ERG theory of Alderfer (1969) - Literature review on job satisfaction

Chapter I INTRODUCTION

CHAPTER 2 LITERATURE REVIEW

2.2 Literature review on job satisfaction

2.3.2 ERG theory of Alderfer (1969)

The ERG theory of Alderfer (1969) is somewhat similar to Maslow’s hierarchy of needs. Neverthelss, the ERG theory only consists of 3 needs including (1) existence needs, (2) relatedness needs and (3) growth needs. The existence needs include

Effort Action Rewards Target

physiological needs and safety needs in Maslow’s hierarchy of needs, and they are essential and secondary needs supporting the existence of a person. Relatedness needs are social needs and part of esteem needs proposed by Maslow. The growth needs include some parts of esteem needs and self-actualization needs. According to Alderfer (1969), these are human needs to develop themselves.

In the ERG theory, Alderfer (1969) suggested that needs may appear in a certain period of time and if a need could not be satisfied, it could be made up by other needs.

On the contrary, Maslow’s theory indicates that there would be only one need appearing in a certain period of time, and if a need is satisfied, the next level of need would appear.

The ERG model is shown in Figure 2.4

Rewards Rewards Effort

Effort

Action Action

Satisfy/reinforce Disappointment/ reverse

Satisfy/Enhance

Figure 2.4 ERG model 2.3.3 Equity theory of Adams (1963)

Adams (1963) suggested that workers tend to evaluate fairness through comparison between the efforts they make and what they receive in return, as well as among coworker of the company. If workers perceive that the comparison results are fair enough, they will maintain their efforts and working efficiency. If what they receive are

more than expectation, they would tend to increase their efforts in their jobs. On the contrary, if the reward is lower than expectation, they will reduce their efforts and find reasons to avoid their works. According to this theory, fairness is essential to the close relationship among employees as well as creating encouragement and enforcement for higher working efficiency.

The equity theory suggests that managers should pay attention to factors influencing perception of workers on fairness so that they can come up with appropriate solutions to create job satisfaction for workers.

2.2.3 Studies on job satisfaction

Spector (1985) conducted a study on job satisfaction in service industry using a scale of satisfaction consisting of 9 factors including (1) communication, (2) operating conditions, (3) promotion, (4) fringe benefits, (5) coworkers, (6) supervision, (7) nature of work, (8) contingent rewards, (9) pay

Smith, Kendall & Hulin (1969) built Job Descriptive Index (JDI) to assess job satisfaction of individuals through work itself, pay, opportunity for promotion, coworkers and supervision. Weiss et al. (1967) constructed a job satisfaction index through their Minnesota Satisfaction Questionnaire (MSQ) with questions about the possibility of using individual ability, achievement, advancement, authority, company policy, company treatment, colleagues, creativity, independence, ethical values, recognition, responsibility, assurance, social status, supervision, working condition, etc.

It can be said that JDI and MSQ are widely used in researches to evaluate employees’

satisfaction.

According to a study in the health service industry of Luddy (2005), job satisfaction was influenced by 5 factors in JDI model. Among them, three factors “work itself”, “supervision” and “coworkers” were positively rated by employees while two factors “pay” and “training and promotion” were negatively rated by employees.

In Vietnam, Tran (2005) conducted a research using the JDI model to assess job

satisfaction. Along with the factors from JDI model, Tran (2005) also added two more factors, namely “working condition” and “company benefits”, in the scale. The primary target of this research was to test JDI scale and to identify how these factors influence job satisfaction of employees in Vietnam. Using a 7-point Likert scale, Tran (2005) implemented the exploratory factor analysis (EFA) and the confirmatory factor analysis (CFA) for the study. Respondents of the research were employees of the contiuning education program in University of Economics Ho Chi Minh City. The result of the study showed that “opportunity for promotion” and “work itself” were ranked highest of all factors by the participants.

In a research in Long An Province conducted by Nguyen (2008), the author proposed 6 factors influencing job satisfaction including (1) nature of work, (2) pay, (3) coworkers, (4) leaders, (5) opportunity for training and promotion and (6) work environment.

Pham (2012) suggested that job satisfaction is affected by 4 factors including: (1) training and promotion, (2) colleagues and benefit, (3) working environment and (4) initiative. Among these factors, working environment was rated negatively by respondents.

It can be said that there are many definitions and explanations for job satisfaction.

Nevertheless, they generally suggested that job satisfaction refers to a comfortable feeling of employees toward their jobs. Researches also showed that JDI model have been widely used by research to measure the job satisfaction of people in different regions and countries. It can be said that JDI model in general can help researchers to explain the job satisfaction of employees in various industries and countries.

CHAPTER 3 RESEARCH METHODOLOGY

In this chapter, the author proposed the research methods used in the thesis. There are two phases of the research process: preliminary study and official study. The preliminary study is to test and construct a complete questionnaire for the quantitative research in the official study. This chapter also introduces the analytical methods to be used in the next chapter.

3.1. RESEARCH DESIGN

3.1.1. Data source

In this research, the author uses the primary data collected from the interviews in the preliminary research to construct the questionnaire used for the quantitative research.

The design of the questionnaire also comes from the literature review. In the official research, the author also collects the primary data from the survey to analyze the job satisfaction of the public servants in the government offices in the 5th district of Ho Chi Minh City.

3.1.2. Research process

There are two stages in this study

Stage 1: Qualitative research to build the questionnaire used for the survey in the quantitative research

Stage 2: Quantitative research to examine the reliability of the questionnaire and analyzes the factors influencing the job satisfaction of the public servants.

Figure 3.1 Research Process

Qualitative Research

Literature Review Proposed Scale Interviews

Official Scale Pilot Test Adjusting

Official Study

Cronbach’s Alpha

EFA

Quantitative Research

Official Model

Regression Analysis

3.1.3. Qualitative Research

The qualitative research is conducted using the focus group technique. The aim of this research stage is to explore potential factors that influence the job satisfaction and to construct a questionnaire for the quantitative research stage. The author designs the open-ended questions in the interviews with 7 senior administrative staff and 3 staff in the People’s Committee of the 5th District.

The questions used in the interview are listed as the following:

Question 1: Are you working in a job matching your expertise? Please give some explanations for your answer.

Question 2: Do you understand the characteristics of your job? Please give some explanations for your answer.

Question 3: Do you always try to complete your job on time? Please give some explanations for your answer.

Question 4: Do you want to do the current job? Please give some explanations for your answer.

Question 5: Do you like your job? Please give some explanations for your answer.

After the interviews, the author identified a list of factors that may have influence on the job satisfaction. These factors are: salary and benefits, promotion opportunities, training activities. These factors also have been mentioned in previous studies on job satisfaction. Thus, the author decides to use these factors in the research model of the study. The design of the questionnaire is also based on these factors. The research model is shown in Figure 3.2.

Figure 3.2 Research Model

Based on this model, the author gave 3 hypotheses:

Hypothesis 1: Salary and Benefits have positive relationship with job satisfaction.

Hypothesis 2: Promotion has positive relationship with job satisfaction.

Hypothesis 3: Training has positive relationship with job satisfaction.

3.1.4 Questionnaire Design

After the interviews, the questionnaire consists of two parts:

Part 1: Demographic information

Part 2: Surveying job satisfaction of public servants in the government administrative offices of 5th District of Ho Chi Minh City

According to Hoang & Chu (2008), there are four types of scale: (1) nominal scale, (2) ordinal scale, (3) interval scale and (4) ratio scale. In this study, the author only used nominal scale and interval scale for the questionnaire. For the demographic information, the author used the nominal scale to categorize and explore the differences between groups of the respondents (age, gender, and educational background, etc.). For part 2, the author uses a 5-point Likert scale (with 1 – Totally Disagree, 2 – Disagree, 3 – Neutral, 4 – Agree, 5 – Totally Agree). Likert scale is a type of interval scale, by which it is possible to collect and analyze the collected data to identify the correlation, and the

Salary & Benefits

regression relationship between independent variables and dependent variable.

3.1.5 Sampling Design

According to Hair et al. (1998), it is necessary to have 5 samples for each observed variable and the total sample should not be less than 100. As there are 22 variables in the questionnaire, 110 samples are enough for the study. The author chose the convenient sampling method due to the limited time and resources of the author when doing the research. 140 questionnaires are delivered to participants in the administrative offices of the 5th District of Ho Chi Minh City. Afterwards, there are 10 questionnaires removed because they are not answered properly, and the remaining 130 questionnaires are accepted for the analysis.

3.2. QUANTITATIVE RESEARCH

3.2.1 Reliability test by Cronbach’s Alpha

The purpose of this test is to determine which question really contributes to assess the concepts in the study. The variables that have the item-total correlation less than 0.3 will be removed and only variables with the item-total correlation higher than 0.65 are chosen. Though there are researchers who require the correlation higher than 0.8, however, if the concepts are new to respondents, it is acceptable to have Cronbach’s Alpha value higher than 0.6 (Nunnally, 1978; Peterson, 1994; Slater, 1995).

The formula of the Cronbach αis α = Nρ/[1 + ρ(N – 1)].

3.2.2 Exploratory Factor Analysis (EFA)

After running Cronbach’s Alpha test, the EFA test is used to reduce the number of items. This method is useful for identifying the groups of variables and exploring the relationship between variables. There are several values needed to look at in the EFA test.

Factor Loadings is the correlation coefficient between variables and factors.

System load factor of greater than 0.3 is said to be consistent with a larger sample size of 150 observations.

Bartlett’s Test of Sphericity: The test is used to test the null hypothesis H0. The Barlett’s Test has significant value lower than 0.05, the null hypothesis is rejected.

KMO (Kaiser-Meyer – Olkin) value: The KMO value should range from 0.5 to 1.

If it is lower than 0.5, the factor analysis is no longer suitable.

Cumulative of variance is the variance percentage explained by all the factors. The cumulative variance must be higher than 50% and components must have eigenvalue higher than 1.

Rotated component matrix: Variables that have factor loading value lower than 0.45 will be removed. The study adopts the principle components method that requires the factor loading higher than 0.45 to fit the model.

3.2.3 Regression analysis

After the analysis using the EFA test, the author runs the correlation analysis to determine the correlation between factors of the research model. If the result of the correlation analysis fits the conditions for a regression analysis, the author will proceed to run the multiple regression analysis to investigate the relationship between the independent constructs and the dependent construct. The regression analysis also provides information on the influence level of each independent construct on the dependent factor. The formula of the regression is:

Y = B0 + B1X1 + B 2X2 + B 3X3 + … + B nXn + ei In which:

Y: Job satisfaction of public servants

X = {X1, ..., Xj}: The variable scale factors affect the PUR

B = {B0, ..., Bj}: regression coefficient impact on PUR ei: error term

CHAPTER 4 ANALYSIS RESULT

In this chapter, the author provided the analysis results of the thesis. The results include descriptive statistics, reliability test, EFA, correlation analysis and regression analysis. The author used SPSS software to analyze the data.

4.1. DESCRIPTIVE STATISTICS

The descriptive statistics of the study is shown in Table 4.1

Table 4.1 Descriptive statistics

Frequency Percent

Valid Percent

Cumulative Percent

Gender

Female 52 40 40 40

Male 78 60 60 100.0

Total 130 100.0 100.0

Age

Below 25 years old 20 15.3 15.3 15.3

26-35 years old 45 34.6 34.6 49.9

36-45 years old 25 19.2 19.2 69.1

46-55 years old 25 19.2 19.2 88.3

Above 55 years old 15 11.7 11.7 100.0

Total 130 100.0 100.0

Salary

Below 2 million VND 25 19.2 19.2 19.2

2 - 5 million VND 67 51.5 51.5 70.7

6 - 10 million VND 23 17.6 17.6 88.3

Above 10 million VND 15 11.7 11.7 100.0

Total 130 100.0 100.0

Position

Employee 95 73 73 73

Head of Branch 20 15.3 15.3 88.3

Senior Leader 15 11.7 11.7 100

Total 130 100.0 100.0

Regarding the gender of participants, male participants accounted for 60% and female participants are 40%. For the age of participants, the age group 26 – 35 years old is the largest group (34.6%), followed by the 36-45 years old group (19.2%), the 46 – 55 years old group (19.2%), the below 25 years old group (15.3%) and above 55 years old (11.7%). These figures are reasonable as the frontline employees usually account for the majority of the workforce in an organization. As young people are usually new recruited employees, they usually form the largest group of participants. Regarding the salary of participants, it can be said that the salary of public employees is quite modest as the majority of the public employees have the salary below 5 million VND per month (19.2%

receive lower than 2 million VND per month; 51.5% receive 2 – 5 million VND per month). Only 17.6% of employees receive 6 – 10 million VND per month and 11.7%

receive above 10 million VND per month. Regarding the position in the organization, the majority of participants are Employee (73%), 15.3% are Head of Branch and 11.7% are Senior Leader.

4.2. RELIABILITY TEST

The reliability test is used to check the reliability of the scale used in the study.

The reliability is displayed through the Cronbach’s Alpha coefficient and Item – Total Correlation. As mentioned in Chapter 3, the Cronbach’s Alpha should be higher than 0.6 and the Corrected Item – Total Correlation of observed variables should be higher than 0.3

Table 4.2 Cronbach’s Alpha coefficients

Scale Mean if

Salary & Benefit (SB) Alpha: .801

SB1 11.25 8.255 .635 .720

SB2 11.17 8.318 .622 .786

SB3 11.33 9.415 .637 .757

SB4 11.26 7.656 .620 .730

Promotion (PM) Alpha: .832

PM1 10.62 7.284 .654 .790

PM2 9.94 7.580 .671 .804

PM3 9.88 7.640 .728 .778

PM4 9.89 8.216 .600 .815

Training (TA) Alpha: .754

TA1 11.80 7.634 .565 .697

TA2 10.79 7.506 .608 .685

TA3 11.53 8.347 .581 .698

TA4 11.87 7.693 .641 .604

Job Satisfaction (SAT) Alpha: .780

SAT1 10.36 5.326 .673 .761

SAT2 10.02 4.612 .650 .731

SAT3 10.53 5.362 .622 .642

SAT4 10.11 4.520 .679 .719

Result from Table 4.2 shows that the entire factors have the Cronbach’s Alpha coefficients larger than 0.6 in all factors. No observed variable in the scale has the Item-Total Correlation coefficient less than 0.3 and the Cronbach’s Alpha coefficient would not increase even if there is any observed variable to be deleted. Thus, the author concluded that the reliability test is passed and the scale can be used for the survey.

4.3. EXPLORATORY FACTOR ANALYSIS

After running Cronbach’s Alpha test, the EFA test is used to reduce the number of items. There are several values needed to look at in the EFA test.

First of all, the author observed the KMO and Barlett’s Test results to determine

whether the EFA is suitable for the research. The result is shown in Table 4.3 Table 4.3: KMO và Bartlett’s Test Result

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling

Adequacy. .693

Bartlett's Test of Sphericity

Approx. Chi-Square 798.694

df 66

Sig. .000

Result from Table 4.3 shows that the KMO value was good enough (0.693 > 0.05) and the Sig. of Barlett’s Test also satisfied the condition (Sig. = 0.000 < 0.05). The results indicated that it is possible to run the EFA in this study. Thus, the author continued to look at the Total Variance Explained result in Table 4.4

Table 4.4: Total Variance Explained

Component Initial Eigenvalues

Total % of Variance

Cumulative

1 4,758 39,647 39,647

2 1,725 14,372 54,019

3 1,103 19.164 73,183

4 ,762 6,349 79,531

5 ,678 5,648 85,180

6 ,541 4,509 89,688

7 ,440 3,663 93,351

8 ,316 2,631 95,983

9 ,192 1,603 97,586

10 ,178 1,480 99,066

11 ,112 ,934 100,000

As shown in Table 4.4, there are three factors extracted using the Principal Component Analysis. The factors satisfied the conditions of Eigenvalue (all higher than 1) and the explaination percentage (73.183%). The rotated components are shown in Table 4.5

Table 4.5: Rotated Component Matrix

Component

1 2 3

SB1 .835 SB2 .816 SB3 .795 SB4 .778

PM1 .770

PM2 .760

PM3 .746

PM4 .744

TA1 .730

TA2 .726

TA3 .697

TA4 .680

As shown in Table 4.5, after the rotation using Varimax Rotation Method, the observed variables are converged in the similar components in the research model. Thus, the author adopts the same research model shown in Figure 3.2 with 3 independent variables and 1 dependent variable for the research.

4.4 CORRELATION ANALYSIS AND REGRESSION ANALYSIS

4.4.1 Correlation analysis

Before conducting the regression analysis, the author ran the correlation analysis to find out whether there is correlation between the dependent variable and each of independent variables and between independent variables. If correlation coefficients between independent variables and dependent variables are large, it may be a supporting indication that there is a regression analysis between the variable. However, it should be noted that if the correlation between independent variables is large, the collinearity may occur in the model. The correlation analysis is shown in Table 4.6

Table 4.6: Correlation Analysis

SAT SB PM TA

SAT

Pearson Correlation 1 .574^** .593^** .620^**

Sig. (2-tailed) .000 .000 .000

N 110 110 110 110

Pearson Correlation .574^** 1 .504^** .542^**

Sig. (2-tailed) .000 .000 .000

N 110 110 110 110

Pearson Correlation .593^** .504^** 1 .596^**

Sig. (2-tailed) .000 .000 .000

N 110 110 110 110

Pearson Correlation .620^** .542^** .596^** 1

Sig. (2-tailed) .000 .000 .000

N 110 110 110 110

As shown in Table 4.6, the dependent variable are correlated to the independent variables as the Sig. value in each pair is smaller than 0.05. However, the independent variables are also correlated to one another. Thus, the author should check whether the collinearity occurs in the model through the VIF value. If the collinearity happens, the regression model is not valid.

4.5.2 Regression analysis

Table 4.7 illustrates the Adjusted R Square and the ANOVA Sig. of the model.

The Adjusted R Square can let the author know the prediction percentage of the model and the ANOVA Sig. shows whether the variance of factors is equal or not.

Table 4.7: Model summary

Model R R

Square

Adjusted R Square

ANOVA Sig.

1 .745^a .556 .539 .000^b

Table 4.7 shows that that the Adjusted R Square = 0.539. It indicated that the independent variables of this model can predict 53.9% of the variation of the dependent variable. Besides, the ANOVA Sig. = 0.000 < 0.05, it indicated that the variance of

factors is not equal in the model. This result satisfied the conditions to proceed to the regression analysis. The result of the regression analysis is shown in Table 4.8

Table 4.8: Regression results

Model Unstandardized Coefficients

Standardized Coefficients

t Sig. Collinearity Statistics

B Std.

a. Dependent Variable: SAT

First of all, the VIF coefficients of all independent factors were smaller than 2, therefore, the author can say that the collinearity did not happen in the model. Based on the Sig. values provided in the Table, it can be said that all the independent variables had relationship with the dependent variable (Sig. SB = 0.000 < 0.05, Sig. PM = 0.000 < 0.05, Sig. TA = 0.026 < 0.05). As all the Beta coefficients are positive, it can be said that the independent variables positively influence the dependent variable. Based on the Standardized Beta coefficients, Promotion is the most influential factor (B = 0.271), then Salary & Benefits (B = 0.251), and Training (B = 0.135). Based on Beta coefficients, the regression formula can be written as the following:

SAT = 0.154 + 0.31*SB + 0.341*PM + 0.169*TA

In which:

SAT: Job Satisfaction

SB: Salary & Benefits PM: Promotion TA: Training

From the regression result, the author can confirm the hypotheses as the followings:

For the first hypothesis “Salary and Benefits have positive relationship with job satisfaction”, the result shows that the hypothesis is accepted

For the second hypothesis “Promotion has positive relationship with job satisfaction”, the result shows that the hypothesis is accepted

For the third hypothesis “Training has positive relationship with job satisfaction”, the result shows that the hypothesis is accepted

4.5. Summary

In chapter 4, the author mentioned the descriptive statistics, the scale test using the reliability test and the EFA. Afterwards, the author has conducted the correlation analysis and the regression analysis to identify the relationship between the independent variables and the dependent variable.

CHAPTER 5 CONCLUSION & RECOMMENDATION

5.1. Research Summary

This research is conducted to identify the factors influencing the job satisfaction of public employees in the 5th District of Ho Chi Minh city. Through this research, the author is to suggest some solutions to improve the job satisfaction and thus, improve the work performance of public employees. This is especially important as the 5th District is

在文檔中 Assessing the job satisfaction at the administrative office of the 5th District of Ho Chi Minh City (頁 23-0)