Mbi-gs questionnaire pdf free download

This dimension was included in the model at a later time, as a result of the factor analysis of a preliminary version of Maslachs questionnaire Maslach, Some authors have suggested that professional efcacy reects personal traits rather than being a component of burnout Shirom, Research in recent years has suggested that lack of professional accomplishment may contribute to the development of burnout Salanova, Peir, Schaufeli, , and that in this case it may be treated as an independent variable and not as a component of burnout Bres, The correspondences between professional burnout and biographical variables also turn out to be interesting.

Somewhat higher indicators of burnout have been observed among young people under the age of 30; they often appear at the start of a career and lessen with the growth of professional experience Maslach et al.

At the outset of the research described in the present paper, it was however expected that a signicant negative correlation would be found between age or length of career on one hand, and exhaustion and cynicism on the other, on the assumption that older employees have more realistic. There is a lack of unambiguous results on the subject of the relationship between gender and burnout Maslach et al.

It was originally assumed that womens responses would indicate a higher level of burnout, if only because women full other professional roles and occupy lower positions in professional hierarchies. When the research presented here was begun, it was assumed that there might exist differences between the sexes with regard to the three dimensions of burnout hypothesis 4. It was expected that a higher level of exhaustion would be found among women if only due to the additional burden of housework and a lower level of cynicism women, to a greater degree than men, seek fullment outside of professional work, and gain greater emotional support from their families; additionally, greater levels of aggression are found among men , and that men would gain higher results on the professional efcacy subscale men more frequently obtain recognition at work: they are promoted more quickly and receive higher remuneration for their work in comparison with women employed in equivalent positions.

Additionally, it was supposed that differences in the level of professional burnout would be found between particular professional groups hypothesis 5. Among people who work with data information and ideas, and among people performing human service professions, cynicism and the loss of the feeling of professional efcacy may develop more quickly, on account of the high demands on the competence of such people most frequently requiring higher education and lifelong learning , and the great expectations placed on them, both in work and by themselvesin contrast to line workers, whose expectations with regard to work conditions are more humble, and who can more quickly see the effects of their labour.

Method Sample The research, which was carried out during and the rst half of , involved 1, employees of Polish rms. Listwise deletion of missing values on burnout subscales and outliers yielded a nal sample of participants. They were all working people, with various career lengths ranging from half a year to 45 years, although 35 years was the largest interval,. Polish statistical ofce GUS data indicate that in Poland, men predominate among employees at This results from the fact that the most important criterion for selecting respondents was the type of work they performed, and about two thirds of the respondents were involved in professions that are frequently dominated by women including teachers, economists, and administrative workers.

The greatest difference between the sample and the Polish working population was in age: among the respondents were more young people at the beginning of their professional career Monitoring Rynku Pracy, Departament Pracy, GUS, May, On the basis of their type of work, the respondents were divided into four sets: those who 1 work with people; 2 work with data; 3 work with ideas; and 4 work with things cf.

Prediger, In the group of people activities PEOPLE were placed those whose work is based on producing a change in human behavioursuch as teachers, nurses, doctors, psychologists, and sales people people in total ; in the group of information workers DATA were placed computer programmers and economists, as well as administrative workers and specialists people ; in the group of those who work with physical items THINGS , were found production workers, mechanics, locksmiths, carpenters, drivers, farmers, and technicians people.

The nal groupof people who work with concepts, abstractions, theories, and new means of expressing things IDEAS contained only 39 people including scientists, designers, and philosophers , and so it did not undergo further analysis as a separate group except to test the t of the model.

Finally, the three professional groups PEOPLE, DATA, and THINGS were more or less equal in size, and they possessed similar distributions of age and length of career; the only major difference between the groups was in sex distribution see table 1 , on account of the higher number of females in certain professions. Among the representatives of social professions there were more women, and similarly among the data workers there were somewhat fewer women, while among those who worked with things, men dominated.

Then backtranslation into English was performed, and we compared the original English with the back-translated versions. After discussion with a native speaker of English, a nal Polish version was chosen. Measurement The MBIGS questionnaire consists of 16 items comprising three scales: exhaustion EX, 5 items, including I feel tired when I get up in the morning and have to face another day on the job , cynicism CY, 5 items, including I have become less enthusiastic about my work , and professional efcacy EF, 6 items, including In my opinion I am good at my job.

In the MBIGS questionnaire, like in earlier versions, the items constituting the exhaustion and cynicism scales are formulated negatively, while those constituting the efcacyscale are formulated positively.

Respondents answer the question of how often they feel a particular way with reference to a seven-point 06 Likert scale, where 0 indicates never, and 6 means daily. The results, in accordance with the advice of the creators of the scale, are calculated separately for each of the subscales, according to the key none of the items are reverse coded.

High results on the exhaustion and cynicism subscales, and a low result on the professional efcacysubscale, point to a high level of burnout. Apart from answering the individual items of the questionnaire, the respondents were also requested to give some information concerning background variables, such as sex, age, length of career, and profession. Because the research was carried out mostly on-site at various organizations, some individuals, out of fear losing their anonymity, did not give information that might be used to identify them although we emphasized the condentiality and anonymity of the data.

This most frequently took place with the respondents age and time in current job, and the systemic lack of data was found most commonly among people working with things, mainly production workers.

Analysis We examined the factor structure using exploratory factor analysis techniques prior to working with the subscale. We employed principal component analysis with Varimax rotation. The number of factors was determined using a model developed by the authors of the test, with the result that the basis of the construction of the MBI was shown to be a three-dimensional model.

We expected that all the test items would dene in an unambiguous manner those constituents to which they were originally attributed. We also took into account the issue of purity of measurement in this case this refers to whether a given item has high factor loading for one factor, and low factor loadings for the remaining two , and the degree of saturation of an item by a given factor. Evaluation of each model was based on a consideration of t measures.

We examined the data with conrmatory factor analyses using the R projects lavaan package Rosseel, A value less than. Results Exploratory factor analysis The results of the exploratory factor analysis see table 2 revealed that the identied structure of the factors is similar to that presented by Maslach and Leiter , and that the particular items are most strongly connected with the original factors.

Most of the test items full the criteria of measure purity and saturation that we set, with the exception of item MG6 I feel burned out from my work , which has a high second loading on another factor EX and CY , and item MG13 I just want to do my job and not be bothered , which did not gain a factor loading.

We considered one aspect of reliability: internal consistency. Reliability analysis revealed that for the Polish version, the internal consistencies are quite good see table 3. Since two test items MG6 and MG13 raised certain doubts, we were interested in the reliability of the individual subscales after their removal. For the CY subscale, whose reliability was.

For the EX subscale, the elimination of MG6 slightly improved its reliability from. This item had the highest factor loading. Deletion of this item increases the theoretical validity of the instrument, as was mentioned when the results of the exploratory analysis were presented. As the authors of this inventory have shown, the internal consistencies of the MBIGS are satisfactory, ranging from. This might be caused by the ambiguity of the item. They suggest that a high score may indicate disengagement and social isolation, because respondents have cut themselves off from unnecessary contact with coworkers; it may however also indicate strong motivation and work engagement, as a result of respondents concentrating on their work and not welcoming interruptions Schutte et al.

Because it had loadings of lower than. Maslach et al. Conrmatory factor analysis To examine factorial validity, we assessed the t of different models that is, the one- and two-factor models, along with the originally hypothesized three-factor models, in which each of the 16 items loads on its intended subscale or factor, and then the three-factor model which omit MG6, MG13, and both MG6 and MG13 to identify which models best t the data in our study.

First, we assessed the t of the one-factor solutionwhich assumes that all three aspects of burnout load are on one underlying dimension as well as the t of the two and three factor solutions, which assume that the two or three aspects of burnout EX, CY and EF are independent yet correlated factors mean scores, standard deviations, and intercorrelations for each subscale are presented in table 4.

In other words, we investigated whether work burnout is a three dimensional construct, or whether it has another factorial structure. Our results table 4 show that the t of the three-factor solution appears to be somewhat better than that of the oneand two-factor solutions, which additionally conrms our hypothesis H1.

Conrmatory factor analyses show that the hypothesized three-factor structure of the MBIGS is slightly superior to the one- and two-factor models. Since this analysis demonstrated a lack of good t to the data for the original model, we removed the two ambiguous items from the MBIGS MG6 and MG13 in two separate steps and reexamined the respecied model for t.

Each step resulted. Teresa Chirkowska-Smolak, Pawel Kleka Table 3 Alpha coefcients for subscales, and alphas after removing subsequent items. Fit index. The best t turned out to be the model with the two weak items MG13 and MG6 removedthis ts the data of the total samples reasonably well. Two of the t indices were close to.

However the SMRM value was. When the initial model failed to t the data well, we also conducted the same series of conrmatory factor analyses for each professional group separately.

The t statistics are presented in table 5. An acceptable t has been achieved in the DATA sample, but this analysis found that in each occupational group these statistics were lower than for total group. The results of the analysis for the four occupational groups turned out to be a slightly poorer t to the data than in the total sample the best t for DATA subsample.

However, the proposed item model provided a better t for all the subsamples than the one-factor and two-factors models. Intercorrelations We calculated correlation coefcients to assess how strongly the subscales were associated with each other in the Polish sample. They are comparable to those observed in other studiesthe correlations observed among the MBIGS subscales were within the range of the American test manual Schaufeli et al.

The correlations between EX and CY ranged from. This relationship might suggest that exhaustion may play a stronger role in the development of cynicism however, such a conclusion goes beyond the limits of this correlational analysis , as proposed by Leiter in his model of burnout as a developmental process Leiter, In agreement with the assumptions of the burnout model, cynicism is a reaction to excessively burdensome work demands.

Distancing oneself from work is nonetheless a dysfunctional coping strategy, since it lowers engagement and leads to a worsening of professional function, and in consequence a decreased feeling of professional efcacy.

The results of our research support the assumption concerning the connection between the three components of burnout hypothesis 2. At the same time, we had also expected a direct negative relationship between EX and EF, and not just a relationship indirectly through CY. The indirect effect size was Biographical Variables and Burnout An additional aim of our study is to examine the relationships between age, length of career, gender, and occupation.

Age, length of career, and burnout Hypothesis 3 predicted a negative relationship between age on one hand, and exhaustion and cynicism on the other.

We also expected a positive relationship between age and professional efcacy. Gender and burnout We postulated that there exist differences between the sexes with regard to the three dimensions of burnout hypothesis 4.

The results indicate, however, that there is a lack of signicant difference across the genders in levels of exhaustion and the feeling of professional efcacy. The differences in the level of burnout affected only one dimension of burnout, namely cynicism. Also, instead of comparing all men in the sample with all women, it was interesting to compare just the gender makeup of those who had achieved a high professional position.

To this end, the group of people who occupied managerial positions or who were owners of rms were considered separately. In the surveyed population this constituted people, of whom were women and 92 men. The women also had longer career histories, although these differences. In this subgroup there were no differences between sexes in the level of burnout in any of the dimensions. Differences in the level of burnout between professional groups We expected differences in the level of professional burnout between particular professional groups hypothesis 5.

From these analyses we have excluded those who work with ideas IDEA , on account of the size of this subgroup. These professional groups did not however differ signicantly in their feeling of professional efcacy. The group of social services workers does not differ from the remaining two groups in the level of burnout, which conrms the assumption that professional burnout does not so much result from the specic nature of working with people, but rather a mismatch between the work and the worker Maslach, Leiter, ; Chirkowska-Smolak, Table 6 Means, standard deviations, and intercorrelations for exhaustion, cynicism, and professional efcacy.

Discusion and Conclusions Regarding the psychometric properties of the MBI, the internal consistencies all meet the standard of. The three-dimensional model t increased when items MG6 and MG13 were omitted, providing a mediocre but acceptable t.

These results indicated that the proposed three-factor structure of the MBIGS based on 14 items could be replicated in the total sample, as well as across different occupations. The results of the present research are generally in accordance with most of the previous studies that have examined the factorial validity of MBIGS.

We should however emphasize that our results have given only modest support for the three subscales. The degree of mist observed in the hypothesized three-factor model is not unlike that reported by other researchers who examined the factorial structure of the MBI, and is like that reported in the other Polish study conducted by Pasikowski GFI of. Psychometrically speaking, we are dealing with an instrument that is composed of three separate dimensions, although for practical purposes practitioners want the three factors to be collapsed into onewhich would justify using a composite burnout score one based on the scores from the three separate dimensions.

It seems more appropriate to think of burnout not as a unitary phenomenon, but rather as three separate concepts. In future research it would be worth paying more attention to the way in which the particular test items are formulated, as suggested by, for example, Demerouti et al.

They have argued that phrasing all of the items within subscales in the same direction might have yielded an articial clustering of factors due to positively and negatively worded subscales.

Biographical variables, however, did not turn out to be as relevant to the explanation of the burnout phenomenon as we had expected. Although certain relationships were statistically relevant, it is in practice difcult to appreciate their signicance if only on account of the size of the sample, and the fact that the differences do not exceed one standard deviation. A weak relationship exists which indicates that young workers have a slightly less cynical approach to work, and that the feeling of professional efcacy increases with age.

Yet Schaufeli and Buunk recommend caution in drawing conclusions from this research and, referring to the. The reason for authors caution is that the results of the research he performed in the Netherlands showed a signicantly greater incidence of burnout among older employees than among the American employees surveyed by Maslach. Turning to another biographical variable, gender, we see small difference in the level of cynicism, as men distance themselves somewhat more from their work; this difference disappears when we analyse the differences between men and women controlling for their professional situations e.

This indicates that for explaining burnout, organizational variables or occupational groups might be more signicant than biographical variables. In conclusion, this study provided support for the three-factor structure of the MBIGS in a Polish sample of three different occupational groups, conrming the crossnational validity of this instrument, and demonstrating that burnout is not limited to human service professions.

Given the results of this study, we can recommend the Polish version of the MBIGS based, however, on only 14 items to assess burnout in Poland across occupations. References Bakker, A.

Anxiety, Stress, and Coping, 15, Brenninkmeijer, V. How to conduct research on burnout: advantages and disadvantages of a unidimensional approach to burnout. Occupational and Environmental Medicine, 60, Bres, E. Well-being and Performance in Academic Settings: The predicting role of self-efcacy, unpublished dissertation.

Castelln de la Plana: Universitat Jaume I. Byrne, B. Structural equation modeling with Amos: Basic concepts, applications and programming. Mahwah, NJ: Erlbaum. Chirkowska-Smolak, T.

Organizacyjne czynniki wypalenia zawodowego. Ruch Prawniczy, Ekonomiczny i Spoeczny, 4, Cielak, R. Wsparcie spoeczne a wypalenie zawodowe nowe perspektywy bada.

Cierpiakowska Eds. Problemy teorii, diagnozy i praktyki. Prace dedykowane Pani Profesor Helenie Sk pp. Gdask: GWP. Cudeck, R. Alternative ways of assessing model t. Scott Long Eds. Newbury Park, CA: Sage. Demerouti, E. Journal of Applied Psychology, 86, Kalliath, T. Koeske, G. Construct validity of the Maslach Burnout Inventory: A critical review and reconceptualization. The current study follows the quantitative research tradition.

All items are scored on a seven-point frequency rating scale, A cross-sectional survey design was used in the study.

The three components Questionnaires were used to gather primary data by means of a of the Burnout construct are conceptualised in broader terms non-random field survey. Together, the subscales of the MBI-GS provide a three- Research participants and procedure dimensional perspective on burnout. SA Journal of Industrial Psychology five items was administered. The management, including the Station consistency of items 3 and 7, such items be dropped from the Commissioner and the Human Resource Management Head, Article general seven-item version of the scale.

It is in, the proposed study. English, Afrikaans and Setswana home scored on a seven-point frequency scale, ranging from 0 never language speaking police officers from randomly selected to 6 every day. Setswana The latter two languages are the home language of most of the inhabitants of the North West province, with The Afrikaans and Setswana versions 37—46 The original English language measuring instrument was then compared with the retranslated measuring instrument.

All translating was undertaken Higher The questionnaires were Rank Constable 60 8. Sergeant Firstly, the Senior Superint endent 7 1. Structural equation modelling SEM Other Comparing three models in terms of the relationships The fit indices for models 1 and 3 are provided in Table 2. Model among the 21 items tested the construct validity measurement 2 resulted in negative variances, indicating that the model fit was invariance of the translated MBI-GS.

Multi-group confirmatory poor. No further analyses were undertaken regarding Model 2. The null items significantly better than did the other competing models hypothesis H0 postulates that the specification of the factor Model 3 vs.

It was assumed that the errors of items is the fit between the hypothesised model under H0 and the were uncorrelated. The model was over-identified, yielding perfect fit cf. Byrne, Due to the sensitivity of the likelihood distinct sample moments, 40 distinct parameters to be estimated, ratio test LRT to sample size and it being based on the central and degrees of freedom. Such revealed a poor overall fit of Model 3. Values of 0. Regarding the collected in the study. Furthermore, the observed variables loads onto only one factor.

The indicator R2 values for Cognitive Weariness, Exhaustion and Professional variables in the model were treated as continuous variables. Efficacy were 0.

Hypothesis 1, which postulates that burnout can be characterised as a four-dimensional construct that distinguishes between The following hypothesised models of the adapted MBI-GS Exhaustion, Cognitive Weariness, Cynicism, and Professional were tested: Efficacy, is thus accepted.

The measurement invariance of the adapted Cognitive Weariness, which made up Fatigue. All the were found. This finding confirms the lack of measurement tests of invariance across the samples began with a global test invariance of the adapted MBI-GS for the three samples. In testing for such equivalencies, the sets of parameters are tested Given that the factor structure of the adapted MBI-GS was not in a logical order and by increasing the restrictions in every invariant for the three samples, it was decided to investigate step.

The sets of parameters that are of most interest regarding the standardised regression coefficients of the 20 items of the group variances are: a the factor-loading paths; b the factor adapted MBI-GS in the unconstrained model see Table 4.

The results of the multi-group analyses to assess the measurement The most notable differences between the standardised invariance of the adapted MBI-GS for the Afrikaans, English and regression coefficients in the three groups were evident for CW3 Setswana home language speaking groups are reported in Table 3.

It for the baseline unconstrained model in the three samples. In the was then decided to re-specify the measurement model after the first step of the multi-group analysis, the measurement weights three items had been deleted. In home language speaking groups. Exhaustion 2. Cognitive Weariness 2. Cynicism 2. Professional Efficacy 4. Cognitive Weariness 0. Cynicism 0. Professional Efficacy In the first step of the multi-group Afrikaans samples.

Furthermore, Cynicism is also positively analysis, the measurement weights of the adapted MBI-GS statistically significantly related to Cognitive Weariness large were equally constrained.

In subsequent steps, and statistically significantly related to Cognitive Weariness the measurement weights, structural weights, and structural medium effects and Cynicism medium effects in the English SA Journal of Industrial Psychology covariances were equally constrained between the three samples.

This home language speaking sample, Exhaustion is statistically and finding confirms that the measurement invariance of the adapted practically significantly related to Cognitive Weariness and MBI-GS for the three samples is still not acceptable, even after Cynicism medium effects.

The results support the construct validity of an adapted members is not confirmed. Hypothesis 2 is therefore rejected.

The adapted and translated MBI-GS, however, showed a lack of measurement Descriptive statistics, reliability and the invariance for participants who completed the English, Afrikaans relationship between the burnout constructs and Setswana versions of the instrument. The reliabilities of the Given the lack of measurement invariance of the adapted and Exhaustion and Professional Efficacy scales were acceptable for translated MBI-GS, it was necessary to compute descriptive the three language groups.

However, the reliabilities of the CWS statistics, reliability, and correlations for the English, Afrikaans and Cynicism scale were not acceptable. The first step in achieving this was to determine the construct The descriptive statistics and alpha coefficients of the adapted validity of the adapted burnout model, including exhaustion, and translated versions of the MBI-GS for the three language cognitive weariness, cynicism and professional efficacy.

This groups are reported in Table 5. Cynicism Weariness, and Professional Efficacy, showed acceptable had the highest loading on Burnout, followed by Cognitive reliabilities. However, the alpha coefficient of Cynicism was just Weariness, Exhaustion, and Professional Efficacy.

The results below the recommended value of 0. Such results provide partial explained their predictors. The results support the previous support for Hypothesis 3. The results did not support and statistically significantly related to Cognitive Weariness the measurement invariance of the adapted and translated large effects and Cynicism large effects in the English and MBI-GS.

Such findings were in accordance with a study model. The same tendency was found when structural weights conducted by Schaufeli, Leiter and Kalimo , in which the and structural covariances were equally constrained between researchers found that exhaustion was associated with mental the three language groups.

It could, therefore, be presumed strain. To a certain extent, the results of the study discussed that, even though the translations and backward translations in this article are similar to those of Coetzee and Rothmann were accurate, the different cultures could still ascribe different , who found that Afrikaans and English home language meanings to individual or collective items.

The construct being speakers reported experiencing higher levels of exhaustion and investigated might, therefore, assume a different meaning in the cognitive weariness than did the indigenous language groups. Limitations and recommendations Analysis of the standardised regression coefficients showed The study described in the current article had various that the three translated items made a notable contribution to limitations. Firstly, the participants in the study consisted only the lack of measurement invariance.

Also, the sample size in general was relatively as acceptable. Although it can be concluded that the factor structures of the adapted and translated MBI-GS are acceptable in the English, SA Journal of Industrial Psychology The implication of the results of the study under review is that Afrikaans and Setswana home language speaking groups, the the adapted MBI-GS scores obtained by the English, Afrikaans measurement invariance of the measure is not acceptable.

The and Setswana home language speaking participants were lack of measurement invariance can be attributed to one of the not comparable in terms of the different translations of the following: 1 semantic differences, resulting from the translation instrument.

Therefore, separate norms have to be developed of the original English text into Setswana and Afrikaans; 2 Article to assess the extent of burnout experienced by the English, the cultural differences between the three different language Afrikaans and Setswana home language speaking members groups; 3 both the semantic and cultural differences existing of the SAPS who participated in the study.

As the translated between the three different language groups. However, it is versions of the adapted MBI-GS are clearly not equivalent, more more plausible that measurement invariance in the adapted studies are needed to minimise the semantic differences existing and translated MBI-GS is the result of semantic differences between the different translations of the instrument.

The use associated with the former cause. Storm The results obtained in such studies could then be compared and Rothmann showed the measurement invariance of with those obtained in the present study.

Furthermore, the study the MBI-GS when it was administered in English to samples of reviewed in the current article translated the two instruments SAPS members from different cultural groups.

Translating the MBI-GS into the other eight official The reliability analyses showed that Exhaustion and Professional languages as well would promote the in-depth investigation of Efficacy, being two of the four scales of the adapted and burnout across cultures in South Africa.

Such results support the Since the present study was conducted among a relatively findings made in various international and South African studies homogenous group, other studies, exploring a greater variety e.

Furthermore, it is the results showed that the CWS was not sufficiently reliable for recommended that external variables, such as job characteristics, the Setswana home language speakers, and that the reliability of ill health and personality variables be investigated by means of the Cynicism scale was lower than the recommended guideline additional research into the phenomenon of burnout in South for both the English and Setswana home language speaking Africa. Furthermore, very supported by the National Research Foundation, under reference high correlations were found between Exhaustion, Cognitive number TTK Arbuckle, J.

Amos 7. Burger, D. The land and its people. Retrieved September 27, , from L. Biological Psychology, 69, — Burke, R. Hospital restructuring and Equal Opportunities Schaufeli, W. Mahwah: Erlbaum. Schaufeli, W. The General Cheung, Y.

Developing health-related Burnout Inventory: A self-report questionnaire to assess burnout quality-of-life instruments for use in Asia: The issues.