This chapter presents the findings in regard to the research hypotheses of this research study. The first part includes the descriptive statistical analysis and the second part includes the general characteristics of the three scales. The third part includes the results of one-way ANOVA analysis, to see if there is any significant difference in HR effectiveness and job satisfaction respectively among different companies. The fourth, fifth and sixth parts include the results of multiple regression analysis, to see whether organizational performance and job satisfaction will be influenced by HR effectiveness, and organizational performance will be influenced by job satisfaction.
Demographic Characteristics of the Sample
Descriptive statistics of the sample
The respondents come from five different companies, two in financial industry, two in service industry and one in manufacture industry. Five demographic items are shown in Table 4.1. For age, 45.4% of the respondents are aged between 21 to 30 and 53.8% are aged between 31 to 40. For gender, 42.1% of the respondents are males and 57.9% are females. For marital status, 61.1% of the respondents are single and 38.9% are married. For education, 35.7% of the respondents have master’s degree and 34.9% have bachelor degree. For tenure, 40.5% of the respondents work in the present company for 0-2 years and 39.7% work for 3-5 years.
62 Table 4. 1. Descriptive statistics of the sample
Item Description Frequency Percent
Age 25 Education level Senior high school
Academy
63
General Characteristics of All Scales
Strategic HRM functional tasks scale
According to Table 4.2, the mean of item S8 (M=3.31) is the highest and the mean of item S7 (M=2.58) is the lowest. The range of difference of mean and standard deviation among these items are narrow, which indicate that the respondents demonstrate a sense of consistency while answering to these items. The result also shows that companies need to review and improve their policies on employee relations in order to better their HR effectiveness.
Table 4. 2. Descriptive statistics of strategic HRM functional tasks scale
Items N Mean Std. Deviation
S8 Employee Rights 126 3.31 1.054
S2 Staffing 126 3.06 .978
S6 Compensation 126 2.94 1.064
S4 Performance Appraisal 126 2.93 .989
S1 Work Flows 126 2.84 .916
S5 Training and Development 126 2.77 1.082
S9 International Management 126 2.69 1.016
S7 Employee Relations 126 2.58 .991
64 Work on present job scale
According to Table 4.3, the mean of item PJ1 (M=3.53) is the highest and item PJ13 (M=2.97) is the lowest. This result indicates that respondents consider their jobs to be fascinating but they also desire their job content to be more challenging.
Table 4. 3. Descriptive statistics of work on present job scale
Items N Mean Std. Deviation
PJ 1 Fascinating 126 3.53 .855
PJ 9 Pleasant 126 3.37 1.170
PJ 18 Uses my abilities 126 3.25 .829
PJ 7 Respected 126 3.17 1.010
PJ 6 Creative 126 3.10 1.003
PJ 10 Useful 126 3.06 1.026
PJ 3 Satisfying 126 3.03 1.003
PJ 5 Good 126 2.98 1.047
PJ 13 Challenging 126 2.97 1.058
Pay scale
According to Table 4.4, the mean of item P4 (M=3.22) is the highest and item P9 (M=3.00) is the lowest. The result shows strong consistency that the respondents are clearly not satisfied with their salaries compared to the amount of efforts they put into their work. It is normal that the general perception of salary among employees is inclined to be negative because it is never enough when it comes to money. Nevertheless, the companies can still approach this issue by shifting the focus from economic rewards to mental rewards through
65
the process of delegation, empowerment, or job redesigning to highlight the importance of responsibility and achievement. In other word, the organization can help the employees to pursue higher level of needs by providing motivator factors.
Table 4. 4. Descriptive statistics of pay scale
Items N Mean Std. Deviation
P 4 Bad 126 3.22 1.065
P 7 Less than I deserve 126 3.21 .977
P 9 Under paid 126 3.00 1.028
One-Way ANOVA Analysis
As Table 4.5 shown, for HR effectiveness, the F-value is 16.505 and very significant (sig.=000). The Scheffe comparison shows the mean difference between Company 2 (5) and Company 5 (2) is 10.483 (-10.483), which is the largest compared to others and is significant at 0.1 level (sig.=000). This suggests that there is a significant difference between company 2 and company 5. Hence, hypothesis 1 is supported.
From the integrated perspective, WOPJ and pay can be merged into one variable tagged as job satisfaction. For job satisfaction, the F-value is 9.954 and very significant (sig.=000).
The Scheffe Comparison shows the mean difference between Company 2 (5) and Company 5 (2) is 8.693 (-8.693), which is the largest compared to others and is significant at 0.01 level (sig.=000). This suggests that there is a significant difference between Company 2 and Company 5.
For WOPJ, the F-value is 17.239 and very significant (sig.=000). The Scheffé comparison shows that the mean difference between Company 2 (5) and Company 5 (2) is 11.534 (-11.534), which is the largest compared to others. This suggests that there is a
66
significant difference between Company 2 and Company 5. Hence, hypothesis 2-1 is supported.
For pay, the F-value is 10.280 and very significant (sig.=000). The Scheffé comparison shows the mean difference between Company 3 (5) and Company 5 (3) is -3.924 (3.924), which is the largest compared to others. This suggests that there is a significant difference between Company 3 and Company 5. Hence, hypothesis 2-2 is supported.
Based on the results of ANOVA and post hoc tests on job satisfaction, WOPJ, pay from both integrated and respective perspectives, they all show that there is a significant difference in job satisfaction among different companies, therefore hypothesis 2 is supported
Table 4. 5. ANOVA and post hoc tests of HR effectiveness, job satisfaction, WOPJ, pay among different companies
67 Table 4.5. (Continued)
Sum of
Squares df Mean Square F Sig. Summary of Scheffé comparisons Total 6425.302 125
PAY
Between
Groups 2332.456 4 583.114 10.208 .000
3 > 4 3 > 5
(mean difference= 3.924**) Within
Groups 4092.845 121 33.825 Total 6425.302 125
** p<0.01
From the results of Table 4.6 below, it shows that company 2 has the highest mean for HR effectiveness, WOPJ and job satisfaction except for pay, which reveals a positive tendency: if a company has higher HR effectiveness, the job satisfaction of the employees tends to be higher. This tendency reinforces the supporting of hypothesis 3 shown later in the fourth part of this chapter.
68
Table 4. 6. Summary of means of different companies on HR effectiveness, WOPJ, pay and job satisfaction
Company HRE
(Mean)
WOPJ (Mean)
PAY (Mean)
JOB (Mean)
1 23.71 30.57 8.63 39.20
2 28.55 33.41 8.91 42.32
3 25.35 31.52 7.83 39.35
4 19.43 23.53 10.77 34.30
5 18.06 21.88 11.75 33.62
Regression Analysis of HR Effectiveness on Job Satisfaction
Influence of HR effectiveness on WOPJ
Stepwise regression was used to test the influence of HR effectiveness on WOPJ (Work on Present Job). The results of Table 4.7 show that three items (S2, S6 and S5) of HR effectiveness scale could explain 48.9% of variance (F=38.918, p < 0.01) and they have the significantly positive beta value of .363, .312 and .171 respectively, meaning the influence of S2 (staffing), S6 (compensation) and S5 (training and development) are positive to WOPJ and S2 has bigger influence than S6 and S5. This indicates that the better the employees’
perceptions of staffing, compensation and training and development in the organization, the more satisfied the employees will feel toward the content of their jobs. For collinearity, tolerance, VIF and CI in the model are .663(>.01), 1.520(<10) and 8.639(<30), which present no problem with collinearity. Hence, hypothesis 3-1 is supported by the results.
69
Table 4. 7. Summary of regression analysis of HR effectiveness on WOPJ
Independent variables Model
value/Sig.
S2 Staffing .363/.000**
S6 Compensation .312/.000**
S5 Training and Development .171/0.026*
R2 .489
F 38.198**
Tolerance/VIF .663/1.520
CI 8.639
*p<0.05, ** p<0.01
Influence of HR effectiveness on pay
Stepwise regression was used to test the influence of HR effectiveness on pay. The results of Table 4.8 show that two items (S6 and S8) of HR effectiveness scale could explain 28% of variance (F=23.861, p < 0.01) and they have the significantly negative beta value of -.368 and -.214 respectively, meaning the influence of S6 (compensation) and S8 (employee rights) are positive to pay and S6 has bigger influence than S8. This indicates that the better the employees’ perceptions of compensation and employee rights in the organization, the more satisfied the employees will feel toward their salaries. For collinearity, tolerance, VIF and CI in the model are .608(>.01), 1.645(<10) and 7.785(<30), which present no problem with collinearity. Hence, hypothesis 3-2 is supported by the results.
70
Table 4. 8. Summary of regression analysis of HR effectiveness on pay
Independent variables Model
value/Sig.
S6 Compensation -.368/.000**
S8 Employee Rights -.214/0.031*
R2 .28
F 23.861**
Tolerance/VIF .608/1.645
CI 7.785
*p<0.05, ** p<0.01
Summary of regression analysis results of HR effectiveness on job satisfaction
For testing the influence of HR effectiveness on WOPJ, item S2 (staffing), S6 (compensation) and S5 (training and development) are found to be predictive of how the employees feel toward the content of their jobs, hypothesis 3-1 is therefore supported by this result.
For testing the influence of HR effectiveness on pay, item S6 (compensation) and S8 (employee rights) are found to be predictive of how the employees feel toward their salaries, hypothesis 3-2 is therefore supported by this result.
In sum, the overall results suggest that hypothesis 3 is supported.
71
Regression Analysis of HR Effectiveness on Organizational Performance
Influence of HR effectiveness on PGR
Stepwise regression was used to test the influence of HR effectiveness on PGR. The results of Table 4.5 show that two items (S9 and S1) of HR effectiveness scale could explain 18.3% of variance (F=13.756, p < 0.01) and they have the significantly positive and negative beta value of .469 and -.193 respectively, meaning the influence of S9 (international management) and S1 (work flows) are positive to PGR and S9 has bigger influence than S.
This indicates that the better the employees’ perceptions of international management and work flows in the organization, the higher the PGR will become. For collinearity, tolerance, VIF and CI in the model are .832(>.01), 1.202(<10) and 7.067(<30), which present no problem with collinearity. Hence, hypothesis 4-1 is supported by the results.
Table 4. 9. Summary of regression analysis of HR effectiveness on PGR
Independent variables Model
value/Sig.
S9 International Management S1 Work Flows
.469/.000**
-.193/.033*
R2 .183
13.756**
.832/1.202 7.067 F
Tolerance/VIF CI
*p<0.05, ** p<0.01
72 Influence of HR effectiveness on SGR
Stepwise regression was used to test the influence of HR effectiveness on SGR. The eight items of HR effectiveness scale could only explain 8.9% of variance (F=1.436, p>0.05) and the beta value of the 8 items range from -.183 to .212, all of them are insignificant (p>0.05).
Due to these results, the influence of HR effectiveness on SGR can be summarized as insignificant. Hence, hypothesis 4-2 is rejected.
Influence of HR effectiveness on ROE
Stepwise regression was used to test the influence of HR effectiveness on ROE. The results shown in Table 4.6 show that two items (S5 and S7) of HR effectiveness and job satisfaction could explain 15.7% of variance (F=11.444, p < 0.01) and they have the significantly positive beta value of .275 and .203 respectively, meaning the influence of S5 (training and development) and S7 (employee relations) are positive to ROE and S5 is bigger than S7. This indicates that the better the employees’ perceptions of training and development, employee relations in the organization, the higher the ROE will become. For collinearity, tolerance, VIF and CI in the model are .870(>.01), 1.149(<10) and 6.465(<30), which present no problem with collinearity. Hence, hypothesis 4-3 is supported by the results.
73
Table 4. 10. Summary of regression analysis of HR effectiveness on ROE
Independent variables Model
value/Sig.
S5 Training and Development .275/.002**
S7 Employee Relations .203/.024*
R2 .157
F 11.444**
Tolerance/VIF .870/1.149
CI 6.465
*p<0.05, ** p<0.01
Summary of regression analysis results of HR effectiveness on organizational performance
For testing the influence of HR effectiveness on PGR, item S9 (international management) and S1 (work flows) are found to be predictive of PGR, hypothesis 4-1 is therefore supported by this result.
For testing the influence of HR effectiveness on SGR, none of the items is positive to SGR, hypothesis 4-2 is therefore rejected by this result.
For testing the influence of HR effectiveness on ROE, item S5 (training and development) and S7 (employee relations) are found to be predictive of ROE, hypothesis 4-3 is therefore supported by this result.
In sum, the overall results suggest that hypothesis 4 is partly supported.
74
Regression Analysis of Job Satisfaction on Organizational Performance
Influence of WOPJ and pay on PGR
Stepwise regression was used to test the influence of WOPJ and pay on PGR. The results of Table 4.9 show that two items (PJ3 and P7) of WOPJ and pay scales could explain 8.4% of variance (F=5.666, p < 0.05) and they have the significantly positive beta value of .257 and .240 respectively, meaning the influence of PJ3 (satisfying) and P7 (less than I deserve) are positive to PGR and PJ3 has bigger influence than P7. This indicates that the more satisfied the employees feel toward the content of their jobs and the less they feel underpaid, the higher the PGR will become. For collinearity, tolerance, VIF and CI in the model are .900(>.01), 1.112(<10) and 11.636(<30), which present no problem with collinearity.
Hence, hypothesis 5-1 is supported by the results.
Table 4. 11. Summary of regression analysis of WOPJ and pay on PGR
Independent variables Model
value/Sig.
PJ3 Satisfying .257/.006*
P7 Less than I deserve .240/.010*
R2 .084
F 5.666*
Tolerance/VIF .900/1.112
CI 11.636
*p<0.05
75 Influence of WOPJ and pay on SGR
Stepwise regression was used to test the influence of WOPJ and pay on SGR. The results of Table 4.10 show that two items (PJ9 and PJ10) of WOPJ and pay scales could explain 11%
of variance (F=7.563, p < 0.05) and they have the significantly positive beta value of .378 and -.316 respectively, meaning the influence of PJ9 (pleasant) and PJ10 (useful) are positive to SGR and PJ9 has bigger influence than PJ10. This indicates that the more pleasant and useful the employees feel toward the content of their jobs, the higher the SGR will become. For collinearity, tolerance, VIF and CI in the model are .688(>.01), 1.453(<10) and 7.565(<30), which present no problem with collinearity. Hence, hypothesis 5-2 is supported by the results.
Table 4. 12. Summary of regression analysis of WOPJ and pay on SGR
Independent variables Model
value/Sig.
PJ9 Pleasant .378/.000**
PJ10 Useful -.316/.003*
R2 .11
F 7.563*
Tolerance/VIF .688/1.453
CI 7.565
*p<0.05, ** p<0.01
76 Influence of WOPJ and pay on ROE
Stepwise regression was used to test the influence of WOPJ and pay on ROE. The results of Table 4.11 show that one item (PJ3) of WOPJ and pay scales could explain 5.7% of variance (F=7.476, p < 0.05) and they have the significantly positive beta value of .238, meaning the influence of PJ3 (satisfying) is positive to ROE, indicating that the more satisfied the employees are about the content of their jobs, the higher the ROE will become. For collinearity, tolerance, VIF and CI in the model are 1.000(>.01), 1.000(<10) and 6.372(<30), which present no problem with collinearity. Hence, hypothesis 5-3 is supported by the results.
Table 4. 13. Summary of regression analysis of WOPJ and pay on ROE
Independent variables Model
value/Sig.
PJ3 Satisfying .238/.007**
R2 .057
F 7.476*
Tolerance/VIF 1.000/1.000
CI 6.372
**p<0.05, **p<0.01
77
Summary of regression analysis results of job satisfaction on organizational performance
For testing the influence of WOPJ and pay on PGR, item PJ3 (satisfying) and P7 (less than I deserve) are found to be predictive of PGR, hypothesis 5-1 is therefore supported by this result.
For testing the influence of WOPJ and pay on SGR, item PJ9 (pleasant) and PJ10 (useful) are found to be predictive of SGR, hypothesis 5-2 is therefore supported by this result.
For testing the influence of WOPJ and pay on ROE, item PJ3 (satisfying) is found to be predictive of ROE, hypothesis 5-3 is therefore supported by this result.
In sum, the overall results suggest that hypothesis 5 is supported.
The summary of the results of hypothesis testing is shown in Table 4.14.
78 Table 4. 14. Results of hypothesis testing
Hypothesis Testing Result
H1 There is a significant difference in HR effectiveness among different companies.
Supported
H2 There is a significant difference in job satisfaction among different companies.
Supported
H2-1 There is a significant difference in employees’ perceptions of the content of their jobs among different companies.
Supported
H2-2 There is a significant difference in employees’ perceptions of their salaries among different companies.
Supported
H3 The higher the HR effectiveness, the higher the job satisfaction of the employees will become.
Supported
H3-1 The higher the HR effectiveness, the more satisfied the employees will feel toward the content of their jobs.
Supported
H3-2 The higher the HR effectiveness, the more satisfied the employees will feel toward their salaries. H4-1 The higher the HR effectiveness, the higher the PGR will become. Supported H4-2 The higher the HR effectiveness, the higher the SGR will become. Rejected H4-3 The higher the HR effectiveness, the higher the ROE will become. Supported
H5 The higher the job satisfaction of the employees, the better the
79