional suppression in a social setting. Journal of Personality and Social Psychology 81: 886.

49.

Haver

Akerjordet

Furunes

(2013) Emotion regulation and its implications for leadership: An integrative review and future research agenda. Journal of Leadership and Organizational Studies 20: 287–303.

50.

Hobfoll

(1989) Conservation of resources: A new attempt at conceptualizing stress. American Psychologist 44: 513–524.

51.

Hochschild

(1983) The Managed Heart – Commercialization of Human Feeling. Berkeley, CA: University of California Press.

52.

Hofstede

(2001) Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations across Nations. Thousand Oaks, CA: Sage Publications.

53.

Hsieh

C-W

Jin

Guy

(2012) Consequences of work-related emotions analysis of a cross-section of public service workers. The American Review of Public Administration 42: 39–53.

54.

L-T

Bentler

(1999) Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal 6: 1–55.

55.

Hülsheger

Alberts

HJEM

Feinholdt

, et al. (2013) Benefits of mindfulness at work: The role of mindfulness in emotion regulation, emotional exhaustion, and job satisfaction. Journal of Applied Psychology 98: 310–325.

56.

Hülsheger

Lang

JWB

Maier

(2011) Emotional labor, strain, and performance: Testing reciprocal relationships in a longitudinal panel study. Journal of Occupational Health Psychology 15: 505–521.

57.

Hülsheger

Schewe

(2011) On the costs and benefits of emotional labor: A meta-analysis of three decades of research. Journal of Occupational Health Psychology 16: 361–389.

58.

Humphrey

(2012) How do leaders use emotional labor? Journal of Organizational Behavior 33: 740–744.

59.

Humphrey

Pollack

Hawver

(2008) Leading with emotional labor. Journal of Managerial Psychology 23: 151–168.

60.

James

Vince

(2001) Developing the leadership capability of headteachers. Educational Management Administration & Leadership 29: 307–317.

61.

Johnson

Cooper

Cartwright

, et al. (2005) The experience of work-related stress across occupations. Journal of Managerial Psychology 20: 178–187.

62.

Joyce

(2014) New research reveals Australian principals face increasing burnout, abuse and bullying [Press release]. Available at: https://www.teachershealth.com.au/news/research-reveals-australian-principals-face-increasing-burnout-abuse-bullying/.

63.

Karasek

(1979) Job demands, job decision latitude, and mental strain: Implications for job redesign. Administrative Science Quarterly 24: 285–308.

64.

Kinman

McFall

Rodriguez

(2011a) The cost of caring? Emotional labour, wellbeing and the clergy. Pastoral Psychology 60: 671–680.

65.

Kinman

Wray

Strange

(2011b) Emotional labour, burnout and job satisfaction in UK teachers: The role of workplace social support. Educational Psychology 31: 843–856.

66.

Kuper

Marmot

(2003) Job strain, job demands, decision latitude, and risk of coronary heart disease within the Whitehall II study. Journal of Epidemiology and Community Health 57: 147–153.

67.

Lasalvia

(2011) Occupational stress, professional burnout and job satisfaction among psychiatrists. In: Langan-Fox

Cooper

(eds) Handbook of Stress in the Occupations. Cheltenham, UK: Edward Elgar Publishing, pp.49–68.

68.

Lazarus

(1991) Psychological stress in the workplace. Journal of Social Behavior and Personality 6: 1–13.

69.

Lee

Ashforth

(1993) A longitudinal study of burnout among supervisors and managers: Comparisons between the Leiter and Maslach (1988) and Golembiewski et al. (1986) models. Organizational Behavior and Human Decision Processes 54: 369–398.

70.

Lee

Brotheridge

(2011) Words from the heart speak to the heart: A study of deep acting, faking, and hiding among child care workers. Career Development International 16: 401–420.

71.

Lee

Lovell

Brotheridge

(2010) Tenderness and steadiness: Relating job and interpersonal demands and resources with burnout and physical symptoms of stress in Canadian physicians. Journal of Applied Social Psychology 40: 2319–2342.

72.

Leithwood

Harris

Hopkins

(2008) Seven strong claims about successful school leadership. School Leadership & Management 28: 27–42.

73.

Lingard

(2010) Policy borrowing, policy learning: testing times in Australian schooling. Critical Studies in Education 51: 129–147.

74.

Lingard

Martino

Rezai-Rashti

(2013) Testing regimes, accountabilities and education policy: Commensurate global and national developments. Journal of Education Policy 28: 539–556.

75.

Man

(1999) Emotion at work: To what extent are we expressing, suppressing, or faking it? European Journal of Work and Organizational Psychology 8: 347–369.

76.

Maslach

(2003) Job burnout: New directions in research and intervention. Current Directions in Psychological Science 12: 189–192.

77.

Maslach

Jackson

(1986) Maslach Burnout Inventory: Manual. Palo Alto, CA: Consulting Psychologists Press.

78.

Maslach

Jackson

Leiter

(1996) MBI: Maslach Burnout Inventory Manual. Palo Alto, CA: Consulting Psychologists Press.

79.

Maslach

Schaufeli

Leiter

(2001) Job burnout. Annual Review of Psychology 52: 397–422.

80.

Muthén

(1998–2012) Mplus User’s Guide. Los Angeles, CA: Muthén and Muthén.

81.

Näring

Briët

Brouwers

(2006) Beyond demand–control: Emotional labour and symptoms of burnout in teachers. Work and Stress 20: 303–315.

82.

Näring

Vlerick

Van de Ven

(2012) Emotion work and emotional exhaustion in teachers: The job and individual perspective. Educational Studies 38: 63–72.

83.

Page

Vella-Brodrick

(2009) The ‘what’, ‘why’ and ‘how’ of employee well-being: A new model. Social Indicators Research 90: 441–458.

84.

Pejtersen

Bjorner

Hasle

(2010a) Determining minimally important score differences in scales of the Copenhagen Psychosocial Questionnaire. Scandinavian Journal of Public Health 38: 33–41.

85.

Pejtersen

Kristensen

Borg

, et al. (2010b) The second version of the Copenhagen Psychosocial Questionnaire. Scandinavian Journal of Public Health 38: 8–24.

86.

Philipp

Schüpbach

(2010) Longitudinal effects of emotional labour on emotional exhaustion and dedication of teachers. Journal of Occupational Health Psychology 15: 494–504.

87.

Phillips

Sen

McNamee

(2007) Prevalence and causes of self-reported work-related stress in head teachers. Occupational Medicine 57: 367–376.

88.

Podsakoff

MacKenzie

Podsakoff

(2012) Sources of method bias in social science research and recommendations on how to control it. Annual Review of Psychology 63: 539–569.

89.

Poirel

Lapointe

Yvon

(2012) Coping with administrative constraints by Quebec school principals. Canadian Journal of School Psychology 27: 302–318.

90.

Pugliesi

(1999) The consequences of emotional labor: Effects on work stress, job satisfaction, and well-being. Motivation and Emotion 23: 125–154.

91.

Rafaeli

Sutton

(1987) Expression of emotion as part of the work role. The Academy of Management Review 12: 23–37.

92.

Rajah

Song

Arvey

(2011) Emotionality and leadership: Taking stock of the past decade of research. The Leadership Quarterly 22(6): 1107–1119.

93.

Rhodes

Greenway

(2010). Dramatis personae: Enactment and performance in primary school headship. Management in Education 24: 149–153. DOI: 10.1177/0892020610379145.

94.

Richardson

Iezzi

Khan

, et al. (2013) Validity and reliability of the Assessment of Quality of Life (AQoL)-8D multi-attribute utility instrument. The Patient – Patient-Centered Outcomes Research: 1–12.

95.

Richardson

Khan

Chen

, et al. (2012) Population norms and Australian profile using the Assessment of Quality of Life (AQoL) 8D utility instrument. Melbourne, Australia: Monash University.

96.

Riley

Langan-Fox

(2013) Bullying, stress and health in school principals and medical professionals: Experiences at the ‘front-line’. In: Burke

Cooper

Fox

(eds) Human frailties: Wrong turns on the road to success Farnham, UK: Gower, pp.181–199.

97.

Riley

(2014) The Australian Principal occupational health, safety and wellbeing survey: 2011–2014 data. [Report]. Available at: http://www.principalhealth.org/au/2011-14%20Report_FINAL.pdf.

98.

Rugulies

Aust

Pejtersen

(2010) Do psychosocial work environment factors measured with scales from the Copenhagen Psychosocial Questionnaire predict register-based sickness absence of 3 weeks or more in Denmark? Scandinavian Journal of Public Health 38: 42–50.

99.

Schaufeli

Bakker

(2010) Defining and measuring work engagement: Bringing clarity to the concept. In: Work Engagement: A Handbook of Essential Theory and Research. New York: Psychology Press, pp.10–24.

100.

Schutz

Aultman

Williams-Johnson

(2009) Educational psychology perspectives on teachers’ emotions. In: Schutz

Zembylas

(eds) Advances in Teacher Emotion Research. Springer US, pp.195–212.

101.

Seery

Corrigall

(2009) Emotional labor: Links to work attitudes and emotional exhaustion. Journal of Managerial Psychology 24: 797–813.

102.

Siegrist

(2001) A theory of occupational stress. In: Dunham

(ed.) Stress in the Workplace: Past, Present, and Future. Philadelphia, PA: Whurr, pp.52–66.

103.

Sloan

(2012) The consequences of emotional labor for public sector workers and the mitigating role of self-efficacy. The American Review of Public Administration 44: 274–290.

104.

Ten Bruggencate

Luyten

Scheerens

, et al. (2012) Modeling the influence of school leaders on student achievement. Educational Administration Quarterly 48: 699–732.

105.

Totterdell

Holman

(2003) Emotion regulation in customer service roles: Testing a model of emotional labor. Journal of Occupational Health Psychology 8: 55–73.

106.

Wang

Seibert

Boles

(2011) Synthesizing what we know and looking ahead: A meta-analytical review of 30 years of emotional labor research. Research on Emotion in Organizations 7: 15–43.

107.

West

Peck

Reitzug

(2010) Limited control and relentless accountability: Examining historical changes in urban school principal pressure. Journal of School Leadership 20: 238–266.

108.

West

Taylor

(2012) Model fit and model selection in structural equation modeling. In: Hoyle

(ed.) Handbook of Structural Equation Modeling. Guilford Press.

109.

Williams

Hartman

Cavazotte

(2010) Method variance and marker variables: A review and comprehensive CFA marker technique. Organizational Research Methods 13: 477–514.

110.

Yilmaz

Altinkurt

Guner

, et al. (2015) The relationship between teachers’ emotional labor and burnout level. Eurasian Journal of Educational Research 59: 75–90.

111.

Zapf

(2002) Emotion work and psychological well-being. A review of the literature and some conceptual considerations. Human Resource Management Review 12: 237–268.

112.

Zikhali

Perumal

(2015) Leading in disadvantaged Zimbabwean school contexts: Female school heads’ experiences of emotional labour. Educational Management Administration and Leadership. DOI 10.1177/1741143214558572, published 27 February 2016.