**4. Results and discussion**

#### **4.1 PCA**

#### *4.1.1 The intrinsic determinants of job satisfaction*

From the 14 selected items, the factorial analysis highlighted a solution of five factors. The value of Kaiser – Mayer – Olkin (KMO) is 0.847, the anti-image correlation matrix, and the significance of Bartlett's sphericity test at the 1% threshold invite us to continue the analysis. The total explained variance is 74.8%, and the lowest factorial weight is 0.445. As a result, according to the scree-test and the Kaiser criterion, the PCA confirms a multidimensional structure with five factors [55, 56].

The determinant "Achievement" represents the first factorial axis which accounts for 29.2% of the total variance explained. However, "Advancement", which explains 15.7% of the total variance explained, is the second factorial axis. The variance explained by the determinant "Work itself" is 14.4% of the total variance explained. The fourth factorial axis, "Recognition", accounts for 12.8% of the total variance explained. Finally, the determinants related to "Growth" explains 12.7% of the total variance explained.

To confirm the consistency of each of these dimensions, we performed the Cronbach's alpha test. **Table 2** shows that the reliability coefficient α is slightly higher than the selected threshold (0.8799122); it is all the more acceptable because this scale is exploratory.

#### *4.1.2 Extrinsic determinants of satisfaction*

A first PCA based on the covariance matrix, on the 14 items, is launched without specifying the number of axes requested. According to the Kaiser criterion (eigenvalue greater than 1), five factors are extracted and allow to resituate 74.8% of the total variance. The results of the anti-image correlation matrix indicate that not all items are presented by two diagonally opposite points, and this indicates that they are not linked by a strong correlation, and therefore no item is duplicated. Furthermore, the factor analysis conducted on the 14 items shows that the first factorial axis is the "relationship with supervisor and colleagues", which explains 22.3% of the total variance explained. The second is the "company policy", explaining 19.4% of the total variance explained. The "working conditions" is the third factorial axis accounting for 16.1% of the total variance explained, and "Salary" is the fourth factorial axis explaining 11.9% of the total variance explained (**Table 3**).

#### *4.1.3 Job satisfaction*

To study the factor structure of job satisfaction, we used the KMO and Bartlett tests. The value of KMO which is of the order of 0.72284, the examination of the



#### **Table 2.**

*PCA of intrinsic determinants (standardized Varimax).*

anti-image matrix of correlations, and the significance of Bartlett's sphericity test at the threshold of 0.01 invite us to continue the analysis. The factorial structure obtained is satisfactory and explains 89.36% of the results. The lowest factorial score is 0.599 (**Table 4**).

To properly analyze the joint link between intrinsic and extrinsic determinants and job satisfaction, we used a multiple regression [57].

#### **4.2 Analysis and discussion of multiple regression results**

To test the validity of our nine research hypotheses, we chose modeling the job satisfaction of accounting professionals using the Fully Modified Ordinary Least Square (FMOLS) approach, which allows us to estimate a set of relationships by taking into account the problem of the autocorrelation of errors and the presence of omitted variables [57].


#### **Table 3.**

*PCA of extrinsic determinants (standardized Varimax).*


**Table 4.**

*PCA of Job satisfaction (standardized Varimax).*

The relationships between the intrinsic/extrinsic variables and job satisfaction are estimated using the following two tests:


The results of these two tests are presented in **Table 5**. As can be seen in this table, the problem of the autocorrelation of errors for all intrinsic and extrinsic variables is verified.

Indeed, the Durbin-Watson statistic shows a value of 2.31, this indicates that we will accept the hypothesis of the existence of autocorrelated errors problem. The Ramsey Reset test confirms the absence of omitted variables in our model, which implies a good prediction of the job satisfaction of the accounting professionals in our sample.

The overall significance of the model is also verified using the Fischer statistic at the 0.1% threshold.

Hypothesis (H2–4) regarding the positive and significant effect of the company policy on job satisfaction is validated (α = 0.41105; p = 0.000), which corroborates this result of [56].

Hypothesis (H2–2) regarding the positive effect of relationship with supervisor and co-workers on job satisfaction is corroborated (α = 0.18456; p = 0.084, p < 0.1). This result is in line with that of [8].

The hypothesis related to the positive effect of salary on job satisfaction (H2–1) is validated (α = 0.3841; p = 0.0011). Thus, confirming a previous result by [36].

Moreover, the result related to Hypothesis (H2.3) postulating the relationship between working conditions and job satisfaction shows a positive and significant relationship (α = 0.3382; p = 0.064, p < 0.1). This last finding is consistent with the result found by [46].


#### **Table 5.**

*Result of FMOLS estimation.*

Hypothesis (H1–1) concerning the positive and significant effect of achievement on job satisfaction is validated (α = 0.2822; p = 0.088, p < 0.1), which is consistent with the finding of [16].

The hypothesis of a positive effect of work itself on job satisfaction is also accepted (H1–3) (α = 0.6554; p = 0.000). This result seems to be justified by the fact that work itself is the most important variable in predicting job satisfaction. This result is in agreement with that of [23].

Our results show the absence of a significant relationship between promotion (H1–2), recognition (H1–4), and growth (H1–5) and job satisfaction which have a P > 0.1. In fact, a survey of accounting firms revealed how intrinsic and extrinsic motivational factors can have an impact on the accounting professionals' job satisfaction. Therefore, faced with the competitiveness requirements, personal achievement, responsibility, initiative, recognition, as well as good working conditions and salary policy are ultimately important adjustment variables for professional accountants because they can contribute to the value creation. However, the survey results showed that variables, such as promotion, recognition and growth do not have any impact on the professional accountants' job satisfaction. More specifically, professional accountants seek personal growth and development much more than the desire for recognition or promotion.

## **5. Conclusion**

The work aimed to investigate the determinants of the accounting professionals' job satisfaction in the Tunisian context. For this end, we posed two main hypotheses that deal with the relationship between the intrinsic and extrinsic factors and the accounting professionals' job satisfaction, respectively. The first hypothesis was subdivided into five sub-hypotheses, which present the intrinsic factors. The second was composed of four sub-hypotheses about the extrinsic factors. The results showed that three intrinsic factors, namely achievement, the work itself, recognition, along with four extrinsic ones, i.e., salary, relationship with supervisor and co-workers, working conditions, and the company policy, significantly affect the job satisfaction. However, we did not prove the existence of a significant relationship between growth, promotion, and recognition as well as job satisfaction. The research presented two limitations: The first was related to the small size of the sample. The second limitation concerned the imbalance in the composition of the sample and the selection of accounting professionals. Finally, new avenues of research could be explored by studying the relationship between the intrinsic and extrinsic factors, and the audit team job satisfaction.

## **A.Appendix**

#### **B.Intrinsic satisfaction factors**

#### **B.1 Achievement**


*Determinants of Job Satisfaction of Accounting Professionals in Tunisia DOI: http://dx.doi.org/10.5772/intechopen.99331*
