Factors Related to Organizational Commitment of Jordanian Auditors: A Structural Model

The relationships between neuroticism personality and workload and organizational commitment are explored in this study, with job burnout as the potential mediator. The study sample comprised 183 Jordanian auditors and the obtained data prove full mediation of job burnout on the association between workload and organizational commitment, and partial mediation of job burnout on the association between neuroticism personality and organizational commitment. Job burnout and organizational commitment are directly and negatively related. An integrated model grounded by three factors contributing to organizational commitment of auditors is presented in this study.


Introduction
Organizational commitment of employees is influenced by countless of factors, which, many employers have long attempted to comprehend (Muldoon et al., 2017). In the extant literature, Djafri and Noordin (2017), reported organizational commitment as among the most extensively scrutinised variables. In this regard, Kaur (2017), mentioned that most researches on organizational commitment were concentrating on its defining antecedents and consequences. Such extensive interest may be caused by the impact imparted by organizational commitment on attitudes and behaviours such as: turnover, intention to leave, attitudes toward organizational changes, organizational citizenship behaviours (OCBs), and retaining high performance level in organization (Allen and Meyer, 1996;Meyer and Herscovitch, 2001).
Organizational commitment is a concept that has varied definitions. In the work by Mowday et al. (1979), this concept is seen as the comparative strength of the identification and engrossment of a person inside a given organization, whereas in Mowday et al. (1979), and Sanders III et al. (2005), organizational commitment relates to how far the goals and values embraced by employees correspond with the values and norms upheld by the organization. In the context of public accounting companies, the attitudes and behaviours exhibited by audit staff greatly influence the success (or failure) of the company. Appositely, Donnelly et al. (2011), mentioned organizational commitment as a very important attitude of auditors. In fact, as mentioned in several past works (Mathieu and Zajac, 1990;Norris and Niebuhr, 1984;Omar and Ahmad, 2014), commitment decreases absenteeism aside from improving job performance.
Organizational commitment plays a crucial role in an organization, and as mentioned in Martinow et al. (2016), organizational commitment may also become an indicator to the delivered audit quality while also reflecting the reputation of audit profession. Accordingly, lack of commitment may cause turnover to increase and in the context of audit quality, such lack can be a threat. However, it should be noted that low commitment will not necessarily cause auditors to abandon their profession, but having low level of commitment causes these auditors to lack motivation to perform as diligent as they should and this can affect audit quality. Hence, auditors who have low level of commitment but still decide to stay are arguably a bigger threat to the company as opposed to their peers who left the firm or the profession; in fact, the impact of those who left is not readily observable. As organizational commitment impacts the performance of employees, Meyer et al. (2002), suggested that it should be made a priority in the relationship between employee and management.
Furthermore, Savery and Syme (1996), reported that any effort to improve organizational commitment is useful in in reducing of stress levels in the job particularly in staffing and also in the perceived job pressure. In the context of developing countries particularly, reducing the level of stress is crucial (Labatmedienė et al., 2007).
Auditors are continually pressed to produce quality work (Brown and Mendenhall, 1995), and more output within lesser time (Larson, 2011). As a result, auditors are obliged to work long hours as the deadlines are tight, which, leads to high level of work stress. High level of work stress affects the attitude of employees towards their job as well as how they feel towards their organization. Furthermore, how the employees feel about the organization reflects their organizational commitment. On the other hand, job burnout is construed the attitude that individual has towards the job. As reported by Brown and Campbell (1994), and Maslach et al. (2001), both high levels of occupational stress and burnout appear to lead to decreased productivity. In addition, studies have found that job-related behaviours including job satisfaction, job performance, learning, motivation, as well as organizational commitment has negative association with high stress and burnout levels (Al Shbail et al., 2018b;Cropanzano et al., 2003;Davey et al., 2001;Demerouti et al., 2002;Llorens et al., 2006). Also, playing an important role, job burnout also may be reflected in the quality of the conveyed audit as well as in the reputation of the profession (Al Shbail et al., 2018b). As mentioned in past works (Chong and Monroe, 2015;Fogarty et al., 2000;Fogarty and Kalbers, 2006;Sweeney and Summers, 2002). (Almer and Kaplan, 2002;King'ori, 2016), job burnout is a critical factor that impacts the behaviours exhibited by auditors. It is thus possible that auditors experiencing burnout of high levels bring higher threat to audit quality. In this regard, George (2015), expressed the importance of examining the psychological aspects of employees, and relevant to these mentioned aspects, the present study has chosen to focus on job burnout and organizational commitment.
The preliminary studies on burnout were generally anecdotal and descriptive, and as mentioned in Silbiger et al. (2017), many of these works did not present appropriate theoretical contributions or any empirical evidence in lending support to improving the subject. In specific, Smith and Emerson (2017), mentioned workload to be a constant burnout predictor in the context of public accounting. Relevantly, Maslach et al. (2001), reported workout to predict burnout in other professions. Attempts have been made to conceptualize and test the frameworks of burnout within the domain of auditing, and as shown by the extant literature, there exist a number of possible burnout antecedents such as role ambiguity and role conflict (Chong and Monroe, 2015;Low et al., 2001).
Many researchers have extensively discussed the contributing factors of burnout and in this context; the burnout paradigm of Maslach has been the primary focus (Maslach and Leiter, 1997). Here, burnout is regarded as an output of environmental conditions. Somehow, as highlighted in Ghorpade et al. (2011), the paradigm does not provide a comprehensive account for burnout. The authors mentioned the increased attention to the correlation of role of personality with burnout. In the work of Bühler and Land (2003), the authors attempted to comprehend why even within the exact working conditions, not everyone experience burnout. For this reason, including both environment and personality as burnout determinants would seem appropriate in expanding this line of inquiry.
In Swider and Zimmerman (2010), burnout antecedents were discussed. According to Maslach et al. (2001), andShirom (2003), burnout antecedents fall into three different levels: organisational, occupational, and individual. In addition, Swider and Zimmerman (2010) criticised the biased focus on organisational factors in addition to the exclusion of individual level causes such as personality.
The past works and the presented models assisted in describing the connexion that relates to burnout (Chong and Monroe, 2015;Smith and Emerson, 2017). Accordingly, the present study is unique in a sense that it examines burnout as a possible mediating attribute in workload, neuroticism personality trait with apposite outcome (organizational commitment) within Jordanian public accounting.
Hence, in this study, the question of whether or not burnout can partially or fully mediate such stressorsoutcome relationships is explored. For the purpose, Job-Person fit theory is employed in the present study to determine if auditors that have higher neuroticism personality level will experience high level of job burnout and low organizational commitment level. As such, the present study will attempt to empirically and conceptually contribute in the knowledge of burnout in the context of Jordan. Gilbert (2013), highlighted a combination of work-factors and personality-factors as predictive antecedents of burnout. As described in Cervone and Winer (2010), personality entails psychological attributes which contribute to the persistent and unique patterns of an individual pertaining to his/her feelings, thoughts and behaviour. Meanwhile, Swider and Zimmerman (2010) highlighted personality as a primary individual-level antecedent of job burnout. Considering that personality is fairly unchanging as opposed to other situational predictors, for instance, workload, it should be treated as the main research focus so that the scope of findings could be expanded.

Neuroticism, Burnout and Organizational Commitment
In this paper, the Big-Five Personality factors are referred. Among the factors (neuroticism, extraversion, agreeableness, conscientiousness and openness to experience), this study has chosen to explore the factor of neuroticism because past studies have indicated this factor being the predictor of countless outcomes that are associated with work including career success and deviance, while also denoting maladaptive and adaptive coping styles (Barrick and Mount, 2000;Colbert et al., 2004;Judge et al., 1999). Hence, Connor- Smith and Flachsbart (2007), suggested the high possibility of these styles to impact the accessibility and usage of coping resources. Relevantly, Schaufeli and Enzmann (1998) highlighted the presence of findings that lend support to the view that possessing coping resources in dealing with work-related demands will decrease the likelihood of individuals in showing strains including burnout (Hobfoll, 2002). Hence, in regards to neuroticism personality and job burnout taking in to account past findings, the hypothesis to be tested in this study is as follows; H1a: Neuroticism personality will be positively related to job burnout.
The situational antecedents of commitment have been explored in many studies (Meyer et al., 2002;Morrow, 2011). On the other hand, the dispositional antecedents of commitment such as personality traits and locus of control have not been sufficiently explored; which different compared to the case of job burnout. Taking into consideration the strong linkage between job burnout and organisational commitment, such discovery is rather unexpected. More empirical researches on the relationship between personality and organisational commitment should therefore be carried out. Apart from that, there have been insufficient works that look into the linkage between Big Five traits and commitment, while the ones available in the literature are only a few such as Erdheim et al. (2006), and Panaccio and Vandenberghe (2012), and were primarily carried out in developed nations. Hence, the present paper will highlight the mechanisms through neuroticism personality as it has been reported to have potential to predispose auditors to experience commitment.
Meanwhile, the knowledge pertaining to the contribution of auditor's personality in organisational commitment development of would provide interesting information for practitioners particularly with respect to auditor selection and recruitment. In actuality, the focus of the extant literature has been on the linkage between personality and behavioural outcomes, including performance such as in Barrick and Mount (1991), Salgado (1997), and Detrick et al. (2004), whereas the impact of personality on job attitudes has not been addressed.
Neuroticism relates to an inclination to demonstrate poor emotional adjustment and feel adverse impact (Judge et al., 2002b). According to Judge et al. (2002a), neurotic person partly intentionally places him/herself into situations that creates negative impact. In addition, Connor- Smith and Flachsbart (2007), reported the linkage between neuroticism and various maladaptive coping mechanisms (e.g., withdrawal, disengagement, and wishful thinking), while Zimmerman et al. (in press) reported the linkage between neuroticism and low job search selfefficacy. Also, neurotic individual has the inclination to mull over the negative side of things (Bono and Judge, 2004), and appears to see neutral events as problematic (Duffy et al., 2006), apart from having the inclination to feel angry and have mood swings (Watson et al., 1994). They are also inclined to exhibit more dysfunctional behaviour (Obeid et al., 2017), feel lower well-being, more stress Ozer and Benet-Martinez (2006), and more burnout (Al Shbail et al., 2018a;Zimmerman et al., 2012). White et al. (2004), further added that neurotic person appears to have poor relationship satisfaction with their partners. Work-wise, such outcomes may cause neurotics to have few productive work experiences (productive work experience contributes to the generation of positive commitment to the organization).
The reviewed literature shows the dearth of studies that explored the linkage between personality and organisational commitment. Erdheim et al. (2006), are among the few that reported the significant linkage between neuroticism personality and continuance commitment. Furthermore, negative relationship was reported between neuroticism and organizational commitment in Raja et al. (2004), Gelade et al. (2006), and Kumar and Bakhshi (2010). In addition, the authors described neuroticism as the primary source of negative affectivity. As such, taking in to account past findings, the hypothesis to be tested in this study is as follows; H1b: Neuroticism personality will be negatively related to organizational commitment.

Workload, Burnout and Organizational Commitment
In public accounting, the "busy season" is usually during the first calendar quarter. During this time, most audit works are performed, tax returns are to be submitted, and professional services are sought after. During this period, the workload is increased, leading to substantial level of stress. Jones et al. (2010), reported that during the busy season each year, those working in public accounting sector have to deal with the long hours and packed workloads. As reported in Sweeney and Summers (2002), and Jones et al. (2010), in general, the period is marked by elevated stress and possible degradation in performance of auditor.
As shown in past studies, pressures of busy season including auditor burnout can cause audit quality to degrade at the individual auditor level (Sweeney and Summers, 2002). The busy season is crucial in the context of public accounting industry. Still, in the literature, there have been insufficient studies that look into the influence of the substantial workload during this period on critical behavioural variables, attitudes, or outcomes.
Also, among public accountants, burnout is the term that appears commonly during the busy season. In this context, burnout is often used to illustrate the feeling of exhaustion caused by workplace demands. On the other hand, "Job burnout," as explained in Cordes and Dougherty (1993), Maslach (1982), and Sweeney and Summers (2002), is a specified syndrome of psychological stress syndrome in which a pattern of negative responses are the outcomes of work demands or stressors.
Possibly the most frequently cited antecedent to burnout is workload or role overload which was described in Bowling and Kirkendall (2012), as a broad term which encompasses variables that denote the amount or complexity of one's work. Meanwhile, workload was described in Bacharach et al. (1991) as the resultant of having to achieve too many things in a stipulated duration of time. Authors of past studies are all in agreement that too much extended work demands will cause drainage to emotional resources and energy, leading to burnout, job tension, in addition to job dissatisfaction (Cordes and Dougherty, 1993). Lee and Ashforth (1996) and Schaufeli and Enzmann (1998) further added that workload reflects a job demand and often, it is regarded as the main predictor of burnout. Smith and Emerson (2017), reported the presence of high work overload problem among auditors. Also, some studies including Sweeney and Summers (2002), regard workload as predictor to burnout levels among auditors. In addition, among public accountant, Smith and Emerson (2017), found a significant relationship between burnout and work overload. Therefore, the hypothesis below is presented; H2a: Workload will be positively related to job burnout.
Higher job stressors experienced by employees cause their organizational commitment to reduce. Fitzgerald et al. (2003), noted that job stressors impact both behaviours and also attitudes of workplace. Accordingly, facing workloads can cause employees to feel astounded by the expectation of successful task completion and this may reduce their positive feelings and attachment toward their organization (Pooja et al., 2016). Meanwhile, Hakanen et al. (2008), found that higher workloads cause work engagement to decrease. In addition, Saks (2006) reported a positive relation between engagement and organizational commitment. In Ahuja et al. (2007), excessive workload appears to result in lower organizational commitment levels amongst employees. Also, work overload and organizational commitment was reported to have significant negative linkage in Jamal (1990), Fisher (2014), and Pooja et al. (2016). As can be deduced from past findings, workload should have negative linkage to organizational commitment. The following hypothesis is therefore proposed; H2b: Workload will be negatively related to organizational commitment.

Job Burnout and Organizational Commitment
Organizational psychology is about a comprehension regarding attitudes and behaviours within the environments of work Gilliland and Chan (2001). Meanwhile, being regarded as a vital organizational variable, job burnout has been associated with the deterioration of audit quality Smith and Emerson (2017) and of organizational commitment (Akar, 2018;Hakanen et al., 2008;Li, 2014). As indicated in Leiter (2009), due to its practical significance from the viewpoint of an organization, the connexion that burnout has with organizational commitment has been of particular interest. In past studies (Mathieu and Zajac, 1990;Shore et al., 1995;Whitener and Walz, 1993), correlations between organizational commitment and other variables linked to organization (e.g., turnover intentions, absenteeism, and job performance) have been reported. Hence, as noted by Randall (1987), work attitude therefore is related to organizations, and also significant to the well-being of individual employees. Chen and Yu (2014), mentioned the mixed finding regarding the relation between burnout and organizational commitment. For this reason, questions regarding if burnout actually has significant linkage to organizational commitment has emerged (Ashill and Rod, 2011;Hakanen et al., 2008). In addition, the association between burnout and organizational commitment still needs to be explored further. Similarly, more studies need to also be carried out on the relationship between job burnout and organizational commitment in the domain of auditing.
As suggested by past studies (Burke et al., 1984;Chong and Monroe, 2015;Fogarty et al., 2000;Leiter and Maslach, 1988;Maslach, 1982;Salehi and Gholtash, 2011), reduced organisational commitment level is a crucial outcome of job burnout. Such linkage has been reported among police, public accountants, faculty members as well as employees from agencies of government. As reported in , lower organisational commitment seems to be present among internal auditors who feel depersonalised and less accomplished personally. The hypothesis below is therefore presented; H3: Job burnout will be negatively related to organizational commitment.

Job Burnout as a Mediator
As posited in modern theories such as Mischel and Shoda (1995), Mischel and Shoda (1998), cognitiveaffective personality system (CAPS), the personality of a person impacts the manner in which he or she encodes or assesses the surrounding information. As suggested, the mental encodings of individuals regarding their expectancies and beliefs, affective and physiological response following events, as well as self-regulatory plans, govern the individual's impulsive behavioural inclinations, frustrations, and also fears. Known as cognitive-affective units, these encodings mediate the deliberations on personality-behaviour connexions (Mischel and Ayduk, 2002). It is possible for job burnout to be a set of mental encodings possessed by individuals regarding their reactions and responses towards the constant stress they experience at work. The value added by dispositional-and also processbased investigations to the knowledge regarding the manner in which people cope and react to stress was highlighted by Mischel and Shoda (1998), in detail.
Just like coping, job burnout might be applicable to dispositional-and process-based models in the description of the behaviour of people at work. Hence, this study will attempt to describe how individuals' stable personality traits impact how they respond to situational aspects at work and the behaviours that they show as a response. For instance, it is possible that neurotic individuals will assess and encode a change in their environment (e.g., increased workload) in manners that are distinct as opposed those who have more emotional stability. Neurotic individuals may be inclined to encode change and this causes them to feel emotionally drained and demonstrate aloofness from the job, or these individuals may fail to attain past performance levels which may impact the ensuing outcomes of work.
In addition, in describing the characteristics of stress and outcomes, theories on stress often recognise the importance of person and environment. This can be seen in the research by Riedo et al. (2017), that employs constructs associated with personality traits in studying people experiencing stress. Meanwhile, in other related studies, Rotter (1966), employed locus of control whereas Kobasa (1979), used the construct of hardiness. In past studies, stress has also been associated with workload and under load (French and Caplan, 1972), job demands and decision leverage Karasek and Theorell (1992), and time pressure (Mark and Smith, 2008).
The person and environment make up the bi-factor stress, and in stress studies, and bi-factor stress becomes a facet of the interactive psychological perspective (Lewin, 1951;Magnusson and Endler, 1977;Pervin, 1989). This perspective highlights the joint determination of the individual and the environment regarding the individual's behaviour, attitudes, and well-being. This theory encompasses several dimensions which enable estimation of general individual outcomes such as stress, environmental, and adjustments, which ultimately cause burnout at workplace.
The application of burnout as a mediator of stressors amongst auditors began from the models proposed in Fogarty et al. (2000), Jones et al. (2010), and Smith and Emerson (2017). As mentioned in Smith and Emerson (2017), initial works on role stress were mainly concerned with the direct associations between job-related stressors (e.g., workload) and key outcomes which include organizational commitment and performance. With respect to these works, the outcomes are often contradictory. With respect to these mixed outcomes, Fogarty et al. (2000), related them to misspecification bias caused by the exclusion of primary variables that relate role stressors with job outcomes.
Relevantly, studies since the last two decades have used the primary mediator variables. Such usage has been to keep refining the stress paradigm among both auditors and accountants, and these include job burnout. For this reason, it is possible for job burnout to be projected by workload and neuroticism and become a mediating connexion between workload, neuroticism personality and organizational commitment. As such, this study proposes the following hypotheses; H4a: Job burnout mediates the relationship between neuroticism and organizational commitment. H4b: Job burnout mediates the relationship between workload and organizational commitment.

Participants
Partakers of this study were auditors from several local and international audit firms in Jordan, and the survey was carried out in Spring 2018. The questionnaires as an instrument of this study were disseminated to these participating auditors. A cover letter was included with the questionnaire. This letter was to inform the partakers of the assurance of confidentiality towards the responses that they provide. The partakers were also assured that they could not be individually identified by the researcher, and that the data would be analysed in summative manner and reported for academic research purposes. The partakers were to return the finished instrument in an enclosed envelope to an administrative staff. The returned questionnaires were then collected by the researchers.
This study reduced the prospective of single-source method bias by having the items comprising the dependent and independent variables separated, while items in each set were combined all through the questionnaire (Podsakoff et al., 2003). In this study, the independent variables are neuroticism personality and workload, whereas the dependent variables are signified by job burnout and organizational commitment. Pilot study was also carried out involving 40 auditors. Following that, minor changes were made to the wording. Then, the formal survey was carried out.
This study disseminated a total of 380 instruments, and obtained 183 usable responses which is equivalent to 48% response rate. Participants in this study were classed into 2 groups of identical size namely early participants group and late participants group. As revealed from the comparison made to the demographic information and survey responses for the two groups, no statistically significant differences were present. The rate of response seems judicious for survey research. However, the comprehensive demographic information for the pool of prospective respondents was not obtained. For this reason, problems of common method bias need to be addressed (Podsakoff et al., 2012). For the purpose, a full collinearity test (VIF) 1 was executed. The use of VIF has been proposed in some studies (Al Shbail et al., 2018a;Kock, 2015;Rasoolimanesh et al., 2015). As evidenced by the outcome of analysis, there was no problem of common method bias.
With respect to the respondents participating in this study, 43 of them were audit seniors or "in-charge" auditors, while 37 of them were audit managers. The mean age of the respondents was 29, while in terms of experience in the profession, the mean attained was five years. Male respondents made up the majority at 82%, and with respect to level of education, the majority (74%) were bachelor degree's holder, while the remainder (26%) had master's degree.

Measures and Scales
Organisational Commitment: This construct was measured using an instrument containing 9 items from Mowday et al. (1979). The instrument employs seven-point Likert-type scale (from "1" for strongly disagree to "7" for strongly agree), whereby the scale of 7 denotes organizational commitment of high level. The reliability and validity of for Mowday et al. (1979), nine-item instrument have been reported by past works (Baack et al., 1993;Donnelly et al., 2003;Kwon and Banks, 2004;McManus and Subramaniam, 2014;Sommer et al., 1996;Steers, 1977). Appositely, the Cronbach alpha for this instrument was 0.864 in this study.
Burnout: This construct was measured using the scale-short version of burnout containing 10 items that Malach-Pines (2005), had developed. In this study, the items were presented in Arabic language. This is to ease the measurement of auditors' perceptions for vocational burnout. The items employ 7-point scale for the respondents to indicate their degree of agreement. In this work, the items were tested again for their scale validity and reliability and as shown by the outcomes of the confirmatory factor analysis, the construct of the scale has compatibility with the research data. Here, the scale achieved Cronbach Alpha coefficient of 0.903.
Neuroticism: This construct was measured using 12 items developed by Goldberg (1992), and the items are in semantic differential arrangement, for instance, at ease-nervous, discontented-contented. In responding to the items of this construct, the respondents were provided with 7 options response. Cronbach alpha of 0.871 was attained for this construct.
Workload: This construct was measured using four items obtained from past works (Janssen, 2001;Pooja et al., 2016). The response options for respondents to choose for these items are: "often worked under time pressure" or "often had problems with the pace of work." Cronbach alpha coefficient of 0.887 was attained for this construct scale.

Data Analysis
PLS Path Modeling (PLS-PM) was the method of analysis chosen in this study. The decision to employ PLS-PM was underpinned by the following reasons: 1) PLS-PM shows aptness in testing mediation impacts where a feature for completing an indirect effect with various options is available (Al-Shbiel et al., 2018;Obeid et al., 2017). PLS-PM has advanced in a swift manner and therefore with the indices of goodness of fit, it has the capacity to test the theory (Al-Shbiel et al., 2018). The use of PLS-PM allows simultaneous testing of connexions among variables (Ringle et al., 2018). Hence, using PLS-PM, the research hypotheses 1 through 4 are simultaneously tested. Also, consistent estimators (consistent PLS (PLSc)) as proposed by Dijkstra and Henseler (2015), are used to simulate the approach of covariance-based SEM to test or affirm the theory.
It should be noted that the use of PLSc necessitates large sample size (at least 100 cases) (Latan and Ghozali, 2015;Latan et al., 2017;Latan et al., 2018). PLS-PM has been used by past similar works for analysis purposes (Al-Shbiel, 2016;Al-Shbiel et al., 2018;Al Shbail et al., 2018a;2018b;Obeid et al., 2017). As opposed to other multivariate techniques, PLS-PM has no reliance upon the normality assumption (distribution-free). This is owes to PLS-PM being non-parametric in nature (Latan et al., 2018). Still, for assessment of local model, there should be consideration on assumptions including multicollinearity and goodness of fit indices.
In general, there are four stages of hypothesis data analysis. In the first stage, the measurement model is evaluated. This is to assure the reliability and validity of the construct indicators. During the second stage, the structural model is evaluated. This is to determine the aptness of the model with observational data. Next in the third stage, the direct impacts are tested. This is for ascertaining the relationship between each predictor variable and the obtained result particularly for all direct hypotheses. In the last stage, the indirect impacts on the relationship between neuroticism personality, workload and organizational commitment is tested, with job burnout as mediator.

Measurement Model
Test was carried out on the measurement model. This was to evaluate the constructs' internal consistency reliability, convergent validity and discriminant validity that this study uses. As mentioned in Al-Shbiel et al. (2018), and Hair et al. (2014), internal consistency reliability assesses the degree to which the items function as a measure of the latent constructs. Hair et al. (2017), describe composite reliability as a measure of internal consistency. Furthermore, for each construct, the model of measurement with composite reliability/rho A greater than the threshold value of 0.7 is regarded as satisfactory (Nunnally and Bernstein, 1994;Richter et al., 2016). As can be seen from the obtained outcomes, the composite reliability for all of the constructs is higher than the cut-off value of 0.70. In specific, neuroticism personality scores 0.894, workload scores 0.896, job burnout scores 0.905, and organizational commitment scores 0.895. All these obtained scores denote the measures' high internal consistency.
Convergent validity is also important to address. As explained in Hair et al. (2017), convergent validity evaluates "the extent to which a measure correlates positively with alternative measures of the same construct" (p. 112). In order to evaluate convergent validity, the outer loading of the items and the average variance extracted (AVE) are checked. The loading factor, average variance extracted (AVE) as well as reliability value obtained from the measurement model analysis for each variable had loading factor > 0.70 and AVE > 0.50, and according to Henseler (2017), obtaining such values means that rule of thumb proposed is satisfied. It should be noted that several questions had a loading factor < 0.70 but if AVE value > 0.50 they are tolerable. In Hair et al. (2017), it is suggested that indicators that have weaker factor loadings is retained if other indicators with high loadings can describe 50% of the variance (AVE=0.50), at minimum. The internal consistency reliability and convergent validity outcomes are provided in Table 1. The discriminant validity of the reflective constructs is determined using two criteria namely the Fornell-Larcker criterion and the heterotrait-monotrait (HTMT) ratio Voorhees et al., 2016). Accordingly, Table 2 shows that the square root of the AVE on diagonal lines is higher than the correlation between the constructs in the model. What can be implied from the presented values is that all variables in this research model satisfy the requirement of the discriminant validity. Using HTMT in the discriminant validity testing, the value obtained is less than 0.85, denoting the adherence to the rule of thumb recommended (Al-Shbiel et al., 2018;Henseler et al., 2017).

Structural Model
The model fit needs to be tested using the standardized root mean square residual (SRMR) as the root mean square discrepancy between the perceived correlations and the model-implied correlations (Hu and Bentler, 1998). as well as other fit indices (i.e., normed fit index (NFI)) for testing the research's model fit (Henseler et al., 2014). This model fit test is done prior to having the model tested. The test yielded SRMR of 0.063 < 0.080 and the normed fix index (NFI) 0.913 > 0.80. What can be construed from these outcomes is that the model proposed in this work fits the empirical data (Al-Shbiel et al., 2018;Hu and Bentler, 1999). Problems of collinearity need to be examined as well. For the purpose, the values of VIF of all predictor constructs are examined. From the obtained outcomes, all values of VIF values are smaller than the established threshold of 3.3 (Kock, 2015). As mentioned in Kock (2015), the model can be regarded as having no common method bias if all VIFs obtained from a full collinearity test are equivalent to or less than 3.3. As such, among the constructs within the structural model, it appears that collinearity is not a serious problem. Hence, the results can be examined further.
The structural model was further assessed by scrutinizing the determination coefficient (R 2 or adjusted R 2 ), f2 and Q 2 which measures the model's predictive power. Determination coefficient denotes the amount of variance in the endogenous variable which is describable using all exogenous variables. Accordingly, in some disciplines, the value of higher than 0.20 for determination coefficient is deemed as high, whereas good values of determination coefficient should fall between 0.25 and 0.50.
Meanwhile, standard errors and t-statistics are produced using the technique of bootstrapping. Standard errors and t-statistics ease the assessment of statistical significance of the path coefficients. The outcomes obtained denote the significance of the conjectured relationships with the exclusion of hypothesis 2b. The use of percentile bootstrapping also led to similar results at 95% confidence interval (Henseler et al., 2016;Henseler et al., 2017). An interval without zero denotes the structural path coefficient being significantly different from zero, at a confidence level of 95% (Castro and Roldán, 2015). As such, the path coefficient can be deemed to be significant.  The obtained results are the complete affirmation of this study's projection on the effect of neuroticism personality as a predictor of job burnout and organizational commitment. As shown in Table 4, the standardized path coefficients demonstrate that neuroticism has a significant positive connexion with job burnout and a significant negative linkage with organizational commitment where: β=0.791, -0.233, p-value < 0.05 respectively. Furthermore, the results from the method of bootstrap also have no absolute zero value, which means affirmation to H1a and H1b. In addition, workload has a significant positive linkage to job burnout (H2a, β = 0.248, p < 0.05) and no relation with organizational commitment, which means support to H2a. This study also provides affirmation of the effect of job burnout, where it is found to have significant and adverse relation (-0.433) to organizational commitment (p-value of < 0.05). Furthermore, bootstrapping confidence intervals yields outcomes of -0.645 and -0.367, which means support to Hypothesis 3.
In the assessment of the model, evaluation of the impact of job burnout as a mediator between neuroticism personality, workload and organizational commitment is the final step. Here, the indirect effect is computed (ab) and for the purpose, the work of Cepeda et al. (2018), was referred. Accordingly, computation was made to the percentile bootstrap and bias corrected bootstrap. Effect of mediation also determined through the assessment of the significance of the indirect effect. The confidence intervals (Panaccio and Vandenberghe) were used in significance valuation. Here, in order to be significant, the value of 0 must not be present; Table 5 can be referred. The outcomes in the present study show significant indirect effect. This is an affirmation of job burnout acting as mediator between neuroticism, workload and organizational commitment, which denotes support to H4a and H4b.
Following the determination of indirect effect, the significance of the direct effect (c') is examined. This is to determine if the mediation is full or partial. Hence, evaluation is done on the direct effect's significance with no inclusion of the mediating variable between neuroticism, workload and organizational commitment. Table 4 presents the details. Furthermore, considering the insignificance of the direct effect of workload on organizational commitment and the significance of indirect effect through job burnout, there is full mediation relationship.
In addition, there appears significant direct and indirect impact of neuroticism on organizational commitment, and a partial mediation relationship is developed. As affirmation to the result, the Variance Accounted for (VAF) is computed. VAF determines the size of the indirect effect on the total effect. Here, according to Hair et al. (2016), if the VAF obtained is less than 20%, then it means almost zero mediation. Meanwhile, VAF higher than 20% and lower than 80% denotes a classic partial mediation whereas VAF higher than 80% denotes full mediation. Table 5 highlights the details.

Conclusion, Discussion and Recommendations
Workload and neuroticism personality appear to be crucial burnout antecedents and these consequently foretell auditors' organizational commitment. In detail, workload and neuroticism personality are positively correlated with job burnout. However, with organizational commitment, only the neuroticism personality shows significant direct correlation. Job burnout shows a significant positive correlation with organizational commitment. Job burnout is also found to fully mediate the linkage between workload and organizational commitment. On the other hand, with respect to the relationship between neuroticism and organizational commitment, job burnout appears to partially mediate. These findings are affirmation and expansion of linkage that past works have established between burnout antecedents and its outcomes.
Appositely, several limitations have been identified in this study, and these limitations could be addressed in forthcoming works. To begin with, the data obtained in this work are cross-sectional data and therefore, deduction on the causal direction of relationships cannot be drawn. It is in fact worthy to perform longitudinal studies within the duration of several months or years even, as this will allow the examination of cross-lagged relationships. The second limitation of this study is its dependency towards self-reported information particularly pertaining to traits, burnout, and organizational commitment, which causes the possible occurrence of common method bias. However, in this study, several statistical steps were taken in order to lessen common method bias. Still, longitudinal designs and multi-source data are of value as they can comprehensively rule out its effects.
Additionally, burnout appears to be the primary outcome for consideration (mediator variable). Furthermore, considering that engagement has linkage with (even though not opposite of) job burnout (Hakanen et al., 2006), the contribution of neuroticism personality to (lower) engagement level might also be beneficial.
Apart from the above suggestions, future works should consider different personal traits as predictor variables of burnout. This will enrich the understanding regarding the interrelatedness between personal resources for burnout (and engagement). Also, this study is proving the value of auditor burnout as a construct which could further be employed in other domains as well. In this regard, other possible predictors could be scrutinized. Notably, only two factors linked to burnout were used in this work and for this reason, future studies could consider including more predictors of job burnout for instance leadership behaviours.