Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey

G Model

NS-2370; No. of Pages 12 International Journal of Nursing Studies xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Nursing Studies journal homepage: www.elsevier.com/ijns

Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey Klaus-Helmut Schmidt *, Stefan Diestel Leibniz Research Centre for Working Environment and Human Factors at the Technical University of Dortmund, Germany

A R T I C L E I N F O

A B S T R A C T

Article history: Received 18 June 2013 Received in revised form 6 March 2014 Accepted 6 March 2014

Background: Emotional labour is an integral part of the role of nurses with surface and deep acting as the core emotional labour strategies. Surface acting involves the regulation of emotional expression with the aim of bringing it in line with organizational display rules, whereas deep acting aims at modifying the situation or perception of situations in order to change felt emotions, accordingly. Objectives: Since surface acting is thought to consume more cognitive control resources than deep acting, the study examines whether nurses’ cognitive control deficits as a stable personal vulnerability factor exert stronger adverse interactive influences with surface acting on job strain than with deep acting. Design: A cross-sectional questionnaire survey was conducted in combination with absence data referring to a time period of 12 months after administration of the questionnaires. Setting: A hospital and three nursing homes for the elderly located in a federal state in Germany. Participants: Questionnaires were distributed to the whole nursing staff of which 195 nurses filled in the questionnaire and permitted collecting their absence data (73% participation rate). Methods: In addition to descriptive statistics and bivariate correlations, confirmatory factor analyses and hierarchical moderated regression analyses were performed. Results: Findings revealed significant positive relations of surface acting and cognitive control deficits with indicators of job strain, whereas deep acting was unrelated to strain. In addition, surface acting was found to interact stronger with cognitive control deficits in its relation to strain than deep acting. Compared to nurses with low levels of cognitive control deficits, the adverse influences of surface acting on burnout, depressive symptoms, and sum of days absent were much stronger when high levels of cognitive control deficits were reported. Conclusions: Consequently, interventions are needed to inform nurses on the detrimental influences of surface acting and train them in the daily use of deep acting as the healthier emotional regulation strategy. In addition, interventions targeted at the enhancement of the individual capacity for self-control would be beneficial to nurses’ health. ß 2014 Elsevier Ltd. All rights reserved.

Keywords: Surface acting Deep acting Control resources Self-regulation Nursing staff

What is already known about the topic? * Corresponding author at: Leibniz Research Centre for Working Environment and Human Factors at the Technical University of Dortmund, Ardeystr. 67, 44139 Dortmund, Germany. Tel.: +49 2311084327. E-mail address: [email protected] (K.-H. Schmidt).

 Emotional labour is an integral part of the role of nurses.  In performing emotional labour, two strategies are used in order to meet organizational rules of displaying certain emotions and hiding or suppressing others.

http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003 0020-7489/ß 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 2

K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

 Surface acting has been found to exert stronger adverse influences on nurses’ job strain and well-being than deep acting. What this paper adds  In line with the assumption that surface acting puts higher demands on limited self-control resources than deep acting, surface acting was found to be stronger related to burnout, depressive symptoms, and absenteeism, especially when nurses’ cognitive control resources are small.  These observations suggest practical measures aiming at (a) promoting the use of deep acting instead of surface acting and (b) strengthening nurses’ individual control resources. 1. Introduction In the nursing literature, the concept of emotional labour has received increasing attention over the past few years (Hunter and Smith, 2007; Mann, 2005). Emotional labour refers to the regulation of feelings in order to create and express a specific facial and bodily display (Hochschild, 1983). It is widely recognized that emotional labour is an integral part of the role of nurses. In their daily interactions with patients, nurses perform emotional labour when they have to regulate their experience and expression of emotions in order to meet organizational rules of displaying certain emotions and hiding or suppressing others. For example, nurses often experience negative emotions such as disgust, anger, and frustration, the expression of which would not be beneficial in the process of delivering care. Thus, they engage in emotional labour to create or express emotions that match patient or social expectations (Mann, 2005; Smith, 1991). In research on emotional labour, two key observations did emerge. First, emotional labour can be stressful and impair psychological well-being and health, especially when emotions have to be portrayed, which are not genuinely felt. Second, the adverse relations of emotional labour to well-being and health reveal a considerable interindividual variability. To explain this interindividual variability, research has started to analyze emotional labour from a perspective of person-related traits (Heuven et al., 2006). While skills, such as emotional competence (Giardini and Frese, 2006), and personal traits, such as extraversion (Judge et al., 2009), have received increasing attention, the role of employees’ control resources in performing emotional labour has largely been neglected so far (Richards and Gross, 2000). Cognitive control resources refer to one’s limited cognitive capacity required for exerting different cognitive control or self-regulatory processes, such as suppression of emotions, inhibition of habits, or attention regulation. Because, as suggested by the strength model of self-control (Baumeister et al., 2007), people differ in their limited capacity of control resources, some people are stronger adversely affected by emotional labour than others. Accordingly, control resources of people with a small capacity will be faster depleted by emotional labour than control resources of people with a greater capacity. This is

supposed to become manifest in an moderating influence of the individual control capacity on the adverse relations of emotional labour to well-being and health, with corresponding large amounts of interindividual variability in that relations. To assess interindividual differences in the limited control capacity, daily cognitive control deficits in the form of self-reported failures in perception, action, self-regulation, and affective control have been repeatedly used as a valid measure for that stable trait (Broadbent et al., 1982). However, not all emotional labour processes exert adverse influences on well-being and health to the same extent. On the basis of Hochschild’s (1983) distinction between surface and deep acting as core emotional labour strategies, two recent meta-analyses revealed substantial positive relationships between surface acting and indicators of impaired personal and job-related well-being and a small negative relationship with performance outcomes. In contrast to this, deep acting was found to be only weakly related to indicators of impaired well-being, but positively associated with emotional performance and customer satisfaction (Hu¨lsheger and Schewe, 2011; MesmerMagnus et al., 2012). Surface acting involves the regulation of emotional expression with the aim of bringing it in line with given display rules. According to Gross’s (1998) process model of emotional regulation, surface acting is a response-focused form of emotional regulation that leaves genuinely felt emotions unchanged. In contrast, deep acting as an antecedent-focused form of emotional regulation (Gross, 1998) aims at modifying the situation or perception of situations in order to change felt emotions according to external display rules. As a result of engaging in deep acting, required and genuinely felt emotions harmonize (Holman et al., 2008). Whereas most of the meta-analyzed studies focused on customer service employees, in nursing contexts quite similar differential relationships of both emotional labour strategies have been identified. Among a sample of registered nurses of two urban hospitals in the Midwestern United States, Diefendorff et al. (2011) observed surface acting to be positively linked to burnout (emotional exhaustion) and negatively linked to job satisfaction. Deep acting showed weaker relations to burnout and no associations with nurses’ job satisfaction. Pisaniello et al. (2012) and Chou et al. (2012) have provided further evidence for differential links of both emotional labour strategies and indicators of job strain and well-being in nursing settings. To further advance our understanding of the differential relationships of both emotional labour strategies and drawing on the strength model of self-control (Baumeister et al., 2007), several scholars have argued that engaging in surface acting puts higher demands on self-control resources than engaging in deep acting and thus depletes ˜ igo et al., 2007; resources to a greater extent (Martı´nez-In Totterdell and Holman, 2003). Whereas surface acting involves the continuous monitoring of genuinely felt and required emotions combined with the continuous investment of mental effort to alter ones emotional expression, deep acting calls for control resources only at the onset of

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

an emotion by deploying attention or reappraising situations in order to induce the expected emotion (Grandey, 2000; Gross, 1998). Since the resource demands of surface and deep acting cannot be directly assessed thus far, the present study adopts a moderation approach to provide evidence for the differential regulatory features of both emotional labour strategies. According to Inzlicht and Schmeichel (2012), analyzing a moderatoring effect can provide a wellfounded understanding of underlying processes, especially when a putative mediator cannot be assessed (see also Spencer et al., 2005). We propose that surface acting consumes more control resources than deep acting and, thus, exerts stronger adverse influences on employees’ well-being and health. Our moderator approach integrates the argument of differential resource demands of surface and deep acting and the assumption of a limited individual capacity of self-control resources. On the basis of the integration of both arguments, we predict that the limited, individual control capacity exerts a stronger moderating influence on the relations of surface acting to well-being and health than on the corresponding relations of deep acting, because surface acting will tax the limited individual capacity to a greater extent than deep acting. In the following, we first review the literature on emotional labour strategies. Then, the concept of cognitive control deficits as an indicator of the individual control capacity will be briefly discussed. Finally, we integrate emotional labour and control resources research and develop the hypotheses in more detail. 1.1. Emotional labour strategies In order to explain the differential relationships of surface acting and deep acting with job strain (e.g., burnout), job-related well-being (e.g., job satisfaction), and performance outcomes (e.g., customer satisfaction), several theoretical arguments have been developed (for reviews, see Holman et al., 2008; Hu¨lsheger and Schewe, 2011). First, since surface acting involves by its very nature the experience of a discrepancy between genuinely felt and displayed emotions, surface acting is thought to threaten and impair employees’ sense of self-authenticity, with corresponding adverse influences on job strain and well-being (Ashforth and Humphrey, 1993). In contrast, engaging in deep acting is not associated with such kinds of discrepancy experiences and resulting impairments of well-being. Second, the discrepancy between experienced and displayed emotions is also thought to impact other persons’ perception of the authenticity of emotion workers. Experimental and field studies have found other persons to be able to differentiate between authentic and inauthentic emotional expression, and only a perceived authentic emotion stimulates the desired reactions of interaction partners such as patients or clients (HennigThurau et al., 2006). Since surface acting is associated with inauthentic emotional behaviour, it was expected and found to affect interaction partners negatively, cause interpersonal conflicts, and, as a consequence, impair emotion workers’ own well-being and performance (Holman et al., 2008).

3

Finally, several scholars have argued that (like any other forms of self-control) emotional labour draws on and depletes limited control resources (Baumeister et al., 1998; Hagger et al., 2010) and that surface acting should consume more of those resources than deep acting (Holman et al., 2008; Totterdell and Holman, 2003). As mentioned above, surface acting involves the expression of emotions which are not genuinely felt in order to meet organizationally desired display rules. This implies continuous monitoring of felt and desired emotions, inhibition or suppression of experienced emotions and a continuous altering of the emotional expression. All these processes can be thought to be effortful and drain control resources. And indeed, findings from laboratory research have revealed that suppressing felt emotions and exaggerating a required emotional display are effortful processes depleting limited control resources (for reviews, Gross, 1998; Schmeichel, 2007). With regard to consequences for well-being and health, frequent exertion of self-control in the form of surface acting have been repeatedly observed to result in psychological and physiological strain (Gross and Levenson, 1997). For example, Robinson and Demaree (2007) found portraying emotions that are not genuinely felt to cause high sympathetic arousal and impaired well-being. Sympathetic arousal is a physiological process associated with a relative activation of the sympathetic nervous system, which manifests in responses, such as increased heart rate, increased arterial blood pressure, and increased glucose levels released by the liver for extra energy. All these responses reflect elevated mental and physical effort. As demonstrated by Gross and Levenson (1997), these increased effort investments are, in turn, strongly related to impaired well-being and health. Thus, in the case of surface acting, employees engage in self-regulatory efforts consuming and depleting cognitive control resources. In contrast, deep acting has been thought to consume not as much resources than surface acting (Totterdell and Holman, 2003). This notion is mainly derived from findings showing that reappraisal as a component of deep acting requires less effort than emotion suppression as a component of surface acting (Richards and Gross, 1999, 2000). However, whether deep acting actually puts lower demands on control resources than surface acting is an open question so far. This is mainly due to the fact that the resource demands of both emotional labour strategies elude an immediate measurement. To by-pass this measurement problem, the present study provides a moderating approach to disclose the resource demands of surface and deep acting and thus shed light on their differential relationships with job strain and well-being. This approach draws on the empirical observation that people differ in their capacity to exert selfcontrol in general (Baumeister et al., 2007) and to control emotions in particular (Judge et al., 2009). 1.2. Cognitive control capacity A growing body of evidence on self-regulatory functioning and control resources suggests that cognitive control deficits in the form of daily failures in perception,

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 4

K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

action, self-regulatory and affective control are a valid indicator for the individual capacity of control resources (Broadbent et al., 1982; McVay and Kane, 2009). Cognitive control deficits involve a broad spectrum of impairments in cognitive control that encompasses, for example, switching between tasks, monitoring multiple sources of information, resisting distractions and sudden impulses, as well as hiding or suppressing emotions (Bridger et al., 2011). For example, the inability to concentrate on complex tasks or to overcome frustrations and inner blockades are typical manifestations of cognitive control deficits. In addition, a high interindividual stability of cognitive control deficits has been repeatedly observed, which led most researchers to conclude that those deficits constitute a person-related trait reflecting stable interindividual differences in the capacity of control resources (Broadbent et al., 1982; Wallace, 2004). Consequently, people with a lower control capacity should experience higher levels of job strain and impaired well-being due to meeting self-control demands than those with a greater control capacity. In broader terms, the risk of experiencing psychological strain and impaired well-being can be thought to increase with increasing control demands and decreasing employees’ control capacity. Statistically, corresponding relationships reflect an interaction effect between individual control capacity and job-related control demands. In line with these theoretical arguments and empirical findings, Schmidt et al. (2007) observed that cognitive control deficits interact with job-related self-control demands in the prediction of job strain. The measured demands addressed the extent to which a given job requires employees to suppress and inhibit spontaneous response tendencies and associated affect states in order to maintain a controlled and purposive behaviour. Increasing levels of cognitive control deficits strengthened the adverse influence of those demands on facets of burnout. That is, cognitive control deficits were found to be a person-related vulnerability factor, which strengthens the relation of self-control demands to job strain (Broadbent et al., 1982; Bridger et al., 2011). Employees with high control deficits are obviously more susceptible to the adverse influence of self-control demands and, thus, show higher levels of strain than those with low control deficits. Consequently, high cognitive control deficits indicate that an employee does not have enough resources or suffers from an insufficient capacity for meeting the regulatory requirements of a given job role. 1.3. Conclusions and predictions Integrating both lines of research, the following arguments appear to be obvious: on the one hand, research on emotional labour strategies suggests that surface acting draws on and depletes control resources to a greater extent than deep acting and, thus, exerts stronger adverse influences on job strain and well-being. On the other hand, research on self-regulation suggests that (a) people differ in their (limited) control capacity as indicated by control deficits, and (b) high control deficits interact with control demands in predicting job strain in such a way that

high control deficits amplify the relation of control demands to strain. On the basis of these arguments, we derive the following hypothesis: in comparison with the interaction effect of cognitive control deficits and deep acting on job strain and well-being, the interaction of cognitive control deficits with surface acting explains greater amounts of variance in job strain and well-being. In other words, we expect differential interactive influences of cognitive control deficits with surface and deep acting. 1.4. The present study In the present study, we examine interaction effects of cognitive control deficits with either surface or deep acting on emotional exhaustion and depersonalization (the core dimensions of burnout), depressive symptoms, and absence behaviour as outcomes. Burnout is an oftenstudied strain outcome of emotional labour. Emotional exhaustion as a ‘‘state of depletion and fatigue’’ (Grandey, 2003, p. 89) is the most direct manifestation of the energy exertion aspect of emotional labour and reflects chronic impairments of the limited cognitive resource (Diestel et al., 2013; van Dam et al., 2011). In comparison with exhaustion, depersonalization is more an attitude by which individuals try to prevent further losses of energy by retreating emotionally and viewing others as objects rather than persons. As previous research has found personal accomplishment to be only weakly related to emotional labour strategies (Hu¨lsheger and Schewe, 2011), this third dimension of burnout was not considered as an outcome. Recently, research on organizational stress has also focused on depressive symptoms as a result of emotional labour. According to the control theory of depression (Hyland, 1987), depressive symptoms result from chronic self-regulatory failures due to decrements in the regulatory resource. In support of this view, Oaten and Cheng (2005, 2006) have found continuous self-control efforts to longitudinally manifest in several forms of psychological strain, such as somatic complaints and depressive symptoms. Consequently, the adverse effects of frequent selfcontrol and repeated resource depletion do not only limit to immediate impairments of well-being, but also become chronically manifest in psychological strain over longer time periods, especially when certain circumstances – such as recurrent demands on self-control – prevent recovery of the limited resource capacity (Baumeister et al., 2006). Finally, resource perspectives (Muraven and Baumeister, 2000) and conservation of resources theory (Hobfoll, 1989, 2002) suggest including absenteeism as a third outcome of the present study. Both lines of theorizing converge in the notion that when valued resources are threatened or lost as a result of emotional labour, employees will strive to prevent further resource losses by withdrawing from resource depleting situations. Furthermore, Staw and Oldham (1978) claimed that absenteeism can fulfil a ‘‘maintenance’’ or ‘‘restorative’’ function, which serves to recovery of a depleted regulatory resource. In a recent study among nurses of a public hospital in Australia, Nguyen et al. (2014) found evidence

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

in favour of these arguments by demonstrating direct effects of surface acting on absenteeism (i.e., number of days absent). Nurses who reported engaging in more surface acting had higher absenteeism during the subsequent 12 months and thus were away from their workplace for longer periods of time. Similar relationships have been found for other forms of withdrawal such as turnover intentions and turnover (Chau et al., 2009; Goodwin et al., 2011).

5

Wallace, 2004). Each of the 12 items refers to a particular type of failure (e.g., forgetting names and appointments, dropping things, failing to regulate attention), and participants are asked to indicate how often they have made that particular error within the last four weeks (e.g., ‘‘How often did you lose your temper and regret it?’’; ‘‘How often did you start doing one thing and got distracted into doing something else?’’). All items are answered on a fivepoint Likert-format covering a range from 1 (never) to 5 (very often).

2. Method 2.1. Participants and procedure The present study was conducted among a sample of nurses, employed at one hospital and three nursing homes for the elderly located in a federal state of Germany. Participants were recruited through announcements at staff meetings and an individual invitation letter sent jointly by the management and local members of work council. A participation rate of 73% yielded a final sample of 195 nurses who completed a questionnaire and gave their consent for using their absence data. Questionnaires were administered in small groups of about 15 persons during normal working hours. During the group sessions, a member of the research team was present who collected the completed questionnaires. All participating nurses were assured that their data would remain confidential. Matching responses with individual absence data was made possible through an individual code number that remained exclusively in the hands of the researchers. The age of participants ranged from 20 to 58 years (M = 37.29, SD = 10.60). The sample comprised 85% women and 34% full-time employees. All participants had been trained as hospital or geriatric nurses. 2.2. Measures 2.2.1. Emotional labour A German version of Brotheridge and Lee’s (2003) sixitem Emotional Labour Scale was used to assess surface acting and deep acting. The three items measuring surface acting refer to modifying and faking expressions (e.g., ‘‘How often do you pretend to have emotions that you don’t really have?’’). The three items measuring deep acting tap the extent to which employees modify inner feelings to meet display rules (e.g., ‘‘How often do you try to actually experience the emotions that you must show?’’). The response format of all items covers a range from 1 (never) to 5 (always). 2.2.2. Cognitive control deficits As an indicator of cognitive control deficits, the frequency of everyday slips and errors in perception, memory, and action was assessed using a German version of the Cognitive Failures Questionnaire developed by Broadbent et al. (1982). Several factor analyses and psychometrical tests across different samples have repeatedly provided strong evidence for the construct validity of this instrument in terms of dimensionality as well as convergent and divergent validity (Larson et al., 1997;

2.2.3. Indicators of strain The two burnout dimensions of emotional exhaustion and depersonalization were measured by a German translation of the Maslach Burnout Inventory (Maslach and Jackson, 1984, 1986). Emotional exhaustion (nine items) refers to feelings of being overextended and drained by psychological work demands (e.g., ‘‘I feel emotionally drained from my work’’). Depersonalization (five items) is characterized by a detached, indifferent, and cynical attitude towards people with whom one has to interact at work (e.g., ‘‘I have become more callous towards people since I took this job’’). All items are scored on a six-point frequency rating format ranging from 1 (never) to 6 (very often). We assessed depressive symptoms with a shortened German version of the Beck Depression Inventory (Beck et al., 1988). The 15 items address various symptoms such as reduced initiative, irritation, sadness and listlessness. Intensity/severity of symptoms is rated on a six-point frequency scale with a range from 0 (never) to 5 (very often). Sum of days absent (the total sum of days absent from work) was used as a measure of absence behaviour. This measure did not count absences due to holidays, maternity leave and participation in training courses and as such would reflect more involuntary absence events (e.g., certificated sick days owing to illness) rather than voluntary absences (Johns, 2011). According to German state regulations, all employees have to call in sick the first day, followed by the presentation of a sickness-certificate issued by a doctor if an absence period comprises more than two days. The corresponding data were drawn from personal records and related to a period of 12 months after the survey. Since the distribution of the absence measure deviated from normality, leading to biases in parameter estimations, all raw scores were subjected to a square root transformation (see Clegg, 1983). After transformation, skewness and kurtosis were less than 1 and 2, respectively, and thus met the criteria for applying covariance-based analyses (see Geurts et al., 1994). 2.2.4. Control variables As age, gender, and working time status have been found to account for differences in strain measures and absenteeism (Cheung and Tang, 2007; Gonza´lez-Morales et al., 2010), these variables were also collected and included in our analyses to control for the possibility of spurious relations among the study variables. Working time status was assessed as dichotomous variable (1 = part-time; 2 = full-time). In line with German regulations, full-time

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

6

two-way interaction terms (i.e., cross-products of the main predictors). Prior to the computation of the interaction terms, the predictors were centred around their mean scores to deal with problems of multicollinearity that may arise from cross-product terms (see Aiken and West, 1991).

employees work 38.5 h per week, whereas part-time employment covers 19.25 h work per week. In addition, we included negative affectivity as a further control variable because those nurses who have dispositionally negative emotions may report higher levels of emotional labour (e.g., Abraham, 1998) as well as stronger strain symptoms (e.g., Burke et al., 1993). Negative affectivity was assessed with seven items selected from the corresponding subscale of a validated German version (Krohne et al., 1996) of a questionnaire developed by Watson et al. (1988). Respondents are asked to indicate the frequency to which they generally experience a variety of feelings including ‘‘distressed’’, ‘‘jittery’’, and ‘‘nervous’’. The response format of all items covers a range from 1 (never) to 5 (very often).

3. Results 3.1. Descriptive statistics and dimensionality of predictors and outcomes Means, standard deviations, reliability coefficients, and intercorrelations of all study variables are presented in Table 1. As expected, cognitive control deficits and surface acting were positively related to all outcomes, whereas deep acting revealed no significant associations. Furthermore, surface and deep acting displayed a weak and not significant positive intercorrelation. Confirmatory factor analyses provided support for the differentiability between both of the emotional labour strategies, cognitive control deficits, and the three indicators of psychological strain (exhaustion, depersonalization, and depressive symptoms). A 6-factor model, treating deep acting, surface acting, cognitive control deficits, and the three strain outcomes as separate dimensions, yielded an adequate data approximation (x2 (174) = 238.04, p < .01, RMSEA = .04 (90% CI = .03–.06), CFI = .96, SRMR = .04, Gamma Hat = .97). All other models, which integrate several factors into one construct, showed a worse fit (4-factor model with the three strain outcomes as one factor: x2 (183) = 350.53, p < .01, RMSEA = .07 (90% CI = .06–.08), CFI = .91, SRMR = .06, Gamma Hat = .92; 4factor model with all predictors as one factor: x2 (183) = 640.46, p < .01, RMSEA = .11 (90% CI = .10–.12), CFI = .74, SRMR = .10, Gamma Hat = .81; 5-factor model with both emotional labour strategies as one factor: x2 (179) = 416.80, p < .01, RMSEA = .08 (90% CI = .07–.09), CFI = .87, SRMR = .08, Gamma Hat = .89; 2-factor model with one predictor factor and one strain factor: x2

2.3. Statistical analyses Before testing the hypotheses, we conducted confirmatory factor analyses for both predictors and outcomes in order to analyze potential confounding effects of common method variance. The corresponding measurement models were evaluated on the basis of conventional fit indices (see Schermelleh-Engel et al., 2003). Then, for answering the question whether surface acting and deep acting interact with cognitive control deficits in predicting strain outcomes, two sets of hierarchical moderated regression analyses (Aiken and West, 1991) were performed separately for each outcome. The first set served to test the main effects of and interaction between cognitive control deficits and surface acting; for the second set, deep acting was included instead of surface acting. The predictor variables were entered into the equations in three successive steps. In the first step, demographic background variables (age, gender, working time status) and negative affectivity were introduced. The next step implied the entry of the main predictor variables (cognitive control deficits either combined with surface acting or deep acting), followed by the introduction of the respective

Table 1 Descriptive statistics, internal consistencies, and intercorrelations of all study variables. Variable

Intercorrelations 1

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

Age Gendera Working time statusb Negative affectivity Surface acting Deep acting Control deficits Emotional exhaustion Depersonalization Depressive symptoms Sum of days absent M SD

– .11 .22** .00 .00 .08 .03 .11 .08 .06 .09 37.29 10.60

2

3

4

5

6

7

8

9

10

– .03 .15* .01 .23** .08 .04 .01 .08 .10

– .06 .01 .11 .02 .01 .03 .08 .02

(.79) .44** .02 .55** .61** .39** .66** .28**

(.77) .10 .35** .45** .42** .40** .28**

(.80) .11 .01 .05 .04 .08

(.77) .47** .44** .58** .34**

(.81) .56** .71** .24**

(.64) .42** .13

(.92) .29**

1.85 0.36

1.34 0.48

2.30 0.51

2.61 0.71

2.94 0.81

2.22 0.55

2.22 0.79

1.92 0.77

1.02 0.76

11

7.91 14.43

Note: N = 195. Descriptive statistics of absence data represent nontransformed scores. Internal consistency estimates (Cronbach’s alpha) are in the parentheses on the diagonal. a Gender (1 = male, 2 = female). b Working time status (1 = part-time, 2 = full-time). * p < .05 (two-tailed test). ** p < .01 (two-tailed test).

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

7

associations with surface acting. An increase in surface acting went along with higher levels of exhaustion, depersonalization, and absenteeism. More importantly, however, the introduction of the two-way interaction between surface acting and cognitive control deficits in Step 3 did also add a significant amount of incremental variance to the prediction of at least three of the strain measures (exhaustion: DR2 = .05, p < .01; depressive symptoms: DR2 = .04, p < .01; sum of days absent: DR2 = .04, p < .01). Only in the measure of depersonalization, the corresponding interaction failed to reach significance. The results of the regression analyses with deep acting and cognitive control deficits are summarized in Table 3. As before, the entry of both predictors in Step 2 contributed a significant amount of incremental variance to the prediction of each outcome. In the final regression equations, cognitive control deficits again had significant positive beta weights. In contrast, deep acting displayed no significant associations with the outcomes. Finally, no significant interaction between deep acting and cognitive control deficits did emerge in any of the outcomes. To facilitate the interpretation of the significant interaction effects of surface acting and cognitive control deficits, interaction plots were generated using the method recommended by Aiken and West (1991). Accordingly, values of the predictors were chosen one standard deviation above and below the standardized means. Then, for each outcome measure reflecting a significant interactive effect, simple regression lines were generated by inserting these values into the regression equation. The resulting plots are depicted in Fig. 1, supplemented by the results of simple slope tests for examining whether the slopes differed significantly from zero. It is evident from the figure that surface acting and cognitive control deficits exerted a quite similar interactive influence on exhaustion, depressive symptoms, and sum of days absent (square root transformed). Compared to nurses with low levels of cognitive control deficits, the adverse effects of surface acting on that outcomes were much stronger when high levels of cognitive control

(188) = 744.67, p < .01, RMSEA = .12 (90% CI = .11–.13), CFI = .69, SRMR = .11, Gamma Hat = .78). In addition, we tested a 7-factor model with an unmeasured, latent factor for all indicators (Podsakoff et al., 2003). The unmeasured, latent method factor test showed that biasing or confounding effects due to common method variance are rather unlikely, because no substantial improvement in model fit could be confirmed (x2 (154) = 203.33, p < .01, RMSEA = .04 (90% CI = .02–.06), CFI = .97, SRMR = .04, Gamma Hat = .97). For both cognitive control deficits and the three outcomes, we were required to parcel the items due to the large number of items (Little et al., 2002). Our parcelling procedure draws on the item-to-construct balance method that combines lower loaded items with higher loaded items and thus reduces the loading differences between the manifest variables. In particular, we created four parcels (each with three items) to estimate cognitive control deficits. The latent variance of emotional exhaustion was estimated on the basis of three parcels (each with three items), whereas two parcels (each with two items) were specified to reflect depersonalization. Finally, to determine the latent variance of depressive symptoms, we constructed three parcels (each with five items). 3.2. Moderated regression analyses The results for cognitive control deficits and surface acting as main predictors are summarized in Table 2. As can be seen, the background variables and negative affectivity (Step 1) accounted for a significant proportion of variance in all outcomes. The entry of surface acting and cognitive control deficits in Step 2 contributed a significant amount of incremental variance to the prediction of each outcome, ranging from 6% (for emotional exhaustion) to 12% (for depersonalization). In the final equations for all outcomes, cognitive control deficits displayed significant positive beta weights. Thus, an increase in cognitive control deficits was associated with higher levels of strain. With the exception of depressive symptoms, all other outcome measures did also reflect significant positive

Table 2 Results from hierarchical moderated regression analyses: main and interaction effects of surface acting and cognitive control deficits on indicators of job strain, controlling for background factors and negative affectivity. Regression steps and effect tested

Emotional exhaustion

Depersonalization

DR2

b

DR2

b

Depressive symptoms

Sum of days absent

DR2

b

DR2

b

1.

Age Gender Working time status Negative affectivity

.01** .13 .15 .31**

.39**

.00 .05 .05 .09

.17**

.00 .04 .04 .32**

.44**

.02 .36 .26 .05

.09**

2.

Surface acting Control deficits

.15** .12*

.06**

.20** .21**

.12**

.07 .21**

.07**

.37** .48**

.07**

3.

Interaction

.15**

.05**

.04

.01

.13**

.04**

.36**

.04**

R2 F

.50 26.22**

.30 11.24**

.55 32.64**

.20 6.48**

Note: N = 195. * p < .05. ** p < .01.

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

8

Table 3 Results from hierarchical moderated regression analyses: main and interaction effects of deep acting and cognitive control deficits on indicators of job strain, controlling for background factors and negative affectivity. Regression steps and effect tested

Emotional exhaustion

Depersonalization

DR2

b

DR2

b

Depressive symptoms

Sum of days absent

DR2

b

DR2

b

1.

Age Gender Working time status Negative affectivity

.01** .19 .14 .41**

.39**

.00 .09 .07 .18**

.17**

.00 .06 .05 .38**

.44**

.02 .13 .15 .21

.09**

2.

Deep acting Control deficits

.01 .15**

.03**

.02 .24**

.07**

.01 .23**

.07**

.27 .66**

.06**

Interaction

.03

.00

.00

.00

.02

.00

.12

.00

3.

R2 F

.42 19.57**

.24 8.31**

.51 32.64**

.15 4.85**

Note: N = 195. ** p < .01.

deficits were reported. As can be seen from the respective regression lines, the simple slopes analyses indicate that the positive relations of surface acting to exhaustion, depressive symptoms, and sum of days absent were only significant at high levels of cognitive control deficits.

4. Discussion The present study examined whether cognitive control deficits as an individual vulnerability factor moderate (amplify) the relation of surface and deep acting with job strain and well-being in different strengths, depending on the emotional labour strategy used. The conceptual background of the study draws on recent theoretical developments in emotional labour and self-control research, according to which we propose that (a) surface acting depletes and consumes more control resources than deep acting, (b) cognitive control deficits as a stable trait reflect a low and limited individual control capacity and as such (c) in interaction with surface acting exert stronger adverse influences on job strain and well-being than in interaction with deep acting. Consequently, cognitive control deficits were expected to make nurses more vulnerable to the adverse impact of emotional labour,

especially when surface acting is used compared with deep acting. The results of our study provide strong support for that main prediction. In interaction with cognitive control deficits, surface acting was found to produce stronger adverse influences on emotional exhaustion, depressive symptoms, and absenteeism than deep acting. By revealing such differential interactive influences, the study suggests that surface acting indeed consumes more control resources than deep acting, which may explain their differential relationships with job strain and well-being. However, in contrast to the other outcomes considered, depersonalization did not reflect a similar interaction of surface acting with cognitive control deficits. From a methodological point of view, this finding may be explained by the non-satisfying operationalization of this burnout symptom (Salanova et al., 2005). In our sample (and in many others from different work settings), Cronbach’s alpha was rather low indicating that the participating nurses responded to the items inconsistently. Thus, the depersonalization scale may not be sensitive enough to reflect a clear and homogeneous state of strain. From a (related) theoretical point of view, given the multiple functions of depersonalization in developing burnout (Diestel and Schmidt, 2010; Taris et al., 2005),

Fig. 1. Interaction effects of surface acting and control deficits on emotional exhaustion, depressive symptoms, and sum of days absent.

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

this burnout symptom has been repeatedly thought to reflect a dysfunctional coping strategy aiming at reducing feelings of exhaustion by retreating and regarding persons with whom one has to interact at work as objects (Salanova et al., 2005). In line with this idea, depersonalization has been found to moderate relations among other burnout symptoms (Diestel and Schmidt, 2010). However that may be, the theoretical status of depersonalization appears to be rather unclear and needs further clarification. Drawing on the present results, our research offers several contributions to the existing knowledge. First, our research extends recent experimental findings from the laboratory to organizational contexts such as nursing. Experimentally, the psychological costs of emotional labour have been revealed to depend on both the specific emotional labour strategy used and a limited control resource capacity, which differs between individuals (Schmeichel et al., 2008). Moreover, a low control resource has been shown to become manifest in cognitive control deficits. Research on emotional labour in nursing contexts has largely failed to consider both findings thus far. Second, in the face of a continuous rise of the services sector in general, and health care in particular, our findings suggests interpreting cognitive control deficits as an indicator for a lack of resources and capacities in performing emotional labour, which becomes manifest in job strain especially when the strategy of surface acting is used. Finally, the differential interactive influences of cognitive control deficits and both emotional labour strategies on absenteeism also imply organizational costs. It is just the combination of cognitive control deficits and surface acting that cause considerable financial costs and thus affect organizational efficiency. 4.1. Limitations and suggestions for future research Of course, the present study has certain limitations. First, most of the study variables were assessed by selfreports. Thus, common method variance or a self-report bias might have contaminated the relations observed (Podsakoff et al., 2003). However, the unmeasured latent method factor test revealed common method influences to be not a severe biasing factor in our data. Furthermore, the differential influences resulting from the combination of cognitive control deficits with either surface acting or deep acting are unlikely to be attributable to common method variance because such variance tends to blur differential relationships (Demerouti et al., 2005). Finally, using a measure of absenteeism that reflected a similar pattern of differential relationships as did the self-report measures of exhaustion and depressive symptoms, largely mitigated the risk of mutual contamination of the constructs. Nevertheless, future research could gain more methodological clarity and practical significance by applying more event-related and situation-based approaches for measuring the use of emotional labour strategies (Reis and Gable, 2000). Furthermore, cognitive control deficits as a trait measure of the individual control resource capacity may benefit from a more precise and careful elaboration of its nomological network. Although that measure has been

9

found to have good psychometrical properties and is wellestablished in both basic and applied research (Bridger et al., 2011; McVay and Kane, 2009), it is widely unknown how cognitive control deficits are related to other individual trait measures. Diefendorff et al.’s, (2000) observation that individual differences in the ability to initiate, maintain and protect intentions from competing action tendencies share small, yet significant portions of variance with cognitive control deficits set an example of how to achieve a better understanding of the specific scope and content of that construct. Beyond that, instead of using with cognitive control deficits a proxy self-report measure of the individual control resource, (short) standardized tests of cognitive functioning might be an alternative and more direct approach to assess limited control resources (Schmeichel et al., 2008). In addition, the relationships of surface acting (alone and in interaction with cognitive control deficits) with the strain outcomes may be argued to be spurious and emerged due to the influence of third variables (Bono and Vey, 2005). For example, the trait-like construct of negative affectivity may operate as such a third variable. Indeed, research has repeatedly revealed that employees’ negative affectivity is positively associated with both surface acting (Brotheridge and Lee, 2002) and indicators of job strain (Brief et al., 1988). However, our results clearly show that even after controlling for negative affectivity, the hypothesized relationships of surface acting did emerge. Similarly, one might argue that other third variables may also influence both nurses’ cognitive control deficits and job strain, or even that job strain itself may exerting influences on control deficits. For example, Diestel et al. (2013) observed that nurses with high and low levels of burnout (emotional exhaustion) differed significantly in their performance of tasks putting high demands on cognitive control. Although the cross-sectional design of this study prevents strong causal inferences, van Dam et al. (2011) have offered an explanation of this observation, which is based on a complex pattern of multi-directional cause–effect relations. Accordingly, high control demands at work can lead to both burnout and impaired control because of increased allostatic load and associated deficits in the limited control capacity. However, since individuals differ in their control capacity and thus in the ability to exert control, low control capacity can amplify the adverse effect of control demands on strain and well-being. Longitudinal studies would be best suited for struggling against validity threats by third variable influences and for disclosing the exact causal mechanisms underlying emotion work. The need for more longitudinal studies combined with control of third variable influences is, last but not least, emphasized by the fact that the few existing longitudinal studies have revealed mixed results. Whereas, for example, Hu¨lsheger et al. (2010) observed surface acting to lead to increases in subsequent strain without indication of reverse causation, Philipp and Schu¨pbach (2010) found ‘‘standard’’ and reverse effects as well as jointly occurring ‘‘standard’’ and reverse relationships of surface acting with job strain.

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

10

4.2. Practical implications

Ethical approval

The current findings show that meeting emotional labour demands is not always and not unconditionally associated with high levels of job strain and impaired wellbeing. Such negative outcomes do only emerge when nurses tend to adopt surface acting as preferred strategy and when they have low control resources at their disposal. In contrast, deep acting is obviously not harmful for nurses and even the adverse influence of surface acting can be buffered by high amounts of control resources as indicated by low cognitive control deficits. These observations suggest practical measures aiming at (a) promoting deep acting instead of surface acting, and (b) strengthening nurses’ control capacity. Consequently, interventions are needed to inform nurses on the detrimental effects of surface acting and train them in the daily use of deep acting as the healthier emotional regulation strategy. Such interventions seem to be promising as both emotional labour strategies are not stable, trait-like individual characteristics but can rather be adopted flexibly (Philipp and Schu¨pbach, 2010). In addition, interventions targeted at the enhancement of the resource capacity for self-control have been found to be effective. For example, Oaten and Cheng (2006) had participants enter a two-month physical exercise programme that was intended to improve their selfcontrol. The programme was tailored to suit the individual by gym staff, and included aerobic classes, free weights, and resistance training. After that programme, participants showed considerable improvements in laboratory self-control tasks. Participants also reported significant decreases in perceived stress, emotional distress, smoking, alcohol and caffeine consumption, and an increase in healthy eating, emotional control and attendance to commitments. The idea that the capacity for self-control can be improved is therefore of considerable practical importance, not at least in the field of nursing. Besides such interventions, field studies on emotional labour have identified several organizational and personal variables which buffer the adverse effects of emotional labour demands on job strain and well-being, such as emotional support from supervisors and colleagues (de Jonge et al., 2008) or emotional competence of employees (Giardini and Frese, 2006; Bulmer Smith et al., 2009). Consequently, both factors provide additional starting points for the development of preventive interventions.

As this was a questionnaire-based study of nursing staff and had no patient involvement, approval from an ethics committee is not required in Germany.

Conflict of interest None declared.

Funding The present research received no specific grant from any funding agency in the public, commercial, or non-forprofit sectors.

References Abraham, R., 1998. Emotional dissonance in organizations: antecedents, consequences, and moderators. Genet. Soc. Gen. Psychol. Monogr. 124 (2), 229–246. Aiken, L.S., West, S.G., 1991. Multiple Regression: Testing and Interpreting Interactions. Sage, Newbury Park, CA. Ashforth, B.E., Humphrey, R.H., 1993. Emotional labour in service roles: the influence of identity. Acad. Manage. Rev. 18 (1), 88–155. Baumeister, R.F., Bratslavsky, E., Muraven, M., Tice, D.M., 1998. Ego depletion: is the active self a limited resource? J. Pers. Soc. Psychol. 74 (5), 1252–1265. Baumeister, R.F., Gailliot, M., DeWall, C.N., Oaten, M., 2006. Self-regulation and personality: how interventions increase regulatory success, and how depletion moderates the effects of traits on behavior. J. Pers. 74 (6), 1773–1802. Baumeister, R.F., Vohs, K.D., Tice, D.M., 2007. The strength model of selfcontrol. Curr. Direct. Psychol. Sci. 16 (6), 351–355. Beck, A.T., Steer, R.A., Garbin, M.G., 1988. Psychometric properties of the Beck Depression Inventory: twenty-five years of evaluation. Clin. Psychol. Rev. 8 (1), 77–100. Bono, J.E., Vey, M.A., 2005. Toward understanding emotional management at work: a quantitative review of emotional labour research. In: Hartel, C.E., Zerbe, W.J., Ashkanasy, N.M. (Eds.), Emotions in Organizational Behavior. Lawrence Erlbaum Associates Publishers, Mahwah, NJ, pp. 213–233. Bridger, R.S., Brasher, K., Dew, A., Sparshott, K., 2011. Cumulative psychological strain and future strain in naval personnel: is executive function the elephant in the room? Ergonomics 54 (7), 597–608. Brief, A.P., Burke, M.J., George, J.M., Robinson, B., Webster, J., 1988. Should negative affectivity remain an unmeasured variable in the study of job stress? J. Appl. Psychol. 73 (2), 193–198. Broadbent, D.E., Cooper, P.F., FitzGerald, P., Parker, K.R., 1982. The Cognitive Failures Questionnaire (CFQ) and its correlates. Br. J. Clin. Psychol. 21 (1), 1–16. Brotheridge, C.M., Lee, R.T., 2002. Testing a conservation of resources model of the dynamics of emotional labour. J. Occup. Health Psychol. 7 (1), 57–67. Brotheridge, C.M., Lee, R.T., 2003. Development and validation of the emotional labour scale. J. Occup. Org. Psychol. 76 (3), 365–379. Bulmer Smith, K., Profetto-McGrath, J., Cummings, G.G., 2009. Emotional intelligence and nursing: an integrative literature review. Int. J. Nurs. Stud. 46 (12), 1624–1636. Burke, M.J., Brief, A.P., George, J.M., 1993. The role of negative affectivity in understanding relations between self-reports of stressors and strains: a comment on the applied psychology literature. J. Appl. Psychol. 78 (3), 402–412. Chau, S.L., Dahling, J.J., Levy, P.E., Diefendorff, J.M., 2009. A predictive study of emotional labor and turnover. J. Org. Behav. 30 (8), 1151– 1163. Cheung, F.-Y., Tang, C.S., 2007. The influence of emotional dissonance and resources at work on job burnout among Chinese human service employees. Int. J. Stress Manag. 14 (1), 72–87. Chou, H.Y., Hecker, R., Martin, A., 2012. Predicting nurses’ well-being from job demands and resources: a cross-sectional study of emotional labour. J. Nurs. Manag. 20 (4), 502–511. Clegg, C.W., 1983. Psychology of employee lateness, absence, and turnover: a methodological critique and an empirical study. J. Appl. Psychol. 68 (1), 88–101. de Jonge, J., Le Blanc, P.M., Peeters, M.C.W., Noordam, H., 2008. Emotional job demands and the role of matching job resources: a cross-sectional survey study among health care workers. Int. J. Nurs. Stud. 45 (10), 1460–1469. Demerouti, E., Verbeke, W.J.M.I., Bakker, A.B., 2005. Exploring the relationship between a multidimensional and multifaceted burnout concept and self-rated performance. J. Manag. 31 (2), 186–209. Diefendorff, J.M., Erickson, R.J., Grandey, A.A., Dahling, J.J., 2011. Emotional display rules as work unit norms: a multilevel analysis of emotional labor among nurses. J. Occup. Health Psychol. 16 (2), 170–186.

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx Diefendorff, J.M., Hall, R.J., Lord, R.G., Strean, M.L., 2000. Action-state orientation: construct validity of a revised measure and its relationship to work-related variables. J. Appl. Psychol. 85 (2), 250–263. Diestel, S., Cosmar, M., Schmidt, K.-H., 2013. Burnout and impaired cognitive functioning: the role of executive control in the performance of cognitive tasks. Work Stress 27 (2), 164–180. Diestel, S., Schmidt, K.-H., 2010. Direct and interaction effects among the dimensions of the Maslach Burnout Inventory: results from two German longitudinal samples. Int. J. Stress Manag. 17 (2), 159– 180. Geurts, S.A., Buunk, B.P., Schaufeli, W.B., 1994. Health complaints, social comparisons, and absenteeism. Work Stress 8 (3), 220–234. Giardini, A., Frese, M., 2006. Reducing the negative effects of emotion work in service occupations: emotional competence as a psychological resource. J. Occup. Health Psychol. 11 (1), 63–75. Gonza´lez-Morales, M.G., Rodrı´guez, I., Peiro´, J.M., 2010. A longitudinal study of coping and gender in a female-dominated occupation: predicting teachers’ burnout. J. Occup. Health Psychol. 15 (1), 29–44. Goodwin, R.E., Groth, M., Frenkel, S.J., 2011. Relationships between emotional labor, job performance, and turnover. J. Vocat. Behav. 79 (2), 538–548. Grandey, A.A., 2000. Emotion regulation in the workplace: a new way to conceptualize emotional labour. J. Occup. Health Psychol. 5 (1), 95–110. Grandey, A.A., 2003. When the show must go on: surface acting and deep acting as determinants of emotional exhaustion and peer-rated service delivery. Acad. Manage. J. 46 (1), 86–96. Gross, J.J., 1998. The emerging field of emotion regulation: an integrative review. Rev. Gen. Psychol. 2 (3), 271–299. Gross, J.J., Levenson, R.W., 1997. Hiding feelings: the acute effects of inhibiting negative and positive emotion. J. Abnorm. Psychol. 106 (1), 95–103. Hagger, M.S., Wood, C., Stiff, C., Chatzisarantis, N.L.D., 2010. Ego depletion and the strength model of self-control. Psychol. Bull. 136 (4), 495– 525. Hennig-Thurau, T., Groth, M., Paul, M., Gremler, D.D., 2006. Are all smiles created equal? How emotional contagion and emotional labour affect service relationships. J. Market. 70 (3), 58–73. Heuven, E., Bakker, A.B., Schaufeli, W.B., Huisman, N., 2006. The role of self-efficacy in performing emotion work. J. Vocat. Behav. 69 (2), 222– 235. Hobfoll, S.E., 1989. Conservation of resources: a new attempt at conceptualizing stress. Am. Psychol. 44 (3), 513–524. Hobfoll, S.E., 2002. Social and psychological resources and adaptation. Rev. Gen. Psychol. 6 (4), 307–324. Hochschild, A., 1983. The Managed Heart. University of California Press, Berkeley, CA. ˜ igo, D., Totterdell, P., 2008. Emotional labour, Holman, D., Martı´nez-In well-being and performance. In: Cooper, C.L., Cartwright, S. (Eds.), The Oxford Handbook of Organizational Well-being. Oxford University Press, Oxford, England, pp. 331–355. Hu¨lsheger, U.R., Lang, J.W.B., Maier, G.W., 2010. Emotional labour, strain, and performance: testing reciprocal relationship in a longitudinal panel study. J. Occup. Health Psychol. 15 (4), 505–521. Hu¨lsheger, U.R., Schewe, A.F., 2011. On the costs and benefits of emotional labour: a meta-analysis of three decades of research. J. Occup. Health Psychol. 16 (3), 361–389. Hunter, B., Smith, P., 2007. Emotional labour: just another buzz word? Int. J. Nurs. Stud. 44 (6), 589–861. Hyland, M.E., 1987. Control theory interpretation of psychological mechanisms of depression: comparison and integration of several theories. Psychol. Bull. 102 (1), 109–121. Inzlicht, M., Schmeichel, B.J., 2012. What is ego depletion? Toward a mechanistic revision of the resource model of self-control. Perspect. Psychol. Sci. 7 (5), 450–463. Judge, T.A., Woolf, E.F., Hurst, C., 2009. Is emotional labour more difficult for some than for others? A multilevel, experience-sampling study. Pers. Psychol. 62 (1), 57–88. Johns, G., 2011. Attendance dynamics at work: the antecedents and correlates of presenteeism, absenteeism, and productivity loss. J. Occup. Health Psychol. 16 (4), 483–500. Krohne, H.W., Egloff, B., Kohlmann, C.W., Tausch, A., 1996. Untersuchungen mit einer deutschen Version der Positive and Negative Affect Schedule (PANAS).(Studies with a German version of the positive and negative affectivity scales). Diagnostica 42 (2), 139–156. Larson, G.E., Alderton, D.L., Neideffer, M., Underhill, E., 1997. Further evidence on dimensionality and correlates of the Cognitive Failures Questionnaire. Br. J. Psychol. 88 (1), 29–38.

11

Little, T.D., Cunningham, W.A., Shahar, G., Widaman, K.F., 2002. To parcel or not to parcel: exploring the question, weighting the merits. Struct. Eq. Model. 9, 151–173. Mann, S., 2005. A health-care model of emotional labour. J. Health Org. Manag. 19 (4/5), 304–317. ˜ igo, D., Totterdell, P., Alcover, C.M., Holman, D., 2007. EmoMartı´nez-In tional labour and emotional exhaustion: interpersonal and intrapersonal mechanisms. Work Stress 21 (1), 30–47. Maslach, C., Jackson, S.E., 1984. Burnout in organizational settings. In: Oskamp, S. (Ed.), Applied Social Psychology Annual. Sage, Beverly Hills, CA, pp. 133–153. Maslach, C., Jackson, S.E., 1986. The Maslach Burnout Inventory, 2nd ed. Consulting Psychologists Press, Palo Alto, CA. McVay, J.C., Kane, M.J., 2009. Conducting the train of thought: working memory capacity, goal neglect, and mind wandering in an executivecontrol task. J. Exp. Psychol. Learn. Memory Cogn. 35 (1), 196–204. Mesmer-Magnus, J.R., DeChurch, L.A., Wax, A., 2012. Moving emotional labour beyond surface and deep acting: a discordance–congruence perspective. Org. Psychol. Rev. 2 (1), 6–53. Muraven, M., Baumeister, R.F., 2000. Self-regulation and depletion of limited resources: does self-control resemble a muscle? Psychol. Bull. 126 (2), 247–259. Nguyen, H., Groth, M., Johnson, A., 2014. When the going gets tough, the tough keep working: impact of emotional labor on absenteeism. J. Manag. (in press). Oaten, M., Cheng, K., 2005. Academic examination stress impairs selfcontrol. J. Soc. Clin. Psychol. 24 (2), 254–279. Oaten, M., Cheng, K., 2006. Longitudinal gains in self-regulation from regular physical exercise. Br. J. Health Psychol. 11 (4), 717–733. Philipp, A., Schu¨pbach, H., 2010. Longitudinal effects of emotional labour on emotional exhaustion and dedication of teachers. J. Occup. Health Psychol. 15 (4), 494–504. Pisaniello, S.L., Winefield, H.R., Delfabbro, P.H., 2012. The influence of emotional labour and emotional work on the occupational health and wellbeing of South Australian hospital nurses. J. Vocat. Behav. 80 (3), 579–591. Podsakoff, P.M., MacKenzie, S.B., Podsakoff, N.P., Lee, J.Y., 2003. Common method biases in behavioral research: a critical review of the literature and recommended remedies. J. Appl. Psychol. 88 (5), 879–903. Reis, H.T., Gable, S.L., 2000. Event-sampling and other methods for studying everyday experience. In: Reis, H.T., Judd, C.M. (Eds.), Handbook of Research Methods in Social and Personality Psychology. Cambridge University Press, Cambridge, pp. 190–222. Richards, J.M., Gross, J.J., 1999. Composure at any cost? The cognitive consequences of emotion suppression. Pers. Soc. Psychol. Bull. 25 (8), 1033–1044. Richards, J.M., Gross, J.J., 2000. Emotion regulation and memory: the cognitive costs of keeping one’s cool. J. Pers. Soc. Psychol. 79 (3), 410–424. Robinson, J.L., Demaree, H.A., 2007. Physiological and cognitive effects of expressive dissonance. Brain Cogn. 63 (1), 70–78. Salanova, M., Llorens, S., Garcı´a-Renedo, M., Burriel, R., Breso´, E., Schaufeli, W.B., 2005. Toward a four-dimensional model of burnout: a multigroup factor-analytic study including depersonalization and cynicism. Educ. Psychol. Measur. 65 (5), 807–819. Schermelleh-Engel, K., Moosbrugger, H., Mu¨ller, H., 2003. Evaluating the fit of structural equation models: tests of significance and descriptive goodness-of-fit measures. Methods Psychol. Res. Online 8 (2), 23–74. Schmeichel, B.J., 2007. Attention control, memory updating, and emotion regulation temporarily. J. Exp. Psychol. Gen. 136 (2), 241–255. Schmeichel, B.J., Volokhov, R., Demaree, H.A., 2008. Working memory capacity and the self-regulation of emotional expression and experience. J. Pers. Soc. Psychol. 95 (6), 1526–1540. Schmidt, K.-H., Neubach, B., Heuer, H., 2007. Self-control demands, cognitive control deficits, and burnout. Work Stress 21 (2), 142–154. Smith, P., 1991. The nursing process: raising the profile of emotional care in nurse training. J. Adv. Nurs. 16 (1), 74–81. Spencer, S.J., Zanna, M.P., Fong, G.T., 2005. Establishing a causal chain: why experiments are often more effective than mediational analyses in examining psychological processes. J. Pers. Soc. Psychol. 89 (6), 845–851. Staw, B.M., Oldham, G.R., 1978. Reconsidering our dependent variables: a critique and empirical study. Acad. Manage. J. 21 (4), 539–559. Taris, T.W., Le Blanc, P.M., Schaufeli, W.B., Schreuers, P.J.G., 2005. Are there causal relationships between the dimensions of the Maslach Burnout Inventory? A review and two longitudinal tests. Work Stress 19 (3), 238–255.

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003

G Model

NS-2370; No. of Pages 12 12

K.-H. Schmidt, S. Diestel / International Journal of Nursing Studies xxx (2014) xxx–xxx

Totterdell, P., Holman, D., 2003. Emotion regulation in customer service roles: testing a model of emotional labour. J. Occup. Health Psychol. 8 (1), 55–73. van Dam, A., Keijsers Ger, P.J., Eling, P.A.T.M., Becker, E.S., 2011. Testing whether reduced cognitive performance in burnout can be reversed by a motivational intervention. Work Stress 25 (3), 257–271.

Wallace, J.C., 2004. Confirmatory factor analysis of the cognitive failures questionnaire: evidence for dimensionality and construct validity. Pers. Individual Differ. 37 (2), 307–324. Watson, D.A., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS scales. J. Pers. Soc. Psychol. 54 (6), 1063–1070.

Please cite this article in press as: Schmidt, K.-H., Diestel, S., Are emotional labour strategies by nurses associated with psychological costs? A cross-sectional survey. Int. J. Nurs. Stud. (2014), http://dx.doi.org/10.1016/j.ijnurstu.2014.03.003