Executive Selection Process and Job Satisfaction: The Case of the Independent Authority for Public Revenue (IAPR) in Greece

Abstract

This paper assesses the impact that the degree of acceptance of the selection process, as conducted by the human recourses management (HRM), has upon job satisfaction in the local tax offices (LTOs) in Greece. Furthermore, it examines the effects that various individual and demographic characteristics have on job satisfaction. The research was performed using primary statistical data by distributing and collecting specific questionnaires, which not only recorded the level of job satisfaction but also noted the degree of acceptance of the selection procedure in the IAPR. The collected statistical data were analyzed using the IBM SPSS Statistics 26 Fixpack 1 software program. The results showed an average acceptance value of the selection process of 4.89, indicating a high degree of participant consensus with this process. They also showed that age, work experience, and the acceptance of the selection process have a positive and statistically significant effect, whereas educational level has a negative and statistically significant effect on degree of job satisfaction. The added value of the present paper lies in the fact that this kind of research, conducted in the IAPR in Greece for the very first time, measured both the degree of job satisfaction and the effect that the acceptance of the selection procedure has on this attitude.

1. Introduction

Nowadays, organizations seek to employ individuals that are qualified, innovative, and capable of facing competition in order to succeed in the long run (Frese and Fay 2001). However, evidence shows that an organization has a greater chance of achieving its goals when it manages to attract, retain, and motivate individuals that embrace the same values, norms, beliefs, and share the same goals and priorities (Kristof 1996). Academic research on public administration pays relatively little attention to issues concerning human resource management (HRM) and job satisfaction. Many private sector studies suggest that investing in HRM has a positive effect on worker morale (Steijn 2004). Job satisfaction is a widely researched subject. This is because it is generally believed that higher job satisfaction is associated with better organizational and individual performance, increased productivity, and lower absenteeism and employee turnover (Hackman and Oldham 1975; Iaffaldano and Muchinsky 1985). This concept is far less popular within the field of public administration (DeSantis and Durst 1996; Aristovnik et al. 2018; Cantarelli et al. 2015; Singh and Slack 2016) especially in Greece. The issues of HRM and job satisfaction are linked because many researchers assume that better human resource (HR) practices result in higher levels of job satisfaction and commitment, and hence, an improved organizational performance (Appelbaum et al. 2000). According to Armstrong (2010), commitment, motivation, and job satisfaction, either jointly or separately, will be higher when people positively experience the application of HR policies concerned with creating an able workforce, motivating valued behaviors, and providing opportunities to participate.

Job satisfaction is the most widely investigated job attitude and one of the most extensively researched subjects in the HRM and organizational behavior fields (Judge and Church 2000). Job satisfaction can be described as “the extent to which people like (satisfaction) or dislike (dissatisfaction) their jobs” (Spector 1997). The most widely accepted definition of job satisfaction was presented by Locke (1976), who explained job satisfaction as “the pleasurable emotional state resulting from the appraisal of one’s job as achieving or facilitating the achievement of one’s job values”. It represents a combination of positive or negative feelings that workers have for their work. It is influenced by a series of factors including the nature of work, salary, advancement opportunities, management, work groups, work conditions, and the congruence between the person and the organization. It is considered as one of the main factors when it comes to efficiency and effectiveness of business organizations. Job satisfaction is an important predictor of employee’s intention to stay or leave the organization. If the employee is dissatisfied with his work, he is likely to leave the organization. Further, a lack of job satisfaction may cause a low level of productivity and a high level of absenteeism, occupational accidents, mental and physical health problems, and early retirement. The HRM and organizational behavior literature provide a plethora of empirical evidence showing that job satisfaction is significantly negatively related to intentions to leave an organization (Chen and Spector 1992) and turnover intentions (Rusbult and Farrell 1983; Coomber and Barriball 2007).

The process is intended to identify prospective employees who will fit well with the hiring organization. It is a complex and expensive process. Most often, an organization’s selection process emphasizes the prospective employees’ “hard” technical skills but focuses very little on their “soft” behavioral skills. However, the soft skills are crucial to the effectiveness of quality management practices. Moreover, while employees can generally be quickly trained in tools and techniques, developing soft skills can take a long time. In some instances, depending on the employee’s personality traits, the development of soft skills may not be possible. Therefore, employers need to pay close attention to the behavioral traits of prospective employees during the selection process (Ahmad and Schroeder 2002).

The present paper is organized in five main parts. Initially, the literature on job satisfaction and relevant responses are reviewed in order to formulate the hypotheses under investigation. Then, a brief overview of the Greek IAPR is provided to describe the context in which this research took place. After that, the research methodology presents the sample composition, the procedures followed, and the instruments adopted. Finally, in the results section, the hypotheses are tested and the discussion section offers insights into the theoretical and practical implications of the findings and recognizes the limitations of the study.

The present study was conducted nationwide in Greece among chief managers employed in the public sector, and specifically in local tax offices. The examination of job satisfaction as a result of the acceptance of the selection procedure, in this case, was considered to be particularly important, as some of the characteristics of the Greek public sector—including increased bureaucracy, high centralization, political interference, and institution tenure—create a rather unique setting (Bellou 2009; Spanou 2019; Akrivos et al. 2013; Makrydemetres et al. 2016). This study hopes to demonstrate the positive role of the acceptance of the selection process as related to the degree of job satisfaction of senior managers at the Independent Authority for Public Revenue (IAPR) in Greece. Moreover, given the fact that some demographic and individual characteristics have been proved as factors that affect job satisfaction, the present study examines the effect of these control variables on this particular attitude.

2. Background and Hypotheses Development

According to some previous studies, sex affects job satisfaction. However, the existing bibliography comes to controversial conclusions. Various writers came to the conclusion that females are more satisfied than their male coworkers (Clark 1997; Sloane and Williams 2000; Blanchflower and Oswald 1999; Bartol and Wortman 1975; Clark and Oswald 1996; Koustelios 2001a, 2001b). On the other hand, others concluded that males are generally more satisfied with their jobs (Brief et al. 1977; Coh and Koh 1991; Lumpkin and Tudor 1990; Chiu 1998). Moreover, Varshney and Malpani (2014) and Rana (2015) assumed that sex makes no difference to the level of job satisfaction. Additionally, Bellou (2010) concluded that males and females draw satisfaction from their labor in a different way.

Taking into account the aforementioned publications, the first hypothesis is as follows:

Hypothesis 1 (H1).

Sex has a statistically significant effect on the degree of job satisfaction of the executives in the IAPR.

Some researches demonstrate a negative correlation between pay and job satisfaction (Clark and Oswald 1996), while others, however, establish a positive correlation (Sloane and Williams 2000; Aspioti 2013).

Spector (1997) argues that remuneration has a positive but not significant correlation with job satisfaction. Van Praag and Ferrer-i-Carbonell (2004) conducted a study and found a positive relationship between household income and job satisfaction.

Given the above-mentioned remarks, the second, third, and fourth hypotheses are as follows:

Hypothesis 2 (H2).

Wage has a positive and statistically significant effect on the degree of job satisfaction of the executives in the IAPR.

Hypothesis 3 (H3).

Annual individual income has a positive and statistically significant effect on the job satisfaction of the executives in the IAPR.

Hypothesis 4 (H4).

Annual household income has a positive and statistically significant effect on the job satisfaction level of the executives in the IAPR.

In existing scientific literature, it is generally believed that job satisfaction increases in a linear fashion with age (Mottaz 1987; Hulin and Smith 1965; Near et al. 1978; Weaver 1978).

However, there are convincing arguments, and some empirical evidence, that the relationship is U-shaped, declining from a moderate level in the early years of employment and then increasing steadily up to retirement (Clark et al. 1996; Clark 1996).

Baruch (2004) found that older workers are less motivated by competition than their younger colleagues and this factor leads them to enjoy the workplace environment and to be more satisfied. Additionally, older workers experience lower stress levels than middle-age workers according to Hertel (2013).

Thus, the fifth hypothesis is put forward for debate:

Hypothesis 5 (H5).

Age has a positive and statistically significant effect on the degree of job satisfaction of the executives in the IAPR.

The relationship between job satisfaction and work experience is positive and statistically significant according to Batiou and Valkanos (2013), Oshagbemi (2000), and Bedeian et al. (1992).

Workers who have a longer service history usually have a better chance of promotion, which could lead to them being more satisfied (Bedeian et al. 1992; Lee and Wilbur 1985).

However, according to a study that dealt with a representative sample of librarians in Greece, job satisfaction levels decline as work experience increases (Togia et al. 2004).

Given the aforementioned remarks, the sixth hypothesis is as follows:

Hypothesis 6 (H6).

Work experience has a positive and statistically significant effect on the job satisfaction level of the executives in the IAPR.

This research confirms the results of numerous surveys showing that the more educated employees are, the higher their expectations in the workplace; thus causing low satisfaction whenever these expectations are not fulfilled (Clark 1996; Clark and Oswald 1996; Sloane and Williams 2000; Jung 2013; McGrandle 2019). The educational level is also important for the organization itself, as this research shows that the more educated employees have lower job satisfaction. We reasoned that educational qualifications that do not lead to extrinsic rewards would lead to job dissatisfaction due to unfulfilled aspirations.

In addition, according to equity theory (Adams 1965), individuals compare their job inputs and outcomes with those of others and then respond in a way to eliminate any perceived inequities. If, in the same workplace, the more educated employees receive the same compensation package as those who are less educated, the former will report declining satisfaction levels.

Thus, the following hypothesis was investigated:

Hypothesis 7 (H7).

The education level has a negative and statistically significant effect on the degree of job satisfaction of the executives in the IAPR.

Many researchers demonstrated a positive correlation between the practices of HRM and job satisfaction (Appelbaum et al. 2000; Armstrong 2010; Ashton and Sung 2002; Delaney and Huselid 1996; Pfeffer 1998).

Moreover, numerous investigators demonstrated the positive impact of the acceptance of the selection process on job satisfaction (Steijn 2004; Gould-Williams 2003; Rehman 2012; Ijigu 2015; Bretz and Judge 1994; Liu et al. 2010; Bright 2008; Mostafa and Gould-Williams 2014; Alniacik et al. 2013; Kim 2012; Bellou 2009).

Thus, the following hypothesis was investigated:

Hypothesis 8 (H8).

The acceptance of the selection process of the top executives in the IAPR has a positive and statistically significant effect on their job satisfaction.

3. A Brief Overview of the IAPR

The Independent Authority for Public Revenue (IAPR) was established by Law 4389/2016 in order to confirm the tax payers’ liability and collect public revenue. It started operating as an independent authority on 1 January 2017 replacing the General Secretariat of Public Revenue, which was then accountable to the Finance Ministry. Since then, it enjoys administrative, operational, and financial independence and is accountable only to parliamentary scrutiny.

The internal structure and organization of the IAPR, the responsibilities, the distribution of the personnel in specialized subjects, the qualifications for their appointment, and the general management of the human resources were established by the Government Gazette No 968/B/2017.

The governing bodies of the IAPR are the governor and a five-member management board.

The authority’s framework of activities, in accordance with the “Directorate for Strategic Planning” of the IAPR, includes the following responsibilities:

• The determination, assessment, and collection of taxes, customs, and other public revenue.
• The fight against tax evasion, smuggling, tax fraud, and the shadow economy.
• The protection of public health, the environment, and consumers’ interests.
• Strategic and operational planning and training concerning the implementation of the targets and setting performance indicators.
• The provision and technical support of digital services to individual citizens, businesses, and public sector entities.

According to the IAPR’s organization, the authority consists of the following services:

• Units reporting directly to the governor.
• The general directorates.
• The directorates.
• The special decentralized services and regional services.

The local tax offices (LTO) are regional tax agencies of the general directorate of tax administration and are structured as follows:

• Director of the LTO.
• Deputy director of the LTO.
• Tax audit section.
• Enforcement section.
• Tax collection section.
• Tax compliance section (desks of income tax, VAT, motor vehicles, and real estate transactions).
• Administrative support section (registry offices, secretariat, technical administrators).
• Tax cashier office.
• Deposits and loans office.

There are 13,322 regular staff positions in the IAPR and they are distributed as follows:

Tax officials: 9103; custom officials: 2443; inspectors: 136; chemical agents: 333; technical administrators: 225; ushers: 240; and other employees: 842.

Furthermore, 63% of the staff are women and 37% are men. The figures show that 67% of the workforce are in the 46–65 age group, and only 7% are under 35 years of age. The educational level of employees is particularly high, with 44% having a university degree, while 25% hold a postgraduate or a doctorate degree.

Additionally, the strategic plan of the IAPR was drafted in 2017 and concerns a medium-term goals of four years. The vision of the IAPR is to establish a trusting relationship with the taxpayers and develop into a model body to safeguard public revenues in the country. The IAPR’s mission is to secure public revenues by strengthening tax compliance and combating tax evasion and smuggling, while providing high-quality services to citizens and businesses.

4. Research Methodology

The research included all the heads of the organizational units at the level of management and deputy management of the LTOs in Greece. In order to measure their level of job satisfaction, we used the Spector questionnaire, which is generally used in such kind of research. The most widespread reliability measurement is the internal consistency reliability, which is the degree of correlation between every question on the scale and the other questions. The most common test score reliability coefficient is Cronbach’s alpha. Values over 0.7 were deemed satisfactory (Spector 1992; Nunnally 1978; Fornell and Larcker 1981). The reliability of this questionnaire was measured accordingly and was found to be highly reliable overall (Cronbach’s alpha = 0.958).

During the research period (July–August 2020), 108 LTOs were active with 169 top executives. Out of the 169 questionnaires sent, 132 were filled out and collected ready for processing. The response rate was thus 78%. The synthesis of the sample, in terms of gender, age, education, and tenure, is as follows:

• 56.8% of the sample consisted of men and 43.2% women;
• 15.9% were 36–45, 26.5% were 46–55, 52.3% were 56–65, and 5.3% were over 65 years old;
• 57.6% were graduates from a higher school, 35.6% possessed a postgraduate qualification, and 6.8% held a PhD or graduated from the National School of Public Administration;
• 39% had a seniority from 20 to 27 years, 40% from 28 to 35 years, and 21% from 36 to 37 years.

The questionnaire consisted of three parts:

In the first part, the participant was asked to fill in information on his/her personal data, such as gender, age, previous service, marital status, level of salary and annual individual and family income, level of education, and administrative position (manager or assistant manager).

In the second part, he/she was called upon to answer 36 questions concerning job satisfaction. For this purpose, in the present study, the Spector (1985) questionnaire was used, which measures nine factors in job satisfaction with four questions—variables for each factor. These factors are as follows: (1) pay; (2) promotion; (3) supervision; (4) fringe benefits; (5) contingent rewards; (6) operating conditions; (7) coworkers; (8) nature of work; and (9) communication. The questionnaire consisted of a total of 36 closed-ended questions and answers on a six-point Likert scale (totally disagree, moderately disagree, disagree a little, agree a little, agree moderately, and strongly agree).

In the third part, the degree of acceptance of the selection process for the positions of director or deputy director was assessed. To this end, eight closed-ended questions were created on a six-point Likert scale (totally disagree, moderately disagree, disagree a little, agree a little, agree moderately, and strongly agree). These questions related to the individual formal and substantive qualifications the candidate required in order to apply for the advertised position and his/her attitude towards the overall procedure; they are listed as follows: (1) work experience; (2) level of education; (3) previous service; (4) performance evaluation reports; (5) work experience in the private sector; (6) structured interview; (7) written test; and (8) sense of meritocracy of the process.

In the present study, the regression analysis and the principal component analysis (PCA) were used as statistical methods. The PCA is the most common method used by researchers. PCA starts extracting the maximum variance and puts them into the first factor. After that, it removes the variance explained by the first factors and then starts extracting the maximum variance for the second factor; this process continues to the last factor. These methods are appropriate tools for studying the relationships between variables. In general, factor analysis is suitable for measuring human behavior and is widely used by social science researchers (Dunteman 1989; Bartholomew et al. 2002; Fowler 2014; Bandalos 2018).

In this article, the dependent variable is the degree of job satisfaction. In terms of job satisfaction, a six-point Likert scale was used. The total result (score) of the job satisfaction variable varied between 36 and 216.

The independent variables, for editing using the SPSS program, are listed and marked as follows:

• Sex. Man: 1; Woman: 2;
• Age. 36–45: 1; 46–55: 2; 56–65: 3; >65: 4;
• Education level. Higher education: 1; MSc: 2; PhD: 3;
• Foreign language. No: 1; Yes: 2;
• Marital status. Single, widowed, divorced: 0; Married: 1;
• Number of children. Natural number;
• Monthly salary. Natural number;
• Annual individual income. 15,000–20,000 = 1; 20,001–25,000 = 2; 25,001–30,000 = 3; 30,001–35,000 = 4; 35,001–40,000 = 5; 40,001–45,000 = 6; 45,001–50,000 = 7; >50,000 = 8;
• Annual family income. 15,000–20,000 = 1; 20,001–25,000 = 2; 25,001–30,000 = 3; 30,001–35,000 = 4; 35,001–40,000 = 5; 40,001–45,000 = 6; 45,001–50,000 = 7; 50,00–55,000 = 8; 55,001–60,000 = 9; 60,001–65,000 = 10; 65,001–70,000 = 11; 70,001–75,000 = 12; 75,001–80,000 = 13; >80,000 = 14;
• Administrative position. Manager: 1; Deputy Manager: 2;
• Years of service. Natural number;
• Acceptance of the selection process. Strongly disagree: 1; disagree moderately: 2; disagree slightly: 3; agree slightly: 4; agree moderately: 5; strongly agree: 6.

The total score of the variable regarding the acceptance of the selection process varied between 8 and 48.

Notwithstanding, no research, including this one, is free of limitations. The sample used references to senior executives in the IAPR (directors and deputy directors) who had already been selected and occupied these positions. Therefore, the participants who were not selected during the process were not included. Thus, a degree of bias should be implied, though it was not measured in the present study. In addition, as can be seen from the analysis of the statistical data, the results obtained depend, to a large extent, on the independent variables selected. A study that did not include the selection process of the research model under consideration would make the effect of different variables statistically significant as regards the degree of job satisfaction.

5. Results

From the elaboration of the questionnaires regarding job satisfaction, the following overall and gender-specific results emerged (

Table 1. Mean, median, standard deviation, minimum, maximum, and range of J.S.

Mean	Median	SD	MIN	MAX	Range
3.80	3.79	0.36	2.50	5.14	2.64
Men			Women
3.77			3.84

).

Means according to job satisfaction factor are shown in

Table 2. Mean according to J.S factor.

Factors	In Total	Men	Women
Pay	3.38	3.35	3.43
Promotion	4.03	4.02	4.05
Supervision	4.86	4.90	4.80
Fringe benefits	3.14	3.06	3.24
Contingent rewards	3.44	3.40	3.50
Operating conditions	2.60	2.50	2.73
Coworkers	3.59	3.55	3.64
Nature of work	4.87	4.89	4.85
Communication	4.31	4.30	4.32

below.

From the processing of questionnaires on the acceptance of the selection procedure, the following overall and gender-specific results emerged (

Table 3. Mean, median, standard deviation, minimum, maximum, and range of the selection process.

Mean	Median	SD	MIN	MAX	Range
4.89	5.00	0.82	3.25	6.00	2.75
Men			Women
4.89			4.89

).

The means regarding the acceptance of the recruitment procedure were as follows: (a) work experience = 5.10; (b) level of education = 5.20; (c) previous service = 4.03; (d) performance evaluation reports = 5.30; (e) work experience in the private sector = 3.80; (f) structured interview = 5.40; (g) written test = 5.20; (h) sense of mediocracy of the process = 4.90.

The reliability of the questionnaire on job satisfaction was found to be very high overall (Cronbach’s alpha: 0.958); the reliability of the questionnaire regarding the acceptance of the selection process was also found to be very high (Cronbach’s alpha: 0.912).

A check was made to establish the existence of a correlation between the variables. Pearson correlation coefficients were used to calculate the intensity of the relationship (

Table 4. Pearson correlations.

Pearson Correlations
	1	2	3	4	5	6	7	8	9	10	11	12	13
SEX	1
AGE	ns	1
EDUC.LEVEL	ns	−0.60 **	1
F.LANGUAGE	ns	−0.59 **	0.56 **	1
MAR.STATUS	ns	ns	ns	ns	1
CHILDREN	ns	ns	ns	ns	0.38 **	1
SALARY	ns	0.52 **	−0.26 **	−0.37 **	ns	0.18 *	1
IND. INCOME	ns	ns	ns	−0.19 *	ns	ns	0.33 **	1
FAM.INCOME	ns	ns	ns	ns	0.60 **	0.27 **	0.20 *	0.32 **	1
POSITION	ns	−0.22 *	ns	ns	ns	ns	−0.30 **	ns	ns	1
SERVICE	ns	0.73 **	−0.50 **	−0.56 **	ns	ns	0.51 **	0.19 *	ns	−0.30 **	1
SELECTION	ns	0.72 **	−0.54 **	−0.60 **	ns	ns	0.48 **	ns	ns	ns	0.73 **	1
J.S	ns	0.64 **	−0.52 **	−0.54 **	ns	ns	0.46 **	ns	ns	−0.18 *	0.74 **	0.87 **	1

Notes: * p < 0.05; ** p < 0.01. 1. Sex; 2. Age; 3. Educational level; 4. Foreign language; 5. Marital status; 6. Number of children; 7. Salary level; 8. Individual income; 9. Family income; 10. Hierarchy position; 11. Previous service; 12. Acceptance of selection process; 13. Job satisfaction level.

).

In order to eliminate multicollinearity problems, we applied principal component analysis with varimax rotation in order to group the independent variables into components. According to Hair et al. (1998), during the analysis process, we sequentially deleted the variables which: (a) had a sample adequacy measure value less than 0.5; (b) had loadings in a component of less than 0.5; (c) had loads greater than 0.45 in more than one component; and (d) created components on their own.

By following this process, we came up with a choice of five variables, which formed one component. These variables were: (a) age; (b) level of education; (c) foreign-language knowledge; (d) previous service; and (e) acceptance of the selection process (KMO measure of sampling adequacy: 0.865 and significance of Bartlett’s test of sphericity: 0.000 < 0.001). The communalities, i.e., the variations of the variables interpreted by the component that were greater than 0.50, were as follows: age: 0.78; foreign-language knowledge: 0.63; education level: 0.58; previous service: 0.73; and acceptance of selection process: 0.76. Additionally, the created component interpreted about 69% of the variance of the independent variables. Furthermore, the loadings of the independent variables on the created component that were greater than 0.75 were as follows: age: 0.88; foreign-language knowledge: 0.79; education level: 0.76; previous service: 0.85; and acceptance of selection process: 0.87. We named the component that we created “A_E_T_S” (Age_Education_Tenure_Selection) so that we could then use it in the regression analysis.

A regression analysis of the component (A_E_T_S) on the degree of job satisfaction was conducted in order to test the hypotheses (

Table 5. Regression analysis of the component on J.S.

Model Summary b
Model		R		R Square	Adjusted R Square	Std. Error of the Estimate	Durbin-Watson
1		0.802 a		0.643	0.640	12.98702	1.561
a Predictors: (Constant), REGR factor score, b Dependent Variable: J.S
Coefficients a
Model		Unstandardized Coefficients		Standardized Coefficients	t	Sig.	Collinearity Statistics
		B	Std. Error	Beta			Tolerance	VIF
1	(Constant)	136.894	1.130		121.105	0.000
	REGR factor score	17.366	1.135	0.802	15.305	0.000	1.000	1.000
a Dependent Variable: J.S

). We use regression analysis and principal component analysis because it is a widely accepted statistical method that allows us to examine the relationship between two or more variables of our interest.All types of regression analysis substantially examine the influence of one or more independent variables on a dependent one.

Table 5 shows the summary of our model. The total interpreted correlation amounted to 80.2% (correlation coefficient: 0.802) and the total interpreted variance to 64.3% (coefficient of determination: 0.643). The variable that we created with the PCA analysis explains 64% of the variance of the dependent variable. Furthermore, it assessed the autocorrelation of the residuals with the Durbin–Watson indicator and the value was found to be 1.561, which is acceptable. An analysis of variance was conducted (ANOVA). The significance of the value of “F” was checked and it was found that the null hypothesis can be rejected (F = 234.245 > F_1;130;0.01 = 6.86). On the basis of the results of the Table 5, our model has the following form:

J.S degree = 136.894 + 17.366 * A_E_T_S.

Coefficients (constant and component) were statistically significant (sig < 0.001) and the multicollinearity problems were deleted. Additionally, the residuals were scattered around the line of zero in the relevant scatterplot, their dispersion did not create patterns (heteroscedasticity does not occur), and their successive values were independent (autocorrelation index: Durbin–Watson = 1.561).

6. Discussion and Conclusions

The purpose of this study was to investigate the relationship between job satisfaction and the degree of acceptance of the selection process by the senior executives of the LTOs in the IAPR. At the same time, the degree of effect on job satisfaction of the economic, individual, and demographic factors that characterize the individuals of the sample was examined. For this purpose, 132 questionnaires were collected from a total population of 169 senior managers. The average job satisfaction was measured to be 3.80. According to Spector (1985), this value is an ambiguous and uncertain result. Values above four indicate job satisfaction and values below three indicate discontent. However, a more careful examination of the average values of job satisfaction per factor of the questionnaire shows satisfactory results for both sexes in the following areas: (a) promotion: 4.03; (b) supervision: 4.83; (c) nature of work: 4.87; and (d) communication: 4.31; ambiguous and uncertain results were found in the following sectors: (a) pay: 3.38; (b) fringe benefits: 3.14; (c) contingent rewards: 3.44; and (d) coworkers: 3.59; and negative results (discontent) were recorded in the field of operating conditions: 2.60.

The survey showed an average acceptance value of the selection process of 4.89, which is evidence of a high degree of employee consensus with this process.

Moreover, the average value of total satisfaction, which was measured to be 3.80, was in line with the results obtained in similar surveys in the public sector internationally, e.g., Steijn (2004): 3.56; Ting (1997): 3.83; Rehman (2012): 3.71; Bretz and Judge (1994): 4.48.

In surveys recorded on the Spector website1, several job satisfaction rates were recorded: (a) U.S. police officers: 3.71 (sample of 642 people); (b) social workers: 3.97 (sample 6505 persons); (c) to teachers 3.75 (in a sample of 9507 persons); (d) to a sample of 22,631 civil servants: 3.83.

Table 6. Hypotheses testing.

Hypotheses	Correlation	R2	Verification
H1: sex	0.045	0.003	rejected
H2: pay	0.626 **	0.214	rejected
H3: individual Income	−0.007	0.000	rejected
H4: family income	−0.010	0.000	rejected
H5: age	0.657 **	0.412	confirmed
H6: years of service	0.781 **	0.549	confirmed
H7: education	−0.566 **	−0.268	confirmed
H8: acceptance of selection process	0.845 **	0.761	confirmed

Notes: ** p < 0.01.

summarizes the research results obtained from the analysis of statistical data on factors affecting statistically significant employment satisfaction.

The statistical analysis demonstrated that the degree of acceptance of the selection process has a positive and statistically significant effect on the degree of overall job satisfaction. The process of selecting appropriate staff by applying modern HRM practices leads to the coupling of the goals and values of the individual with the requirements and culture of the organization and has a positive and statistically significant effect on job satisfaction. Public sector entities should take due account of the process of selecting appropriate staff to promote satisfaction in the working environment. Public organizations, in order to find those with appropriate attitudes, should seek, through the process of attracting, selecting, and hiring, the right people, who adapt as much as possible to the values, rules, characteristics, and goals of the organization. These findings are consistent with those of Gould-Williams (2003), Rehman (2012), Liu et al. (2010), Bretz and Judge (1994), Bright (2008), Mostafa and Gould-Williams (2014), Alniacik et al. (2013), Kim (2012), Ijigu (2015), and Bellou (2009).

Satisfaction was measured at the same level in both sexes, confirming the conclusions of García-Bernal et al. (2005), Varshney and Malpani (2014), and Rana (2015). Monthly salaries had no statistically significant effect on the level of work satisfaction, confirming the findings of Adams (1965), Pouliakas and Theodossiou (2005), Clark and Oswald (1994), and Spector (1997). Similarly, annual individual and annual family income had no statistically significant effect on the degree of job satisfaction, in contrast to the findings of Van Praag and Ferrer-i-Carbonell (2004).

Positive and statistically significant dependence on job satisfaction was measured against age. Previous research demonstrated that older workers are more likely to have already proven their worth and now enjoy the benefits of their work (Mottaz 1987; Baruch 2004; Hulin and Smith 1965; Near et al. 1978; Weaver 1978; Clark et al. 1996; Clark 1996).

Regarding the dependence of job satisfaction on the variable that counts the years of service, a positive and statistically significant effect was found. Previous studies concluded that satisfied employees remain in the same job for many years. Greater seniority is also linked to higher promotion opportunities, which has a positive effect on the satisfaction of individuals (Bedeian et al. 1992; Batiou and Valkanos 2013; Oshagbemi 2000; Lee and Wilbur 1985).

A negative and statistically significant effect was measured regarding the dependence between the level of education and the degree of job satisfaction. Both the possession of postgraduate degrees and proven knowledge of foreign languages align with statistically significant variations in job satisfaction (Clark 1996; Clark and Oswald 1996; Sloane and Williams 2000; Jung 2013). The more educated employees ought to be higher paid and receive more personalized awards in order to shift this negative relationship into a positive correlation.

The contribution of the present work is that, for the first time, a survey was conducted on senior executives in the public sector, taking into account both individual and demographic factors that affect job satisfaction and factors that shape the selection process of these executives. This work enriches the existing literature regarding job satisfaction in the public sector by providing data related to the Greek public administration. The results of the present research can be utilized to make comparisons with data that will emerge from similar surveys in Greece and abroad. Furthermore, efficient planning, based on the specific job description of the posts to be filled, is necessary to consolidate trust between public administration and public sector employees. Recruiting executives who share the same values and are ready to work towards the goals set by any public organization is of highest priority to the administration.

The present paper is a useful source of information concerning the impact of HRM practices on the attitudes of employees in the public sector in general but also in the services of the Ministry of Finance and the IAPR.

As previous research showed, the application of modern HRM practices improves the degree of job satisfaction, organizational commitment, work efficiency, and the predisposition to stay at work for many years. In the future, it would be useful to conduct a similar survey that includes section heads. Moreover, similar research could be carried out concerning other branches of the IAPR, such as customs and chemical state laboratory officials. Furthermore, the enrichment of the factors that affect the degree of job satisfaction with other HRM practices, such as training, development, evaluation, policy, and employee remuneration systems, would be useful in order to measure the overall dependence of satisfaction on these practices.