Desarrollo y validacion trasnacional de la Escala de Esfuerzo Emocional (EEF).
Customer service employees are required to show a range of different emotions in order to achieve successful interactions with customers (e.g., Kiffin-Peterson, Jordan, & Soutar, 2011). According to Hochschild (1983), showing the emotions required by the role is an effortful process, which she labelled: "Emotional Labour" (EL). Similarly, Morris and Feldman defined EL as the "effort, planning and control required to display organizationally desired emotions during interpersonal transactions" (1996, p. 987). In spite of the emphasis on effort, EL has been extensively measured in terms of the strategies individuals use to manage their emotions: Deep Acting (DA) and Surface Acting (SA) (e.g., Brotheridge & Lee, 2003). DA describes the process whereby employees change their own feelings to match the emotions they have to display (e.g., Grandey, 2000). In contrast, SA refers to faking the emotional displays.
Quantitative studies measuring the impact of DA and SA have confirmed a strong link between EL and burnout. Overall, SA is regarded as the most harmful strategy based on the consistent associations with emotional exhaustion and cynicism (e.g., Grandey, Fisk, & Steiner, 2005; Hulshegera, Langa, & Maie, 2010; Martinez-Inigo, Totterdell, Alcover, & Holman, 2007). Nevertheless, the predictive ability of SA has been questioned, as some studies show that SA did not add any explanatory value beyond negative affectivity (e.g., Brotheridge & Grandey, 2002; Zammuner & Galli, 2005). According to Goodwin (2011), these inconsistencies can be due to the lack of attention to the "effort" component of EL, which should be assessed separately from the strategies.
In line with the main definitions of EL a sensible operationalization of the construct should incorporate the key element: "effort". In this study, we aim to develop the Emotional Effort Scale (EEF) (Study 1). Secondly, we aim to adapt the EEF to Spanish and demonstrate Measurement Invariance of the scale in the two national groups of study (UK and Spain) (Study 2). Finally, we aim to demonstrate significant relationships between emotional effort and theoretically related outcomes (i.e., surface acting, deep acting and emotional exhaustion) (Study 3).
Conceptual definition of the Emotional Effort construct
Based on the consistent relationships between EL and strain, we build on two stress-related frameworks to conceptualize the emotional effort construct: transactional theory and ego depletion theory. Transactional theory defines stress as "a particular relationship between the person and the environment that is appraised by the person as taxing [...] and endangering well-being" (Lazarus & Folkman, 1984, p. 19). In line with the emphasis on individual appraisals, emotional effort could be conceptualized as the perception of resources invested when meeting emotional requirements of the job (i.e., display rules). Further, transactional stress theory states that the cognitive appraisal mediates the relationship between external demands and stress reactions. Thus, it could be that the perceived effort associated with meeting emotional requirements of the job, explains the resource depletion commonly attributed to emotional labour. In short, we define emotional effort as the explicit perception of resources invested when managing emotions to meet display rules.
On the other hand, the effort associated with emotional requirements of the job, could operate at a less conscious level. In line with this, ego depletion theory suggests that we know that an activity involves high self-controlled effort, when the energy remaining for subsequent activities is diminished (Baumeister, Bratslavsky, Muraven, & Tice, 1998). Thus, it was found that the experimental condition of suppressing emotions whilst watching a film could involve effort, based on the higher number of mistakes made in subsequent cognitive tasks (Baumeister et al., 1998). In view of this, implicit emotional effort is conceptualized as a less conscious resources investment process, that can be inferred from the degree of interference with the execution of other tasks (Martinez-Inigo et al., 2007). In short, we conceptualize Emotional Effort as a two-dimensional construct with an explicit dimension (i.e., perceptions of resource investment) and an implicit dimension (i.e., effort inferred from the degree of interference with other tasks).
Cross-national adaptation and evidence of external validity
Most studies on the EL literature have focused on Anglo-Saxon samples. However, evidence suggests that the "service with a smile" model has been exported to countries where free regulation of emotion is encouraged. Thus, a cross-cultural study found that customer service staff from a country in the Latin-European cluster (i.e., France) actively engaged in SA. However, their levels of SA were significantly lower than those of American employees (Grandey et al., 2005). Prior to this, the authors had demonstrated measurement invariance of the instruments in order to ensure that differences were not due to instrument bias (Byrne, 2008). In this study, we aim to cross-nationally validate the EEF in two different representatives of the aforementioned clusters: Spain and UK. We expect that this scale is metrically and conceptually equivalent in both countries, allowing future comparisons on variable levels.
The stronger link between SA and work-related stress is commonly interpreted as an index of the higher effort involved in this strategy (Austin, Dore, & Donovan, 2008; Grandey, 2000). Thus, suppressing emotions changes the outward display, but the internal activation remains, depleting resources over time (Grandey, 2000). On the other hand, some argue that the active modification of emotions (i.e., DA) requires higher effort (Zapf, 2002). Similarly, Liu, Prati, Perrewe, and Ferris (2008) argue that, unlike with DA, individuals perform SA under resource-loss circumstances, in an effort to conserve their energy. Although both strategies are expected to be associated with effort, there is not enough evidence to hypothesise either strategy to involve higher effort than the other.
In line with the stress models where emotional effort is conceptualised, EEF is expected to be strongly associated with emotional exhaustion. Similarly, to the extent to which meeting emotional requirements of the role interferes with other tasks, emotional exhaustion is also likely to arise (Baumeister et al., 1998). Further, evidence from qualitative studies suggests that conscious "effort" could be a potential intervening variable in the relationship between the EL strategies and exhaustion (Wong & Wang, 2009). Hence, we expect that the relationship between SA and emotional exhaustion is explained by the emotional effort associated with meeting display rules. This will allow us to include EEF in a comprehensive EL model with greater criterion potential.
Participants and procedure
A total of 379 customer service employees from the south of England participated in this study. Recruitment consultants and letting agencies comprised 45% of the sample, 26% retail, 16% bars and restaurants, 7% travel agents, and 6% bank clerks and insurers. Ages ranged from 17 years old to 61, with an average of 33 and 9.3 SD; 60% were female and 40% were male. Once ethical approval was obtained, different workplaces, where employees have to deal with customers on a regular basis, were approached. We asked employees to voluntarily take part in a doctoral study by completing a 12-minute survey on their lunch break. Anonymity and confidentiality were guaranteed. The first author collected the questionnaires after lunch on the same day. All questionnaires were collected and participation was estimated at 89%.
Item construction and qualitative assessment: evidence of content validity. Carretero-Dios and Perez's (2007) recommendations on instrument development were followed. Two frameworks guided item development: the ego depletion theory (Baumeister et al., 1998) and the classic stress theory (Lazarus & Folkman, 1984). First, the authors wrote instructions of the scale inspired by the various EL definitions (e.g., Morris & Feldman, 1996). Instructions were: "MEETING EMOTIONAL DISPLAY RULES refers to the process whereby you show the emotions required by your job (smiling when first meeting a customer, hiding anger when dealing with an unpleasant customer etc.). In the last month, HOW OFTEN have you felt that ...". Respondents were asked to rate their answers on a 5-point Likert Scale ranging from 1 (never) to 5 (very often). In line with Ego Depletion theory, we worded items related to Implicit Effort so that they reflect the effort by interference with other tasks (see Table 1). We also incorporated the two existing items of psychological effort developed by Martinez-Inigo et al. (2007). Then, following Lazarus and Folkman's (1984) transactional theory, we developed items referring to individuals' explicit perception of emotional effort: for example, "How often have you felt that this activity involves a great amount of effort?"
The item specification table with the semantic definition of the construct and its components was given to a panel of four experts. First, we asked them to rate how well the items reflected the construct, and whether the items were understandable. They gave their responses on a scale ranging from 0 (disagree) to 4 (agree). Those items scoring lower than 3 were re-worded. We also conducted a small pilot study with 20 customer-service workers and simplified the instructions based on their suggestions. Finally, we obtained a 13-item scale with 6 and 7 items that a priori seemed to tap into the theoretical dimensions Explicit and Implicit Emotional Effort respectively.
We randomly split the British sample (N = 379), as this is an accepted procedure when sample sizes are of a relevant magnitude and communalities are high (.60 or greater) (MacCallum, Widaman, Preacher, & Hong, 2001). We then conducted descriptive analysis for the items, exploratory factor analysis (EFA) and item reduction with one of the subsamples (N = 197) using SPSS 20. Then, we conducted confirmatory factor analysis (CFA) with the other sub-sample (N = 182) using AMOS 20. Missing cases were 3% (i.e., below the 5% threshold where listwise deletion is not recommended). The "exclude cases pairwise" option was selected in SPSS and deletion of missing cases in AMOS.
Internal structure of the scale
Firstly, we analysed the psychometric properties of the scale. Following Nunnally and Bernstein's (1994) recommendations, four items with a correlation item-total below .300 were deleted (Items 4, 5, 11, 13). EFA was conducted with the remaining nine items. Prior to this, we found that Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was .77 and Bartlett's test of sphericity achieved statistical significance ([chi square] = 350.3, p<.000), supporting the adequacy of factor analysis. Then, we selected principal axis factoring and oblimin rotation. Kaiser's criterion supported the two-dimension solution to retain factors (those with eigenvalues >1). We also used the Monte Carlo principal component analysis for parallel analysis. This also indicated retaining two factors. Finally, we checked that each item loaded above .40 on its factor and below .30 on the other one (Nunnally & Bernstein, 1994), and 2 items that did not meet this criterion were removed. The analyses were repeated for the 7 remaining items and two factors.
Psychometric properties of the items can be seen in Table 1. The four items loading onto the first factor corresponded to what we expected for the implicit effort dimension (Items 4, 5, 6, and 7). Consequently, we labelled this dimension "Implicit Emotional Effort". The three items that loaded onto the second factor grouped items corresponded to appraisal of effort; as a result, we labelled that dimension "Explicit Emotional Effort" (Items 1, 2 and 3). Cronbach's alphas were .73 for explicit effort and .76 for implicit effort.
Confirmatory factor analysis
In order to test the fit of the two factor solution we ran CFA with the second British subsample (N = 182), using AMOS 20 and Maximum Likelihood. The Incremental Fit Index (IFI), the Comparative Fit Index (CFI), the Tucker-Lewis Index (TLI) all must be over .90. Additionally, the Standardized Root Mean Square Residual (SRMR) and the Root Mean Squared Error of Approximation (RMSEA) must be below .08 and [chi square] /df value lower than 3.0. Our results met all the requirements to conclude that the two dimension theoretical model exhibits a good fit (IFI = .963, CFI = .962, TLI = .933, SRMR = .054, RMSEA = .080 and [chi square] /df = 2.14). Cronbach's alphas were .69 for explicit effort and .79 for implicit effort.
We compared the fit of the two factor model to that of a single factor model for the same data. Differences between models were significant ([DELTA] [chi square] = 41.91, p<.001). Thus, we rejected the most parsimonious model and supported the two factor solution. Factor loadings for items corresponding to each dimension of effort were high and loaded in their corresponding factors.
Participants and procedure
A total of 304 customer service employees from Asturias (Spain) participated in this study: 68.8% retail, 11.7% travel agencies and hotels, 7.5% bars and restaurants, 6.8% lettings and insurance agencies, and 5.2 % hairdressing. Their ages ranged from 18 to 65 years with an average age of 34.4; of the sample, 79% were female and 21% were male. Same procedure as in Study 1 was followed.
Emotional Effort Scale Adaptation. We followed the main steps for the adaptation of assessment instruments recommended in the literature (e.g., Muniz & Hambleton, 1996). Once we confirmed with experts that this construct was potentially present in the Spanish population, we started with the back translation procedure. First, the native Spanish bilingual author translated the items from English to Spanish. The rest of the author-team confirmed that the meaning held. Then, a bilingual native English speaker translated the items back to English. Finally, we reviewed the items to determine whether the cross-translations held their original meaning. Minor changes were made to the adapted version. Sixty-five individuals with customer service experience participated in the pilot. They were asked to comment on the instructions and understanding of the items. Then, the instrument was administered to the final sample. Measurement invariance (MI) tests were conducted with multigroup confirmatory factor analysis (MCFA) to ensure that the emotional effort was invariant across the two national groups.
First, multivariate analysis of covariance was performed with country as independent variable, the EEF items as dependent variables, and the demographic variables that were significantly different across samples as covariates (e.g., gender, type of job). The model was significant only for country of origin Wilks' [lambda] = 7.07, p<.001. Then, measurement invariance tests were performed. These involve various hierarchical model testing steps with the two samples simultaneously (Byrne, 2008). Prior to that, the model must have a good fit for each sample independently. Thus, we fit the same model of Study 1 to the Spanish sample. Again, we found good fit (IFI = .983, CFI = .983, TLI = .972, SRMR = .036, RMSEA = .050 and [chi square] /df = 1.73) and support for a better fit of the two factor model ([DELTA] [chi square] = 127.33, p<.001). Cronbach's alpha coefficients were .77 for explicit effort and .74 for implicit effort. Factor loadings can be appreciated in Figure 1b.
Next, the first level of MI or the "configural invariance model" was confirmed by fitting the model across the two groups simultaneously and finding a good fit (Model 1: Table 2). The next model testing stage represents a stronger case for invariance: "metric invariance model". Here we compared the fit of a model with no constraints across groups to one where constraint of equal factor loadings across samples was imposed. As the increment in chi-square was not significant, metric invariance was supported. These steps are sufficient to justify that the item content and metric properties are equivalent across countries. In order to test whether the meaning and structure of the construct were also invariant, we also tested the "strong invariance" step. Thus, we added the constraint of equal intercepts (Models 3 and 4, Table 2) and equal latent variables (Model 5). Full scalar invariance was not supported due to two intercepts (p<.004). According to Byrne, Shavelson, and Muthen (1998), as long as there are two items or more that remain invariant, latent means can be meaningfully compared. Hence, the two intercepts were freed up and the fit of the model was estimated resulting in non-significant chi square differences, which supports (partial) scalar invariance. Finally, the latent means were constrained (Model 5) and the model did not have a significantly worse fit. Hence, equality of structural means is also supported.
Participants and procedure
A total of 185 customer service employees from Madrid (Spain) participated in this study. Participants were employed as follows: 54.1% ride operators, 25.4% bars and restaurants, 15.7% retail, 4.8% customer service staff. Their ages ranged from 17 to 61 with an average age of 29.5 and 9.7 SD. Of these, 47% were female and 53% were male. The same procedure as in Study 1 and 2 was followed.
Emotional Labour. Brotheridge and Lee's (2003) scale was used. Respondents are asked to rate statements on a scale ranging from 1 (never) to 5 (always). A sample item for SA: "Fingir emociones que no siento realmente" [Faking emotions I don't really feel]. A sample item for DA: "Tratar de sentir realmente las emociones que debo mostrar" [Try to really feel the emotions I should show]. Cronbach's alphas were .86 for SA and .80 for DA.
Emotional exhaustion. We measured this with the Spanish adaptation of the Maslach Burnout Inventory-GS (Moreno-Jimenez, Rodriguez-Carvajal, & Escobar-Redonda, 2001). We asked respondents to rate the extent to
which they experience each of the statements on a scale ranging from 1 (never) to 5 (always). The reliability for this scale was .92.
Emotional effort. Cronbach's alphas were .75 for explicit effort and .79 for implicit effort.
Negative affect. Sandin, Chorot, Lostao, Joiner, Santed, & Valiente's (1999) scale was used. Alpha coefficient was .84.
In order to demonstrate that emotional effort and the EL strategies did not converge to the extent of being redundant constructs, we ran CFA with the EL strategies and emotional effort (Ferris, Brown, Berry, & Lian's, 2008). The model achieved a good fit (CFI = .92, IFI = .92, SRMR = .06) and all items loaded significantly into the expected latent variables. Further, for each of the EL strategies and the emotional effort dimensions, a two-factor model yielded a significantly better fit than the one factor model (e.g., explicit effort and SA: 1, [DELTA][chi square] = 54.76, p<.001). These results provide support for the convergent and divergent validity of EEF.
Then, we explored the nomological net of emotional effort with correlation analysis. As expected, emotional effort was significantly associated with SA. In contrast, DA was only significantly associated with the explicit effort dimension. Finally, we found that emotional exhaustion was positively associated with explicit and implicit effort.
Next, using Baron and Kenny's (1986) mediation test, we found that explicit effort fully mediated the association between SA and exhaustion. Thus, when explicit effort was entered in the last step ([beta] = -.427, p<.001), surface acting lost its significance ([beta] = -.121, p. 096). The Sobel test supported this mediation (z = 3.11, p<.001)
The aim of this study was to develop and cross-nationally validate the Emotional Effort Scale. Evidence gathered in three studies supports that the EEF scale is a two-factor construct with good reliability levels across all samples. We also confirm that the instrument meets the requirements to be used as a reliable tool for cross-national hypothesis testing in Spain and UK (Byrne et al., 1989). Thus, future studies may include this instrument for meaningful comparison across the aforementioned countries. Finally, we also found evidence of convergent, divergent and nomological validity. Thus, emotional effort is related, but a different construct from DA and SA. Further, explicit emotional effort seems to be the mechanism that explained the impact of the EL strategy SA on emotional exhaustion.
The development of this emotional effort operationalization fills a gap in the literature regarding the definition of EL as "effort", and the lack of an instrument to assess such a construct independently from the EL strategies (Goodwin, 2011). In line with the experimental evidence, which the studies testing ego-depletion theory provided (e.g., Martinez-Inigo et al., 2007), implicit effort was associated with emotional exhaustion. However, explicit effort was the strongest and only significant predictor when we analysed it simultaneously with the implicit effort dimension. This provides strong support for the relevance of the transactional stress theory to understand the impact of EL on employees' exhaustion.
This manuscript also clarifies the debate about whether the EL strategy seems to involve higher effort (e.g., Liu et al., 2008). Thus, we found a stronger association between emotional effort and SA than with DA; this supports the emotion regulation approach (Grandey, 2000). Thus, suppressing felt emotions (SA) involves a high level of effort, because one has to regulate them after the emotional responses have been triggered. In contrast, DA could operate at a lower level of conscious effort as the regulation comes before the emotion is fully underway. Nevertheless, future research with diary study methodology is encouraged to clarify this further. This will also allow differentiating the perceived effort of the strategies when one implements them, and, the immediate impact on stress rather than the aggregate effect.
Regarding practical applications, we believe that there is room for the use of the EEF scale in recruitment and selection processes. The instrument could be used to assess the potential vulnerability of stress for employees in constant interaction with customers. However, we need more studies that examine the relationships between effort and other desired traits of prospective employees. Thus, even though it was related to negative outcomes in our study, it could be that individuals who score high in effort are also more likely to engage with their work, thereby becoming assets for their organization.
Finally, research is needed on the area of emotion regulation training for customer service employees. Such training should reduce the cognitive and emotional burden placed on employees who are required to perform EL on a regular basis, freeing these resources to focus on other aspects of their role. Further, research is encouraged to study the moderator role of potentially relevant psychological constructs (e.g., empathy) in the relationship between emotional effort and well-being. Thus, it could be that effort is less significantly related to negative outcomes when individuals have a tendency to be highly empathetic with others' emotions.
Among the limitations of this study, the risk of the common method bias must be cited. Nevertheless, we followed Conway and Lang's (2010) steps to minimize this bias (e.g., demonstrating the lack of overlap of the items with related constructs). Additionally, respondent's answers to these self-report measures could have been biased due to social desirability. Finally, the cross-sectional design does not allow demonstrating causal association between the variables. However, this is an appropriate design for the development and validation of a new instrument.
In conclusion, the development and validation study of the Emotional Effort Scale (EEF) provides strong evidence for the internal structure and validity of the relationship with theoretically related variables (SA, DA and emotional exhaustion). We expect that this construct will contribute to explain the intervening mechanisms in the association between EL and employees' wellbeing in future studies.
Austin, E.J., Dore, T.C.P., & O'Donovan, K.M. (2008). Associations of personality and emotional intelligence with display rule perceptions and emotional labour. Personality and Individual Differences, 44, 679688.
Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in social psychology research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.
Baumeister, R., Bratslavsky, E., Muraven, M., & Tice, D. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74, 1252-1265.
Brotheridge, C.M., & Grandey, A.A. (2002). Emotional labor and burnout: Comparing two perspectives of "people work". Journal of Vocational Behavior, 60, 17-39.
Brotheridge, C.M., & Lee, R.T. (2003). Development and validation of the Emotional Labour Scale. Journal of Occupational and Organizational Psychology, 76, 365-379.
Byrne, B.M. (2008). Testing for multigroup equivalence of a measuring instrument: A walk through the process. Psicothema, 20, 872-882.
Byrne, M.B., Shavelson, R.J., & Muthen, B.O. (1989). Testing for the equivalence of factor covariance and mean structures: The issue of partial measurement invariance. Psychological Bulletin, 105, 456-466.
Carretero-Dios, H., & Perez, C. (2007). Standards for the development and review of instrumental studies: Considerations about test selection in psychological research. International Journal of Clinical and Health Psychology, 7, 863-882.
Conway, J.M., & Lance, C.E. (2010). What reviewers should expect from authors regarding common method bias in organizational research. Journal of Business and Psychology, 25, 325-334.
Ferris, D.L., Brown, D.J., Berry, J.W., & Lian, H. (2008).The development and validation of the workplace ostracism scale. Journal of Applied Psychology, 93, 1348-1366.
Goodwin, R.E. (2011). Understanding the relationship between emotional labor and effort. In C.E.J. Hartel, N.M. Ashkanasy, & W.J. Zerbe (Eds.), Research on Emotions in Organizations: What have we learned? Ten years on (pp. 45-71). Bingley: Emerald Group Publishing.
Grandey, A.A. (2000). Emotion regulation in the workplace: A new way to conceptualize emotional labour. Journal of Occupational Health Psychology, 5, 95-110.
Grandey, A.A., Fisk, G.M., & Steiner, D.D. (2005). Must "service with a smile" be stressful? The moderating role of personal control for American and French employees. Journal of Applied Psychology, 90,893-904.
Hochschild, A.R. (1983). The managed heart. London: University of California Press.
Hulshegera, U.R., Langa, J.W.B., & Maie, G.W. (2010). Emotional labor, strain, and performance: Testing reciprocal relationships in longitudinal panel study. Journal of Occupational Health Psychology, 15, 505521.
Kiffin-Petersen, S.A., Jordan, C.L., & Soutar, G.N. (2011). The big five, emotional exhaustion and citizenship behaviors in service settings: The mediating role of emotional labor. Personality and Individual Differences, 50, 43-48.
Lazarus, R.S., & Folkman, S. (1984). Stress, appraisal and coping. New York: Springer.
Liu, Y., Prati, L.M., Perrewe, P.L., & Ferris, G.R. (2008). The relationship between emotional resources and emotional labor: An exploratory study. Journal of Applied Social Psychology, 38, 2410-2439.
MacCallum, R.C., Widaman, K.F., Preacher, K.J., & Hong, S. (2001). Sample size in factor analysis: The role of model error. Multivariate Behavioral Research, 36, 611-637.
Martinez-Inigo, D., Totterdell, P., Alcover, C.M., & Holman, D. (2007). Emotional labour and emotional exhaustion: Interpersonal and intrapersonal mechanisms. Work and Stress, 21, 30-47.
Moreno-Jimenez, B., Rodriguez-Carvajal, R., & Escobar-Redonda, E. (2001). La evaluacion del burnout profesional. Factorializacion del MBI-GS. Un analisis preliminar. Ansiedad y Estres, 7, 69-78.
Morris, J., & Feldman, D.C. (1996). The dimensions, antecedents, and consequences of emotional labor. The Academy of Management Review, 21, 986-1010.
Muniz, J., & Hambleton, R.K. (1996). Directrices para la traduccion y adaptacion de los tests. Papeles del Psicologo, 66, 63-70.
Nunnally, J.C., & Bernstein, I.H. (1994). Psychometric theory (3rd edition). New York: McGraw Hill.
Sandin, B., Chorot, P., Lostao, L., Joiner, T.E., Santed, M.A., & Valiente, R.M. (1999). Escalas PANAS de afecto positivo y negativo: validacion factorial y convergencia transcultural. Psicothema, 11(1), 37-51.
Wong, J.Y., & Wang, C.H. (2009). Emotional labor of the tour leaders: An exploratory study. Tourism Management, 30, 249-259.
Zammuner, V.L., & Galli, C. (2005). Wellbeing: Causes and consequences of emotion regulation in work settings. International. Review of Psychiatry, 17, 355-364.
Zapf, D. (2002). Emotion work and psychological well-being: A review of the literature and some conceptual considerations. Human Resource Management Review, 12, 237-268.
Cristina Quinones-Garcia (1), Raquel Rodriguez-Carvajal (2), Nicholas Clarke (3) and Bernardo Moreno-Jimenez (2)
(1) University of Northampton, (2) Universidad Autonoma de Madrid and (3) University of Southampton
Received: October 16, 2012 * Accepted: April 9, 2013
Corresponding author: Cristina
University of Northampton
nn27al Northampton (Reino Unido)
Table 1 Factors of the EEF scale, item description, item descriptives, Cronbach alpha and item loadings estimated with principal axis factoring and oblimin rotation (Study 1: UK subsample N = 197) Factors Item description M SD DI [alpha] (b) Factor 1: Explicit Emotional Effort 1 this activity 2.88 1.20 .56 .65 involves a great amount of effort? 2 the effort you spend 2.67 1.10 .65 .54 in meeting emotional display rules is greater than the actual task you have to carry out? 3 this activity is the 2.65 1.25 .47 .74 main reason why you feel emotionally drained after work? Factor 2: Implicit Emotional Effort 4 you could deal with 2.74 1.13 .51 .74 a complaint more efficiently if you did not have to focus on meeting emotional display rules? (e.g. remaining calm when dealing with unpleasant customers) 5 you would be doing 2.40 1.13 .66 .65 a better job if you didn't have to meet certain emotional display rules (e.g expressing feelings of sympathy when you don't feel like that)? 6 meeting emotional display rules 2.14 .93 .62 .68 impairs your performance on other tasks? 7 you make more mistakes in other areas due 2.04 .91 .49 .75 to this activity? Factor Eigenvalue Variance [alpha] loading explained Factor 1: Explicit Emotional Effort .73 3.35 40.93% 1 .69 2 .86 3 .50 Factor 2: Implicit Emotional Effort .76 1.04 49.42% 4 .57 5 .88 6 .68 7 .49 Note: Items have been re-numbered for clarity purposes. [alpha] (b) Cronbach alpha if the item was deleted. Loadings <.3 are left blank Table 2 Fit indices of four nested models from multigroup confirmatory factor analysis-measurement invariance of the emotional effort scale ([N.sub.UK]= 182; [N.sub.Spain]= 304) M1: M2: Full M3: Full Model Configural metric scalar specification invariance invariance invariance [chi square] 43.15 55.54 81.35 (df, p) (24, .010) (31, .004) (38, .000) [chi square]/df 1.80 1.79 2.14 IFI .980 .974 .953 CFI .979 .973 .953 SRMR .040 .040 .049 RMSEA .041 .041 .041 [DELTA] -- 12.38 25.81 [chi square] [DELTA] df -- 7 7 Statistical -- ns p<.001 significance [DELTA] CFI .004 .000 M4: Partial Model scalar M5: Latent specification invariance * means [chi square] 58.12 (34, .006) 61.75(36,.005) (df, p) [chi square]/df 1.71 1.72 IFI .972 .972 CFI .972 .972 SRMR .039 .038 RMSEA .039 .039 [DELTA] [chi square] 18.60 3.63 [DELTA] df 12 2 Statistical significance ns ns [DELTA] CFI .026 .002 Note: Df_Degrees of freedom; [chi square] difference_Chi square differences; IFI_Incremental fit Index; GFI _Goodness-of-Fit Index; SRMR_ Standardized Root mean Square Residual; RMSEA_Root Mean Square Error of Approximation. Partial Scalar Inariance *: non-invariant intercepts items 6 and 10 are set free Table 3a Bivariate correlations of the variables of study in the validation of the emotional effort scale in study 3 (N = 185) M SD 1 2 1. Negative 1.37 0.46 affect 2. Explicit 2.47 0.99 .500 ** effort 3. Implicit 1.45 0.29 .399 ** .686 ** effort 4. Surface 2.73 0.76 .320 ** .458 ** acting 5. Deep 2.89 0.79 .143 .151 * acting 6. Emotional 2.10 1.02 .607 ** .648 ** exhaustion 3 4 5 1. Negative affect 2. Explicit effort 3. Implicit effort 4. Surface .383 ** acting 5. Deep .129 .465 ** acting 6. Emotional .456 ** .448 ** .129 exhaustion Note: ** p<.010, * p<.05 Table 3b Results of hierarchical multiple regression and mediation analysis in Study 3 (N = 185) Model summary Dependent vbs Standardized Emotional Explicit Emotional coefficients exhaustion effort exhaustion Step 1. Control variables (.398 ***) (.253 ***) (.398 ***) ([DELTA][R.sub.2]) Age .538 *** .080 .385 *** Gender -.016 -.020 -.050 Negative affect .156 ** .387 *** .164 ** Step 2. EL strategies (.071 ***) (.108 ***) (.071 ***) ([DELTA][R.sub.2]) Surface acting .279 *** .399 *** .121 Deep acting -.076 -.088 -.040 Step 3 Emotional effort (.104 ***) [DELTA][R.sub.2]) Explicit effort 427 *** Implicit effort -.040 Adjusted [R.sub.2] .339 .555 Note: *** p<.001, ** p<.010, * p<.05
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||texto en ingles|
|Author:||Quinones-Garcia, Cristina; Rodriguez-Carvajal, Raquel; Clarke, Nicholas; Moreno-Jimenez, Bernardo|
|Date:||Jul 1, 2013|
|Previous Article:||Preferencia por el olor del etanol tras la interaccion social con un congenere intoxicado en ratas adolescentes expuestas a la droga in utero.|
|Next Article:||Desarrollo y validacion de una escala para la identificacion de actitudes hacia la discapacidad en la Educacion Superior.|