Validity evidence of the Organizational Justice Scale in Spain.
In the Spanish context, two noteworthy organizational justice perception scales are available: the Colquitt's Organizational Justice Scale (Colquitt, 2001, COJS), and the Organizational Justice Scale developed by Moliner (2004, OJS). Firstly, the COJS was developed on the basis of a four-dimensional structure (distributive, procedural, informational, and interpersonal justice), with a second-order latent justice variable being demonstrated (Colquitt & Shaw, 2005). With reference to the psychometric properties of its Spanish version, the four-dimensional structure was recently discovered in a sample of 460 services employees (Diaz-Gracia, Barbaranelli, & Moreno-Jimenez, 2014). Secondly, the OJS was developed drawing from an emphasis on a faceted conception of organizational justice that reflects the concepts of distributive, procedural, and interactional justice (e.g., Moorman, 1991; Schminke, Ambrose, & Cropanzano, 2000). Despite its psychometric properties already analyzed in a hotel employees' sample (Moliner, 2004; Moliner, Martinez-Tur, Peiro, & Ramos, 2005), this study aims to provide further evidence about OJS characteristics.
Taking the above-mentioned information into account, the main goal of the present study is to examine the structure and reliability of the OJS using a wider sample. Given that the OJS has been developed on the basis of three-dimensions, the first hypotheses are:
H1. The OJS will comprise three organizational justice factors (distributive, procedural, and interactional).
H2. The three factors (distributive, procedural, and interactional) of the OJS will constitute a global organizational justice construct.
In addition, several studies have revealed a positive relationship between organizational justice perception and well-being, whereas a negative relationship between organizational justice perception and emotional exhaustion has been found (e.g., Heponiemi, Kuusio, Sinervo, & Elovainio, 2011; Lawson, Noblet, & Rodwell, 2009; Liljegren & Ekberg, 2009). In this sense, the OJS could also be used to study the relationship between organizational justice perception and these significant organizational variables. Thus, the second objective is to discover the criterion-oriented validity of the OJS, and consequently the hypotheses proposed are:
H3. The OJS will correlate in a significant and positive way with well-being.
H4. The OJS will correlate in a significant and negative way with emotional exhaustion.
Participants comprised 849 Spanish workers from different private (63.4%) and public (36.6%) organizations. Most of the public workers belonged to administrative and auxiliary services and security forces (33.5%), whereas most of the private workers belonged to the primary industry (42.8%). Table 1 shows a detailed breakdown in percentages of participants by sector.
Women represented 47.1% of the participants. The mean age was 38.82 years (SD = 12.432), and the mean experience in the job position, 7.33 years. In terms of their occupational characteristics, 44.2% of participants held a low-ranking position, whereas 55.8% were employed in technical or managerial positions.
Participants completed a questionnaire composed of several items regarding sociodemographic data, as well as the instruments listed below:
Organizational justice perception was measured with the OJS (Moliner, 2004), composed by 12 items (Table 2). Response options were delivered on a scale ranging from 1 (to a small extent) to 5 (to a large extent).
Emotional exhaustion was measured with five items extracted from the Spanish adapted version of the Maslach Burnout Inventory General Survey by Salanova, Schaufeli, Llorens, Grau, and Peiro (2000). Response options were delivered on a 7-point scale from 0 (never) to 6 (always). A sample item is: "I feel used up at the end of a workday".
Well-being was measured by means of 12 items of the Spanish version of the General Health Questionnaire by Goldberg and Williams (1996). Response options were delivered on a 4-point scale ranging from 0 (not at all) to 3 (more than usual). A sample item is: "Felt constantly under strain".
Researchers trained several survey takers who approached different organizations located in Asturias (Spain) and asked them for to distribute an anonymous questionnaire among their employees following a non-probabilistic snowball sampling. Participants who were prone to participate were given a paper and pencil questionnaire which they had to return after completion.
The dimensionality of the OJS was analyzed through exploratory (EFA), confirmatory (CFA) and second-order factor analyses splitting the sample up into two random subsamples ([n.sub.1] = 426, [n.sub.2] = 423). According to Friedman (1982), in order to obtain a statistical power of .90 (d = .30, r = .15) a sample of 459 is necessary. Thus, both subsamples were quite close to that value.
Mplus software (version 7, Muthen & Muthen, 2012) was used for the factor analyses. An oblique rotation was applied to interpret the obtained factors, and in order to analyzed the loadings of every item, .40 was taken as the recommended cut-off point (Lloret-Segura, Ferreres-Traver, Hernandez-Baeza, & Tomas-Marco, 2014; Matsunaga, 2010).
Regarding the estimation method, on the one hand, maximum likelihood (MLE) has been traditionally pointed out as adequate. On the other hand, polychoric correlations are highlighted as adequate when dealing with Likert polytomous responses which reflect the elections participants make from a continuous conception of the measured construct (e.g., Diaz-Vilela, Diaz-Cabrera, Isla-Diaz, Hernandez-Fernaud, & Rosales-Sanchez, 2012; Lloret-Segura, et al., 2014; Morata-Ramirez & Holgado-Tello, 2013). Based on the foregoing, it was decided to compare the more traditional MLE estimator using Pearson correlations with the robust unweighted least squares (ULSMV) estimator using polychoric correlations. The comparison was carried out presenting the factor analyses results obtained with both estimators in order to check if both of them reached the same conclusion (e.g., Freiberg, Stover, De la Iglesia, & Fernandez, 2013; Holgado-Tello, Chacon-Moscoso, Barbero-Garcia, & Vila-Abad, 2008).
In "order to analyzed the conditions for the factor analysis, the Kaiser Meyer Olkin index (KMO) and the sphericity Bartlett test were taken into account by means of the SPSS software version 24. Furthermore, the following comparative and adjustment indexes were used (Browne & Cudeck, 1993; Hoyle, 1995; Hu & Bentler, 1999; Tanaka, 1993): (i) comparative fit, and Tucker and Lewis indexes (CFI, and TLI), where values of .90 to .95 indicate acceptable fit and values above .95 indicate good fit; (ii) root mean square error of approximation (RMSEA), where values of .05 or lower indicate a well-fitting model, values of .05 to .08 a moderate fit and .10 or greater a poor fit; (iii) Akaike Information Criteria (AIC), and Bayesian Information Criteria (BIC) to compare models, where the lower values the better fit; and finally (iv) the [chi square]/degree of freedom ratio, where values between one and three indicate a great fit, with values below five being acceptable (Carmines & Mclver, 1981; Joreskog, 1970)".
Scale reliability by means of the Cronbach's alpha index and Pearson correlations of the OJS with well-being and emotional exhaustion were calculated using the SPSS software.
Table 3 shows the descriptive statistics. Asymmetry and kurtosis coefficients were below 1, and histograms and p-p plots graphics showed an adequate adjustment to the normal distribution. However, the Kolmogorov-Smirnov test using the Lilliefors correction resulted significant (p < .05), so normality could not be assumed. Because MLE requires the assumption of normality, a robust MLER estimator was used instead. Regarding the Cronbach's alpha indexes, all reliabilities were adequate being above .80. In relation to the properties of the OJS items, means ranged from 2.51 to 3.61, standard deviations ranged from 1 to 1.18, and in any case the reliability of the scale could be improved if one of the items were deleted.
Exploratory Factor Analysis (EFA)
The conditions for the EFA were adequate in the first subsample ([n.sub.1] = 426): KMO = .911 and Bartlett's test: p < .001. As shown in Table 4, the EFA results indicated that the three-factorial structure was the one which presented the best adjustment for both estimators. For this three-factorial model distributive (F1), procedural (F2) and interactional (F3) factors were differentiated. Reliability coefficients for all the obtained factors proved adequate, with values above .80. Nonetheless, as shown in Table 5, item "D.1" loaded in F2, instead of in F1 as previously expected.
According to the Fisher r-to-z transformation (Lenhard & Lenhard, 2014), there was a significant correlation difference between factors F1 and F2 (p = .01), revealing a higher correlation with the ULSMV estimator ([r.sub.12] = .691) than with the MLER estimator ([r.sub.12] = .598). However, there were non-significant differences for all the remaining correlations obtained with the MLER and the ULSMV estimators respectively: [r.sub.13] = .495 vs .536, p = .21, and [r.sub.23] = .346 vs .350, p = .47.
Confirmatory Factor Analysis (CFA)
Table 6 shows the results of the CFA performed with the second subsample ([n.sub.2] = 423) in order to cross-validate the three-factorial model and specifications obtained in the EFA, as well as to compare its fit with that of the three-factorial model proposed by Moliner (2004). The adjustment with MLER for the three-factorial model following the item distribution by Moliner (2004) was worse than that for the three-factorial item distribution in which item "D.1" loads in the procedural factor. It is worthwhile highlighting that the RMSEA values were poorer for the ULSMV estimator, suggesting a mediocre adjustment of the model.
Moreover, a second-order factor analysis was performed with the three-factorial structure obtained in order to check an organizational justice perception construct. Results indicated that whereas adjustment of the model with MLER was acceptable, results for the ULSMV meant a mediocre adjustment.
In order to check if the loading factors depend on the sample's characteristics, several EFA both with MLER and ULSMV were performed, splitting the sample up with regards to gender, the job position, and the type of organization (Table 7). In all cases, the KMO was well above .50, and the Bartlett's test was significant. On the one hand, regarding the MLER, item "D.1" loaded in both distributive and procedural factors for women, technician and managerial, public and private groups; whereas for men and low-ranking groups, this item only loaded in the procedural factor. Moreover, in the women group, item "D.4" also loaded in both the distributive and interactional factors. On the other hand, regarding the ULSMV, similar results were found except for the men group, in which items "D.2" and "D.4" also loaded in both the distributive and the procedural factors. The latter suggests the possibility of a different interpretation of those items related to cultural and social differences.
Regarding the criterion-oriented validity of the OJS, Table 8 shows the correlations between the three-factorial structure obtained in the CFA and the emotional exhaustion and well-being scales. In addition, these correlations were compared with those obtained according to the three-factorial structure initially proposed by Moliner (2004). Results indicated that all the coefficients were significant in the expected way, and according to the Fisher r-to-z transformation (Lenhard & Lenhard, 2014), there were no significant differences between the correlations found with both item distributions.
In this study, the psychometric properties of the OJS (Moliner, 2004) have been analyzed. Regarding hypothesis H1, an adequate model adjustment and reliability indexes are revealed for the three-factorial structure of the OJS, rejecting the possibility of the one and two-factorial structure. This three-factorial structure has been found using both estimation methods, MLER and ULSMV, with the RMSEA value suggesting a worse adjustment of the model with the ULSMV, as well as a higher correlation between the distributive and the procedural factors for this estimator than for the MLER. Moreover, it must be remarked here that one of the items that was supposed to belong to the distributive factor ("D.1") has been assigned to the procedural factor. Specifically, in the multi-group validation using MLER, it was found that item "D.1" also loaded in the procedural factor; whereas with regards to the ULSMV, the men group also revealed that items "D.2" and "D.4" loaded for both the distributive and the procedural factors. This may be due to the cultural and social understanding perceived for these items, suggesting the need for reconsidering their wording. Moreover, hypothesis H2 reveals the existence of a latent organizational justice perception construct. However, second-order factor analysis showed an acceptable adjustment using MLER, but a mediocre adjustment for the ULSMV, which could be related to the problems with understanding of some of the items as the results of the multi-group validation suggested.
Finally, regarding hypotheses H3 and H4, a positive correlation of the OJS with well-being was found as well as a negative correlation with emotional exhaustion, results already suggested by previous studies. These correlations are similar to those we obtained also following the three-factorial structured proposed by Moliner (2004). Thus, this suggests that no difference exists when an assignment of the item "D.1" is made to the procedural factor. Nevertheless, it must be noted that although the signs of the correlations are in the expected directions, the correlation values are low in general.
To conclude, the OJS presents adequate reliability and validity, and on the basis of the three-factorial structure of organizational justice, it can be used for the study of organizational justice perceptions in the Spanish context. However, item distribution was slightly different from the one suggested by Moliner (2004). Moreover, the second-order latent structure showed an acceptable fit for the MLER, whereas a mediocre fit existed for the ULSMV. The latter suggests the need for further studies to corroborate the plausibility of the second-order organizational justice latent variable regardless of the estimator used.
With respect to the limitations of the study, data collection has been carried out by means of self-reports, and this could exacerbate the common method variance given the mono-method bias (Podsakoff, Mcackenzie, Lee, & Podsakoff, 2003). Nonetheless, as Spector (2006) has noted, there is a legend about the assumption of this method alone is sufficient to produce biases because the nature of shared bias depends on both, the construct of interest and how it is measured.
Regarding future lines of research, the multi-group validation results suggested the probability of the presence of differential item functioning which could give light for an in-depth study looking for potentially biased items with different item characteristics curves across groups.
Moreover, a factor invariance analysis could be of interest given some potential interpretations of those items because of cultural and social differences.
Finally, it would be also recommended a discriminant validity analysis of the OJS in order to complete the study of its psychometric properties.
No potential conflict of interest was reported by the authors.
This work has been supported by Ministerio de Economia y Competitividad and Fondos Sociales Europeos under Grant--PSI-2013-44854-R.
Browne, M., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen, & J. S. Long (Eds.), Testing structural equation models (pp. 136-162). Beverly Hills, CA: Sage.
Carmines, E. G., & McIver, J. P. (1981). Analyzing models with unobserved variables. In G. W. Bohrnstedt, & E. F. Borgatta (Eds.), Social measurement: Current issues (pp. 65-115). Beverly Hills: Sage.
Colquitt, J. (2001). On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86(3), 386-400. doi: http://dx.doi.org/10.1037/0021-9010.86.3.386
Colquitt, J., Scott, B., Rodell, J., Long, D., Zapata, C., Conlon, D., & Wesson, M. (2013). Justice at the millennium, a decade later: A meta-analytic test of social exchange and affect-based perspectives. Journal of Applied Psychology, 98(2), 199-236. doi: http://dx.doi.org/10.1037/a0031757
Colquitt, J. A., & Shaw, J. C. (2005). How should organizational justice be measured? In J. Greenberg & J. A. Colquitt (Eds.), The Handbook of organizational justice (pp. 113-152). Mahwah, NJ: Erlbaum.
Cropanzano, R. S., & Ambrose, M. L. (2015). Organizational Justice: Where We Have Been and Where We Are Going. In R. S. Cropanzano, & M. L. Ambrose (Eds.), The Oxford Handbook of Justice in the Workplace (pp. 3-14). New York: Oxford University Press.
Diaz-Gracia, L., Barbaranelli, C., & Moreno-Jimenez, B. (2014). Spanish version of Colquitt's organizational justice scale. Psicothema, 26(4), 538-544. doi: http://dx.doi.org/10.7334/psicothema2014.110
Diaz-Vilela, L., Diaz-Cabrera, D., Isla-Diaz, R., Hernandez-Fernaud, E., & Rosales-Fernandez, C. (2012). Adaptacion al espanol de la escala de Desempeno Civico de Coleman y Borman (2000) y analisis de la estructura empirica del constructo [Spanish adaptation of the Citizenship Performance Questionnaire by Coleman & Borman (2000) and an analysis of the empiric structure of the construct]. Revista de Psicologia del Trabajo y de las Organizaciones, 28(3), 135-149. doi: https://dx.doi.org/10.5093/tr2012a11
Freiberg, A., Stover, J. B., De la Iglesia, G., & Fernandez, M. (2013). Correlaciones policoricas y tetracoricas en estudios factoriales exploratorios y confirmatorios [Polychoric and tetrachoric correlations in exploratory and confirmatory factorial studies]. Ciencias Psicologicas, 7(2), 151-164.
Friedman, H. (1982). Simplified determinations of statistical power, magnitude of effect and research sample size. Educational and Psychological Measurement, 42(2), 521-526. https://doi.org/10.1177/001316448204200214
Garcia-Izquierdo, A. L., Moscoso, S., & Ramos-Villagrasa, P. J. (2012). Reactions to the Fairness of Promotion Methods: Procedural justice and job satisfaction. International Journal of Selection and Assessment, 20(4), 394-4 03. doi: http://dx.doi.org/10.1111/ijsa.12002
Garcia-Izquierdo, A. L., & Ramos-Villagrasa, P. J. (2012). Equality employment opportunity legislation in Europe. Organisational barriers and the organisational justice framework for promotion of women at work. In S. Borelli, & P. Vielle (Eds.), Quality of Employment in Europe Legal and Normative Perspectives (pp. 237-255). Brussels: P. I. E. Peter Lang.
Goldberg, D., & Williams, P. (1996). Cuestionario de Salud General GHQ (General Health Questionnaire). Guia para el usuario de las distintas versiones [General Health Questionnaire GHQ (General Health Questionnaire). User's guide for the different versions]. Barcelona: Masson.
Heponiemi, T., Kuusio, H., Sinervo, T., & Elovainio, M. (2011). Job attitudes and well-being among public vs. private physicians: organizational justice and job control as mediators. The European Journal of Public Health, 21(4), 520-525. doi: https://doi.org/10.1093/eurpub/ckq107
Holgado-Tello, F. P., Chacon-Moscoso, S., Barbero-Garcia, I., & Vila-Abad, E. (2008). Polychoric versus Pearson correlations in exploratory and confirmatory factor analysis of ordinal variables. Quality & Quantity, 44(1), 153. doi: https://doi.org/10.1007/s11135-008-9190-y
Hoyle, R. (1995). Structural equation modeling. Concepts, issues, and applications. London: Sage.
Hu, L., & Bentler, P. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55. doi: http://dx.doi.org/10.1080/10705519909540118
Joreskog, K. G. (1970). A general method for analysis of covariance structures. Biometrika, 57(2), 239-251. doi: http://dx.doi.org/10.1093/biomet/57.2.239
Kulik, C. T., & Li, Y. (2015). The Fork in the Road: Diversity Management and Organizational Justice. In R. S. Cropanzano, & M. L. Ambrose (Eds.), The Oxford Handbook of Justice in the Workplace (pp. 561-576). New York: Oxford University Press.
Lawson, K. J., Noblet, A. J., & Rodwell, J. J. (2009). Promoting employee wellbeing: the relevance of work characteristics and organizational justice. Health Promotion International, 24(3), 223-233. doi: https://doi.org/10.1093/heapro/dap025
Lenhard, W., & Lenhard, A. (2014). Hypothesis Tests for Comparing Correlations. Retrieved from: https://www.psychometrica.de/correlation.html. doi: https://doi.org/10.13140/RG.2.1.2954.1367
Liljegren, M., & Ekberg, K. (2009). The associations between perceived distributive, procedural, and interactional organizational justice, self-rated health and burnout. Work, 33(1), 43-51. doi: https://doi.org/10.3233/WOR-2009-0842
Lloret-Segura, S., Ferreres-Traver, A., Hernandez-Baeza, A., & Tomas-Marco, I. (2014). El analisis factorial exploratorio de los items: una guia practica, revisada y actualizada [Exploratory Item Factor Analysis: A practical guide revised and up-dated]. Anales de Psicologia, 30(3), 1151-1169. doi: http://dx.doi.org/10.6018/analesps.30.3.199361
Matsunaga, M. (2010). How to Factor-Analyze Your Data Right: Do's, Don'ts, and How-To's. International Journal of Psychological Research, 3(1), 97-110. Retrieved from: https://www.redalyc.org/comocitar.oa?id=299023509007
Moliner, C. (2004). Justicia Organizacional, Bienestar del Empleado y Calidad de Servicio en Organizaciones Turisticas: Una Aproximacion Psicosocial [Organizational Justice, Employee Wellbeing and Quality of Service in Tourist Organizations: A Psychosocial Approach]. Valencia: Universitat de Valencia.
Moliner, C., Martinez-Tur, V., Peiro, J. M., & Ramos, J. (2005). Linking organizational justice to burnout: Are men and women different? Psychologic Reports, 96(1), 805-816. doi: https://doi.org/10.2466/pr0.96.3.805-816
Moorman, R. H. (1991). Relationship between organizational justice and organizationa l citizenship behaviors: Do fair ness perceptions influenc e employee citizenship? Journal of Applied Psychology, 76(6), 845-855. doi: http://dx.doi.org/10.1037/0021-9010.76.6.845
Morata-Ramirez, M., & Holgado-Tello, F. (2013). Construct validity of Likert scales through confirmatory factor analysis: A simulation study comparing different methods of estimation based on Pearson and polychoric correlations. International Journal of Social Science Studies, 1(1), 54-61. doi: http://dx.doi.org/10.11114/ijsss.v1i1.27
Muthen, L. K., & Muthen, B. O. (2012). Mplus User's Guide. Seventh Edition. Los Angeles, CA: Muthen & Muthen.
Osca, A., & Lopez-Araujo, B. (2009). ?La Justicia en Seleccion Predice las Intenciones de los Candidatos? [Does Justice in Selection Predict Candidate's Intentions?]. Revista de Psicologia del Trabajo y de las Organizaciones, 25(3), 219-229. Retrieved from http://scielo.iesciii.es/scielo.chp?script=sci_arttext&pid=S1576-5962200900030003&ling=ef&tlng=es
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879-903. doi: http://dx.doi.org/10.1037/0021-9010.88.5.879
Salanova, M., Llorens, S., Grau, R., Schaufeli, W. B., & Peiro, J. M. (2000). Desde el "burnout" al "engagement": ?una nueva perspectiva? [From "burnout" to "engagement": A new perspective?]. Revista de Psicologia del Trabajo y de las Organizaciones, 16(2), 117-134. Retrieved from http://www.copmadrid.org/web/articulos/2000162/trabajo
Schminke, M., Ambrose, M. L., & Cropanzano, R. (2000). The effect of organizational structure on perceptions of procedural fairness. Journal of Applied Psychology, 85(2), 294-304. doi: http://dx.doi.org/10.1037/0021-9010.85.2.294
Sora, B., Caballer, A., Peiro, J. M., Silla, I., & Gracia, F. J. (2010). Moderating influence of organizational justice on the relationship between job insecurity and its outcomes: A multilevel analysis. Economic and Industrial Democracy, 31(4), 613-637, doi: https://doi.org/10.1177/0143831X10365924
Spector, P. E. (2006). Method Variance in Organizational Research: Truth or Urban Legend? Organizational Research Methods, 9(2), 221-232. doi: https://doi.org/10.1177/1094428105284955
Tanaka, J. (1993). Multifaceted conceptions of fit in structural equation models. In K. A. Bollen, & J. S. Long (Eds.), Testing structural equation model (pp. 10-40). Newbury Park, CA: Sage.
Truxillo, D. M., Bauer, T. N., & McCarthy, J. M. (2015). Applicant Fairness Reactions to the Selection Process. In R. S. Cropanzano, & M. L. Ambrose (Eds.), The Oxford Handbook of Justice in the Workplace (pp. 621-640). New York: Oxford University Press.
Whitman, D. S., Caleo, S., Carpenter, N. C., Horner, M. T., & Bernerth, J. B. (2012). Fairness at the Collective Level: A Meta-Analytic Examination of the Consequences and Boundary Conditions of Organizational Justice Climate. Journal of Applied Psychology, 97(4), 776 -791. doi: https://dx.doi.org/10.1037/a0028021
Ana M. Castano and Antonio L. Garcia-Izquierdo
Universidad de Oviedo
Received: November 29, 2017 * Accepted: March 22, 2018
Corresponding author: Antonio L. Garcia-Izquierdo
Facultad de Psicologia
Universidad de Oviedo
33003 Oviedo (Spain)
Table 1 Percentage of participants regarding their sector Private sector Percentage Public sector Primary industry 42.8 Administrative services and security forces Trade 8.7 Education Hotel 8.4 Health and social services Information and 8.1 Primary industry communication Other services 7.9 Scientific and technical Financial, insurances and 4.2 Transport real-estate sector Transport and storage 3.9 Other services Health services 3.6 Education 3.4 Administrative and 3.4 auxiliary services Scientific and technical 3.4 Household support 2.2 Total 100 Total Private sector Percentage Primary industry 33.5 Trade 25 Hotel 17.9 Information and 8.8 communication Other services 7.6 Financial, insurances and 3.6 real-estate sector Transport and storage 3.6 Health services Education Administrative and auxiliary services Scientific and technical Household support Total 100 N = 849 Table 2 Translated items of the OJS (Moliner, 2004) Dimension Items D.1 The rewards I receive here are fair D.2 My retribution is correlated to the quality of the work I do Distributive D.3 I feel fairly rewarded in my work D.4 I have a fair retribution taking into account the hours I work here P.1 Procedures used in this organization to decide my retribution and other income (premiums, etc.) are fair P.2 Procedures used in this organization to evaluate my work Procedural are fair P.3 Procedures used in this organization to place me in a position and /or to promote me are fair P.4 The procedures for setting my work schedule and tasks are fair I.1 My supervisor is very sincere with me I.2 My supervisor treats me with respect and dignity Interactional I.3 My supervisor offers adequate justification for decisions made about my job I.4 My supervisor listens attentively when I ask him/her a question Table 3 Reliability and descriptive statistics Variable [alpha] Range M SD Min. Max. Distributive justice .865 4 20 10.799 3.510 Procedural justice .929 4 20 10.383 3.913 Interactional justice .903 4 20 13.139 4.058 Organization justice total .923 12 60 34.322 9.627 Emotional exhaustion .873 0 30 9.464 6.153 Well-being .806 3 36 21.009 6.147 Variable Asymmetry Kurtosis Distributive justice .451 -.267 Procedural justice .480 -.432 Interactional justice -.254 -.597 Organization justice total .276 -.188 Emotional exhaustion .888 .661 Well-being -.034 -.039 Note: N = 849 Standard error for the asymmetry and the kurtosis was .084 and .168 respectively for all the scales Table 4 Comparison of fit indexes between MLER and ULSMV estimators in the EFA Fit MLER indexes 1-factor 2-factors 3-factors RMSEA .198 .101 .064 CFI .644 .926 .977 TLI .565 .886 .954 [chi square] 951.967 229.719 90.604 Degrees 54 43 33 [chi square]/ 17.629 5.342 2.746 Degrees AIC (a) 12789.918 11783.156 11596.686 BIC (a) 12935.877 11973.715 11827.789 Fit ULSMV indexes 1-factor 2-factors 3-factors RMSEA .267 .144 .083 CFI .693 .929 .982 TLI .625 .891 .963 [chi square] 1695.849 422.954 130.572 Degrees 54 43 33 [chi square]/ 31.405 9.836 3.957 Degrees AIC (a) - - - BIC (a) - - - Note: [n.sub.1] = 426 (a) BIC and AIC are not available for ULSMV Table 5 Comparison of the three-factorial structure between MLER and ULSMV estimators in the EFA Parameter/I tem MLER ULSMV 1 2 3 1 2 3 D.1 .623 .634 D.2 .693 .736 D.3 .817 .985 D.4 .511 .601 P.1 .894 .915 Pattern matrix P.2 .876 .887 P.3 .673 .697 P.4 .769 .833 I.1 .725 .746 I.2 .856 .895 I.3 .805 .834 I.4 .872 .900 [alpha] .845 .937 .895 .845 .937 .895 Note: [n.sub.1]=426 Loadings lower than .400 are omitted. Loadings in a different factor from expected are highlighted in italics Table 6 Comparison of fit indexes between MLER and ULSMV estimators in the CFA Fit indexes MLER Second-order (a) 3-factors (b) 3-factors (c) RMSEA .071 .100 .070 [chi square] 160.842 265.484 152.706 Degrees 51 51 50 [chi square]/Degrees 3.154 5.206 3.054 TLI .947 .896 .949 CFI .959 .920 .962 AIC (d) 11367.394 11512.921 11360.054 BIC 11525.242 11670.769 11521.949 Fit indexes ULSMV Second-order (a) 3-factors (b) 3-factors (c) RMSEA .104 .145 .098 [chi square] 285.132 502.172 251.071 Degrees 51 51 50 [chi square]/Degrees 5.591 9.847 5.021 TLI .950 .904 .956 CFI .961 .925 .967 AIC (d) - - - BIC - - - Note: [n.sub.2] = 423 (a) The organizational justice construct was tested following the item distribution found with the three factors in this study (b) Item distribution suggested by Moliner (2004) (c) Item D.1 assigned to the procedural factor (d) BIC and AIC are not available for ULSMV Table 7 Pattern matrix comparison among gender, job position, and type of organization for the three-factorial structure Women Men Low-ranking Item n = 400 n = 449 n = 375 1 2 3 1 2 3 1 2 3 D.1 .419 .523 .582 .662 D.2 .742 .748 .685 D.3 .724 .781 .923 D.4 .488 .464 .473 .652 P.1 .929 .910 .888 P.2 .900 .845 .810 P.3 .657 .741 .784 P.4 .800 .760 .925 I.1 .810 .653 .632 I.2 .844 .808 .780 I.3 .843 .708 .737 I.4 .928 .810 .854 Technician and Public Private managerial Item n = 474 n = 311 n = 538 1 2 3 1 2 3 1 2 3 D.1 .541 .420 .480 .480 .441 .536 D.2 .833 .770 .850 D.3 .841 .871 .855 D.4 .516 .498 .594 P.1 .888 .887 .902 P.2 .824 .849 .840 P.3 .583 .617 .703 P.4 .650 .780 .744 I.1 .740 .831 .615 I.2 .842 .844 .814 I.3 .759 .826 .704 I.4 .855 .920 .827 Note: Loadings lower than .400 are omitted. Loadings in a different factor from expected are highlighted in italics MLER estimator was used for these results, and similar structures were found using the ULSMV estimator except for the men sample where items D2 and D4 loaded both in the distributive and the procedural factor Table 8 Criterion- validity: comparison of correlations obtained with the suggested three-factorial structure of the OJS and the three-factorial structure obtained in this study Scale Structure Distributive Suggested (a) -.267 (**) Emotional exhaustion Obtained (b) -.277 (**) Test differences p = .412, q = .011 Suggested (a) .200 (**) Well-being Obtained (b) .212 (**) Test differences p = .398, q = .013 Scale Procedural Interactional OJS Total -.192 (**) Emotional exhaustion -.196 (**) -.326 (**) -.313 (**) p = .466, q = .004 .132 (**) Well-being .134 (**) .274 (**) .242 (**) p = .483, q = .002 Note: N = 849 (a) Item distribution suggested by Moliner (2004) (b) Item D.1 assigned to the procedural factor (**) significant at p < .01
|Printer friendly Cite/link Email Feedback|
|Author:||Castano, Ana M.; Garcia-Izquierdo, Antonio L.|
|Date:||Jul 1, 2018|
|Previous Article:||Validity and reliability of the Parental Homework Management Scale.|
|Next Article:||Ergonomia y Psicosociologia aplicada a la prevencion de riesgos laborales.|