This chapter shows the statistical analysis results of the data collected from Taiwanese employees in various industries in Taiwan. IBM SPSS 23.0 was adopted for data analysis in this study. Two major sections are presented in this chapter: the result section and the discussion section. The chapter starts by presenting the descriptive statistics of the sample, followed by Pearson correlation analysis of control variables and the constructs of this study.
The result of hierarchical regression analysis for the research hypothesis closes the result section. The discussion section provides the summary and discussion from the statistical analysis result.
Result of Descriptive Analysis
Table 4.1 presents the demographic statistics of the participants. There are five demographic variables included in this study: gender, age, marital status, education level, and tenure. The total valid number of respondents is 205 and more than half of them (67.8%) are female. People under the age of 30 account for nearly one-third, those aged from 30 to 40 and 40 to 50 share roughly the same proportions which are approximately 25%. 35 aged above 50 account for 14.6%. Regarding marital status, there are 120 single participants (58.5%) compared to 85 married participants (41.5%). Most of the people have bachelor’s degree (51.7%), master’s degree holders rank second with 45.4%, nearly 3% has high school diploma or doctor’s degree. As far as the work tenure is concerned, people with working experience less than 5 years accounts for more than 50%, followed by those with 5-10 years of work experience (18%), 30 people have worked more than 20 years in their current companies (14.6%) and around 14% of the participants has their work tenure falling within the range of 11 to 20 years.
44 Table 4.1
Participants Demographic Statistics
Variable Category Frequency Percentage (%)
Gender Female 139 67.8
Male 66 32.2
Age (years) <30 60 31.3
30-40 49 25.5
41-50 48 25.0
>50 35 18.2
Marital status Single 120 58.5
Married 85 41.5
Education High school diploma 3 1.5
Bachelor’s degree 106 51.7
Master’s degree 93 45.4
Doctor’s degree 3 1.4
Tenure (years) <5 107 52.2
5-10 37 18.0
11-15 19 9.3
16-20 12 5.9
>20 30 14.6
Note. N = 205
45
Result of Pearson Correlation Analysis
Pearson correlation analysis was applied to examine the correlation degree and the direction among variables in the study. Table 4.2 shows the mean, standard deviation and correlations among measured variables. International reliability coefficient (indicated by Cronbach Alpha) for employee CSR perception, OCB, employee volunteering behavior, prosocial motivation, task significance is .890, .833, .926, .913, .847 respectively. These Cronbach Alpha results are all above .80 indicating good internal reliability of the measurements.
As shown in Table 4.2, the independent variable has a significant positive relationship with dependent variables. Specifically, Employee CSR perception is significantly positively correlated with employee OCB (r = .432, p < .01) and with employee volunteering (r = .422, p
< .01), indicating that employees with better CSR perception are likely to engage in OCB and volunteering behavior. In addition, employee CSR perception has a significant positive relationship with prosocial motivation (r = .492, p < .01). This result implies that employees with better CSR perception tend to be more prosocially motivated. The results also show prosocial motivation is significantly correlated with employee OCB (r = .482, p < .01) and employee volunteering behavior (r = .518, p < .01). This result implies that the higher employees are prosocially motivated, the more likely they engage in OCB and volunteering behavior. Task significance was found significantly positively related to prosocial motivation (r = .538, p < .01), indicating the more task significant is, the more the employees feel prosocially motivated. In addition, task significance has significant correlations with employee OCB (r = .447, p < .01) and employee volunteering behavior (r = .434, p < .01). This finding indicates that employees with higher task significance are likely to express OCB and volunteering behaviors. Regarding control variables, age have positive relationships with some variables, including employee CSR perception (r = .184, p < .05), employee OCB (r = .152, p
< .05), employee volunteering behavior (r = .213, p < .05), prosocial motivation (r = .142, p <
.05) and task significance (r = .140, p < .05). Education level have significant positive correlations with employee volunteering (r = .163, p < .05) and prosocial motivation (r = .140, p < .05).
46 Table 4.2
Mean, Standard Deviation, Correlation and Reliability of Variable
Variable Mean SD 1 2 3 4 5 6 7 8 9 10
Gender 1.68 .47
Age 37.80 11.12 -162*
Tenure 8.59 9.54 -.164* .696**
Education level 2.47 .56 -.096 .038 -.056
Marital status .41 .49 -.225** .528** .458** -.032 Employee CSR
perception 3.25 1.12 -.044 .184** .129 .095 .121 (.890)
Employee OCB 3.55 .94 .059 .152* .101 -.021 .049 .432** (.833)
Employee volunteering 2.55 1.24 -.044 .213** .075 .163* -.001 .422** .420** (.926)
Prosocial motivation 3.40 1.22 -.068 .142* .059 .140* .085 .492** .482** .518** (.913)
Task significance 2.75 1.18 -.062 .140* .039 .086 .044 .429** .447** .434** .538** (.847) Note. N = 205. Numbers in parenthesis are Cronbach Alpha.
values. *p < .05. **p < .01.
47
Result of Hierarchical Regression Analysis
This part is dedicated to discussing the result of regression analysis to verify the hypotheses of this research.
Hypothesis 1 proposed that employee CSR perception has a positive relationship with employee OCB. The summary of the hierarchical regression analysis for Hypothesis 1 is shown in Table 4.3. Two steps were done in this hierarchical regression analysis. In step one, five control variables were entered followed by employee CSR perception in the second step. The results show that 16.7% of the total variance in employee OCB can be explained by employee CSR perception (F = 8.029, p < .001). Employee CSR perception was found to have a significant and positive relationship with employee OCB (β = .419, p < .001). Based on these results, it is implied that if employees have a better CSR perception, they tend to be more involved in OCB. Hypothesis 1 is supported.
Table 4.3
Results of Hierarchical Regression Analysis for Hypothesis 1
Variable β
Note. N = 205. Dependent variable = Employee OCB.
***p < .001.
48
Hypothesis 2 proposed that employee CSR perception has a positive relationship with employee volunteering behavior. The summary of the hierarchical regression analysis for Hypothesis 2 is shown in Table 4.4. Two steps were conducted in this hierarchical regression analysis. In step one, five control variables were entered. In the second step, employee CSR perception was entered. The results show that 14.3% of the total variance in employee volunteering can be explained by employee CSR perception (F = 9.921, p < .001). Employee CSR perception was found to have a significant and positive relationship with employee volunteering (β = .388, p < .001). Based on these results, it is implied that if employees have better CSR perception, they tend to engage more in volunteering behaviors. Hypothesis 2 is supported.
Table 4.4
Results of Hierarchical Regression Analysis for Hypothesis 2
Variable
β
1 2
Step 1
Gender -.026 -.025
Age .338*** .282**
Tenure -.091 -.096
Education level .141 .102
Marital status -.135 -.156
Step 2
Employee CSR perception .388***
F 3.877** 9.921***
R² .090** .233***
Adj. R² .066** .209***
∆R² .090** .143***
∆F 3.877** 36.634***
Note. N = 205. Dependent variable = Employee volunteering.
**p < .01. ***p < .001.
49
Hypothesis 3 proposed that employee CSR perception has a positive relationship with prosocial motivation. Summary of the hierarchical regression analysis for Hypothesis 3 can be found in Table 4.5. Two steps were performed in this hierarchical regression analysis. In step one, five control variables were entered. In the second step, employee CSR perception was entered. The results show that 20.8% of the total variance in prosocial motivation can be explained by employee CSR perception (F = 11.022, p < .001). Employee CSR perception was found to have a significant and positive relationship with prosocial motivation (β = .466, p <
.001). Based on these results, it is implied that if employees have a better CSR perception, they tend to be more prosocially motivated. Hypothesis 3 is supported.
Table 4.5
Results of Hierarchical Regression Analysis for Hypothesis 3
Variable
β
1 2
Step 1
Gender -.035 -.034
Age .156 .088
Tenure -.065 -.070
Education level .137 .090
Marital status .038 .014
Step 2
Employee CSR perception .466***
F 1.839 11.022***
R² .045 .252***
Adj. R² .020 .229***
∆R² .045 .208***
∆F 1.839 54.440***
Note. N = 205. Dependent variable = Prosocial motivation.
***p < .001.
50
Hypothesis 4 proposed that prosocial motivation has a positive relationship with employee OCB. Summary of the hierarchical regression analysis for Hypothesis 4 can be found in Table 4.6. Two steps were performed in this hierarchical regression analysis. In step one, five control variables were entered. In the second step, employee prosocial motivation was entered. The results show that 22.2% of the total variance in employee OCB can be explained by prosocial motivation (F = 11.031, p < .001). The prosocial motivation is found to have a significant and positive relationship with employee OCB (β = .483, p < .001). Based on these results, it is implied that the more prosocially motivated the employees are, the more likely they engage in OCB. Hypothesis 4 is supported.
Table 4.6
Results of Hierarchical Regression Analysis for Hypothesis 4
Variable
β
1 2
Step 1
Gender .080 .097
Age .171 .096
Tenure .002 .033
Education level -.011 -.077
Marital status -.013 -.032
Step 2
Prosocial motivation .483***
F 1.216 11.031***
R² .030 .252***
Adj. R² .005 .230***
∆R² .030 .222***
∆F 1.216 58.335***
Note. N = 205. Dependent variable = Employee OCB.
***p < .001.
51
Hypothesis 5 proposed that prosocial motivation has a positive relationship with employee volunteering behavior. The summary of the hierarchical regression analysis for Hypothesis 5 is shown in Table 4.7. Two steps were done in this hierarchical regression analysis. In step one, five control variables were entered followed by prosocial motivation in the second step. The results showed that 22.4% of the total variance in employee volunteering can be explained by prosocial motivation (F = 14.950, p < .001). The prosocial motivation was found to have a significant and positive relationship with employee volunteering (β = .485, p
< .001). Based on these results, it is implied that the more prosocially motivated the employees are, the more likely they involve in volunteering behaviors. Hypothesis 5 is supported.
Table 4.7
Results of Hierarchical Regression Analysis for Hypothesis 5
Variable
β
1 2
Step 1
Gender -.026 -.009
Age .338*** .263**
Tenure -.091 -.060
Education level .141* .075
Marital status -.135 -.154
Step 2
Prosocial motivation .485***
F 3.877** 14.950***
R² .090** .314***
Adj. R² .066** .293***
∆R² .090** .224***
∆F 3.877** 64.108***
Note. N = 205. Dependent variable = Employee volunteering.
*p < .05. **p < .01. ***p < .001.
52
Hypothesis 6 proposed that prosocial motivation mediates the positive relationship between employee CSR perception and employee OCB. A series of regression analyses were conducted to investigate the mediating effect of prosocial motivation based on suggestions of Baron and Kenny (1986). Accordingly, three regressions were performed: regressing employee OCB on employee CSR perception (1), regressing prosocial motivation on employee CSR perception (2), regressing employee OCB on prosocial motivation (3), regressing OCB on both employee CSR perception and prosocial motivation (4). The regressions (1), (2), and (3) had been done in Hypotheses 1, 3, and 4 respectively. These hypotheses testing results all show supporting results, meaning regression (1), (2), and (3) yield significant results. Table 4.8 shows the result of regression (4). Three steps were done in this analysis. Step one involved entering five demographic variables. In step two, employee CSR perception was entered, and in the final step, the meditator prosocial motivation was entered.
According to the result, the regression coefficient for employee CSR perception is subsequently reduced from Model 2 to Model 3 (.419 to .247) but remains significant (p < .01), and prosocial motivation positively affects employee OCB (β = .367, p < .01), which confirmed the partial mediating effect of prosocial motivation (see Table 4.8). In other words, prosocial motivation plays a partial mediator in the positive relationship between employee CSR perception and employee OCB. Hypothesis 6 is supported.
Table 4.8
Results of Hierarchical Regression Analysis for Hypothesis 6
Variable β
Employee CSR perception .419*** .247***
(continued)
53 Table 4.8
Results of Hierarchical Regression Analysis for Hypothesis 6 (continued)
Variable β
Note. N = 205. Dependent variable = Employee OCB.
***p < .001.
Hypothesis 7 proposed that prosocial motivation mediates the positive relationship between employee CSR perception and employee volunteering. A series of regression analyses were conducted to investigate the mediating effect of prosocial motivation based on suggestions of Baron and Kenny (1986). Accordingly, three regressions were performed:
regressing employee volunteering on employee CSR perception (1), regressing prosocial motivation on employee CSR perception (2), regressing employee volunteering on prosocial motivation (3), regressing OCB on both employee CSR perception and prosocial motivation (4). The regressions (1), (2), and (3) had been done in Hypotheses 2, 3, and 5 respectively.
These hypotheses testing results all show supporting results, meaning regression (1), (2), and (3) yield significant results. Table 4.9 shows the result of regression (4). Three steps were done in this analysis. Step one involved entering five demographic variables. In step two, employee CSR perception was entered and in the final step, the meditator prosocial motivation was entered.
According to the result, the regression coefficient for employee CSR perception is subsequently reduced from Model 2 to Model 3 (.388 to .206) and remains less significant (p
< .05), and prosocial motivation positively affects employee volunteering (β = .388, p < .01), which confirmed the partial mediating effect of prosocial motivation (see Table 4.9). In other words, prosocial motivation plays a partial mediator in the positive relationship between employee CSR perception and employee volunteering. Hypothesis 7 is supported.
54 Table 4.9
Results of Hierarchical Regression Analysis for Hypothesis 7
Variable β
Employee CSR perception .388*** .206**
Step 3
Note. N = 205. Dependent variable = Employee volunteering.
*p < .05. **p < .01. ***p < .001.
Hypothesis 8 proposed that the relationship between employee CSR perception and prosocial motivation is moderated by task significance, such that this relationship is stronger at higher levels of task significance. The hierarchical regression analysis was used to check the moderating effect. The three analysis steps were conducted in IBM SPSS 23.0. In step one, five controlled variables were entered. Next, predictors, which are employee OCB and task significance, were added in in step two. Lastly, employee CSR perception x task significance interaction term was added in step three. The output shown in Table 4.10 indicates that while the second regression including predictors shows a significant increase in R-square change (∆R² = 331, p < .001), regression 3 shows a non-significant increase when the interaction effect was added (∆R² = .005, p > .05). In other words, the interaction term does not add significant explanatory power forModel 3.Thus, Hypothesis 8 is unsupported.
55 Table 4.10
Results of Hierarchical Regression Analysis for Hypothesis 8
Variable β
1 2 3
Step 1
Gender -.035 -.015 -.015
Age .156 .030 .043
Tenure -.065 -.031 -.047
Education level .137 .076 .069
Marital status .038 .031 .037
Step 2
Employee CSR perception .307*** .297***
Task significance .391*** .412***
Step 3
Employee CSR perception x Task
significance -.072
F 1.839 16.784*** 14.901***
R² .045 .376*** .381***
Adj. R² .020 .354*** .355
∆R² .045 .331*** .005
∆F 1.839 51.777*** 1.451
Note. N = 205. Dependent variable = Prosocial motivation.
***p < .001.
56
Discussion
This study aims to investigate the relationship among employee CSR perception, employee OCB, employee volunteering behavior, prosocial motivation, and task significance.
In this section, the results of all hypotheses testing in this research are discussed in detail with possible explanations and implications. Summary of hypothesis testing results can be found in Table 4.11.
Table 4.11
Hypotheses Testing Results Summary
Hypotheses Result
H1 There is a positive relationship between employee CSR perception and OCB.
Supported
H2 There is a positive relationship between employee CSR perception and employee volunteering behavior.
Supported
H3 There is a positive relationship between employee CSR perception and prosocial motivation.
Supported
H4 There is a positive relationship between prosocial motivation and OCB.
Supported
H5 There is a positive relationship between prosocial motivation and employee volunteering behavior.
Supported
H6 Prosocial motivation mediates the relationship between employee CSR perception and OCB.
Supported
H7 Prosocial motivation mediates the relationship between employee CSR perception and employee volunteering behavior.
Supported
H8 Task significance strengthens the positive relationship between employee CSR perception and employee’s prosocial motivation.
Unsupported
57
Effects of Employee CSR Perception on OCB and Employee Volunteering Behavior
As socially responsible activities have gradually become one of the important strategies of the companies, and employees tend to care more about these activities, it is important to investigate the effects of CSR on prosocial behaviors of employees, for example, OCB and volunteering behavior. This study thus aims to enrich this understanding. Based on previous research and theories, this research explains the relation between employee CSR perception and employee prosocial behaviors like OCB and employee volunteering behavior. Specifically, employee CSR perception was proposed to have a positive relationship with OCB and volunteering behaviors. According to the hypotheses testing result, this proposition is supported. This result is supported by the study of Newman et al. (2015) about how CSR perception impacts stakeholders’ behaviors, including employee OCB. Ko et al. (2018) generalized this relation by concluding that CSR perception affects employee prosocial behaviors, not restricted to OCB. This conclusion supports the result of this research on the relationship between CSR perception and volunteering behavior, noting that volunteering behaviors are also prosocial behaviors in nature. Regarding the mechanism between CSR perception and employee prosocial behaviors like OCB and employee volunteering behavior, social identity theory is the most commonly used theory to explain this relationship. Employees are strongly influenced by how they perceive their organization (Cropanzano & Mitchell, 2016). Those who perceive their company as being socially responsible are likely to identify themselves with the organization, which boosts the intrinsic motivation to engage in prosocial behaviors (Baruch & Bozionelos, 2011). Specifically, a good perception of CSR motivates a sense of pride and self-esteem in the employees as they believe they belong to a socially responsible organization, which in turn boosts them to express prosocial behaviors like OCB in order to maintain and develop this organizational identity (Ellemers et al., 2004).
58
Mediating Effect of Prosocial Motivation in the Relationship Between Employee CSR Perception and OCB, Employee Volunteering Behavior
Another objective of this research is to investigate the effect of prosocial motivation in the relationship between CSR perception and prosocial motivations, including OCB and volunteering behaviors. The result shows that prosocial motivation mediates the relationships between employee CSR perception and employee prosocial behaviors like OCB and volunteering behaviors. This result is consistent with the study of Ong et al. (2018). In order to explain this relationship mechanism, two sub-relationships should be looked at. First, the relationship between CSR perception and prosocial motivation. The result of this research indicates that CSR perception has a significant positive relationship with prosocial behaviors, implying that the employees with a better perception of CSR are more prosocially motivated.
According to the research of Shin and Hur (2020), this finding can be explained by ethical working climate and external reputation. CSR activities for the sake of others create an ethical working climate, consequently, promote a sense of community and a norm concerning others’
well-being. This ethical climate triggers employee prosocial motivation. On the other hand, CSR is found to increase a firm’s reputation, which in turn motivates employees to act socially to keep a good image of their firm in external stakeholders’ eyes (Shin & Hur, 2020). Regarding the link between prosocial motivation and prosocial behaviors like OCB and employee volunteering behavior, three explanations are found. Firstly, prosocial employees have their concern for others’ well-being, there is a higher chance that they will seek opportunities to support others and their company (Meglino & Korsgaard, 2005). Secondly, employees with high prosocial motivation will likely perceive that doing good for others is their responsibility, involving in prosocial behaviors can give them a sense of commitment (Grant, 2007, 2008;
Ilies et al., 2006). Thirdly, highly prosocially motivated employees always orient their concern for others, they will likely contribute for a better good at the expense of their interest (Bolino
& Turnley, 2005; Meglino & Korsgaard, 2005; Moon et al., 2008).
59
Moderating Effect of Task Significance in the Relationship Between Employee CSR Perception and Prosocial Motivation
Task significance was proposed to moderate the relation between CSR perception and prosocial motivation, meaning that employees with higher task significance were expected to be more prosocially motivated when they have a better perception of CSR. This hypothesis was formed based on the study of Ong et al. (2018). Ong et al. (2018) found that those with high task significance tend to be sensitive to how their organizations treat others, once organizations show support for external stakeholders via CSR initiatives, they will be prosocially motivated to engage in prosocial behaviors. CSR sensitivity can be explained by the virtuous cycle of making a positive impact (Grant, 2007). This cycle suggests that once the employees, through the tasks of their job, perceive the positive influences on the beneficiaries tend to have a desire to make even more prosocial impacts. However, the result of this study is inconsistent with the findings of previous studies (Grant, 2007; Ong et al., 2018). Task significance itself shares a positive relationship with prosocial motivation, not the interaction effect of CSR perception and task significance. This result implies that the Taiwanese participants in this study regardless of whether their task significance is high, are still prosocially motivated if they have a good CSR perception. The finding can likely be explained by Taiwanese cultural characteristics. In a collectivistic society like Taiwan, people tend to care more about their group’s well-being than their own and consider the group’s interests and goals their priorities (Earley, 1994; Hofstede, 2001). For these characteristics, Taiwanese are more likely to be concerned for other’s well-being (Takeuchi et al., 2015). It is likely that the strong others-oriented nature of the Taiwanese easily makes them prosocially motivated regardless of their task significance influence.
61