• 沒有找到結果。

The sample data collected was empirically analyzed using SPSS 15.0 and Smart PLS 3.0 tools. This chapter presents the findings of the main study. First, it explains the correlation analysis. Second, it provides the reliability and validity analysis result of the main study.

Lastly, regression analysis and PLS method are discussed.

Correlation Analysis

Pearson coefficient correlation shown in Table 5.1. was implemented to examine the direction and strength of linear relationship between variables. According to the results, transformational leadership, organizational trust, customer relationship management, innovation capabilities and business performance are significantly correlated. Furthermore, the relationship between dependent variables is positively related to independent variables.

The correlation analysis tested whether the multicollinearity problem exists in the model. The correlation needs to be below .75 or it is considered as problematic (Kennedy, 1989). Therefore, the correlation table shows .77 values that are greater than .75 so it implies that there is multicollinearity in the table (see table 5.1.).

46 Table 5.1.

Correlation Analysis (Main Study, N=225)

Note. TLIC= Intellectual Consideration, TLIS= Intellectual Stimulation, TLIM= Intellectual Motivation, TLIIA= Idealized influence attributes, TLIIB= Idealized influence behavior, Comp= Competence OH=

Openness/honesty, Cfe= Concern for employees, Reli= Reliability, Iden= Identification, WbM1= Win-back management, PS= Product/Service customization, CIM= customer information management, PI= Product Information, PCI= Process Information, ML= Market Leadership, FP= Financial Performance.

**p<.001.

Constructs Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

1 MTLIC .354 .889 1

2 MTLIS 3.81 1.128 .620** 1

3 MTLIM 3.65 1.021 .635** .574** 1

4 MTLIIA 3.69 .988 .460** .684** .572** 1

5 MTLIIB 3.60 1.036 .597** .770** .638** .727* 1

6 Mcomp 3.52 1.037 .446** .535** .571** .422** .578** 1

7 MOH 3.51 .812 .435** .347** .452** .304** .396** .518** 1

8 Mcfc 3.43 .903 .581** .551** .615** .566** .686** .631** .552** 1

9 Mreli 3.42 1.077 .580** .483** .605** .568** .544** .430** .447** .650** 1

10 Miden 3.18 .932 .492** .366** .621** .349** .428** .472** .599** .598** .578** 1

11 MWbM1 3.08 1.138 .416** .445** .419** .303** .495** .561** .343** .457** .417** .329** 1

12 MPS 3.44 1.006 .419** .395** .344** .239** .429** .471** .394** .455** .385** .401** .666** 1

13 MCIM 3.81 1.059 .499** .421** .385** .435** .460** .372** .427** .532** .526** .413** .515** .563** 1

14 MPI 3.30 0.859 .295** .261** .317** .156** .290** .375** .260** .378** .315** .241** .519** .468** .327** 1

15 MPCI 3.55 0.942 .458** .303** .373** .238** .318** .387** .332** .374** .424** .335** .553** .544** .416** .630** 1

16 MML 3.56 0.951 .372** .280** .344** .147 .331** .439** .268** .296** .209** .271** .507** .476** .258** .459** .531** 1 17 MFP 3.51 1.012 .359** .376** .397** .195** .415** .424** .239** .316** .200** .269** .534** .408** .292** .454** .444** .668** 1

47

Validity and Reliability Analysis

Cronbach’s Alpha reliability was performed for the main study. As it can be seen from the Table 5.2. both countries have values higher than .70 which indicated that the data is reliable. It also implies that all scales have a high level of internal consistency and reliability.

Table 5.3. also represents total sample for the main study (N=225). The Cronbach’s Alpha lowest value is Innovation capabilities (0.773) while the highest is Transformational Leadership (0.894).

Table 5.2.

Cronbach’s Alpha Results for All Dimensions (Main Study, N =225)

Constructs Number of Items Cronbach’s Alpha

TL 20 0.894

OT 29 0.858

CRM 8 0.806

IC 9 0.773

BP 10 0.801

Note. TL= Transformational Leadership, OT= Organizational Trust, CRM= Customer Relationship Management, IC= Innovation Capabilities, BP= Business Performance.

Partial Least Square (PLS) Analysis

The standard PLS model analyzed and interpreted two stages: the assessment of reliability and validity of the measurement model and the assessment of the path coefficients of structural model of the main study. The results for Taiwanese and Bosnian sample will be presented as well as the united sample.

Validity and Reliability

The model was tested using validity and reliability using SmartPLS tool. The composite reliability and average variance extracted (AVE) were examined in order to test convergent validity. The minimum value of .70 is required for composite reliability (Nunally

& Bernstein, 1994).

Average variance extracted (AVE) determined the variance captured by the indicators relative to the measurement error. According to Hair et al. (2011) the values for AVE should be greater than .50. The lowest value is Organizational Trust (0.638) while BP has the highest value (0.834).

48

According to Chin (1998), loadings of 0.50 or higher meet the requirement for convergent validity. Table 5.4. represents PLS loadings for main study (N=225) and points out that all values are higher than 0.50.

Table 5.3.

Measurement Model Results (Main Study, N =225) Constructs Number of

Note. TL= Transformational Leadership, OT= Organizational Trust, CRM= Customer Relationship Management, IC= Innovation Capabilities, BP= Business Performance.

Table 5.4.

PLS Loadings (Main Study N=225)

Note. TLIC= Intellectual Consideration, TLIIA= Idealized influence attributes, TLIIB=

Idealized influence behavior, TLIM= Intellectual Motivation TLIS= Intellectual Stimulation, OH= Openness/honesty, Cfe= Concern for employees, Comp= Competence, Iden=

Identification, Reli= Reliability, WbM1= Win-back management, PS= Product/Service customization, CIM= customer information management, PCI= Process Information, PI=

Product Information, FP= Financial Performance, ML= Market Leadership.

Items Loadings Items Loadings Items Loadings Items Loadings

TLIC .795 OH .755 WbM1 .866 FP .904

TLIIA .810 Cfe .876 PS .879 ML .922

TLIIB .891 Comp .772 CIM .800

TLIM .830 Iden .798 PCI .913

TLIS .862 Reli .789 PI .892

49

Testing Measurement Model

The structural model was tested by computing path coefficients using PLS tools.

According to Hair et al. (2011) the individual path coefficient in the PLS structural model can be interpreted as the standardized beta coefficient in ordinary regression. Moreover, each of the path coefficients’ significant value can be determined by the means of bootstrapping procedures.

The results of the multivariate table of the structural model are presented in the table 5.5. and are clearly configured in figure 5.1. The structural model as a whole explained 58%

of the variance in organizational trust, 40% in customer relationship management, 38% in innovation capabilities and 33% in business performance. According to Nunnally and Bernstein (1994) the explanatory power (R2) explains the dependent variables. In other words, transformational leadership can explain organizational trust and organizational trust can explain 40% of customer relationship management and customer relation management can explain 38% of innovation capabilities. Furthermore, transformational leadership, organizational trust, customer relationship management and innovation capabilities can explain 33% of business performance. These numbers can be considered large enough to prove the relevance of the model.

The results of PLS testing in table 5.5. show that all path coefficients for the variables are significant. Hair et al. (2011) stated the critical t-values for the two-tailed test are 1.65, 1.96 and 2.58 which represent weak, moderate and strong relationship among variables. The results show that there is a positive and significant effect between transformational leadership and organizational trust (β= 0.765, t=31.227, p<0.001) thus null hypothesis 1 was rejected.

Organizational trust also has a positive and significant effect towards customer relationship management (β=0.640, t=15.998, p<0.001) and customer relationship management and innovation capabilities (β=0.621, t=15.842, p<0.001) therefore null hypothesis 2 and 3 have also been rejected. The results also show that innovation capabilities have positive effects on business performance (β=0.575, t=12.775, p<0.001) therefore null hypothesis 4 has also been rejected. The results show that all of them have positive influences that rejected all null hypothesis.

50 Table 5.5.

PLS Path Analysis Results (Main Study, N=225) Path Hypothesis β path Adj.

t-value

Sig. Direction Null Hypothesis

TLOT H1 0.765 31.227 *** + Rejected

OTCRM H2 0.640 15.998 *** + Rejected

CRMIC H3 0.621 15.842 *** + Rejected

ICBP H4 0.575 12.775 *** + Rejected

Note. * p<.01. ** p<.05. *** p<.001.

TL= Transformational Leadership, OT= Organizational Trust, CRM= Customer Relationship Management,IC= Innovation Capabilities, BP= Business Performance.

51 Figure 5.1. PLS structural model (Main study, N= 225) Note. * p<.01. ** p<.05. *** p<.001.

52

The results of PLS findings show that transformational leadership has a significant positive effect on organizational trust, in which organizational trust has a significant positive effect on customer relationship management. The same situation occurs with customer relationship management and innovation capabilities. Furthermore, innovation capabilities have a significant and positive effect on business performance. This implies that all null

H1 Transformational leadership has no effect on organizational trust Rejected H2 Organizational trust has no effect on customer relationship

management (CRM).

Rejected H3 Customer relationship management has no effect on innovation

capabilities

Rejected H4 Innovation capabilities have no effect on organizational performance. Rejected

Focus only at the transformational leadership, the dominant factor of transformational leadership is idealized influence behavior. It follows by intellectual stimulation, inspirational motivation, idealized influence attributes and intellectual consideration. The dominant factor of organizational trust is concerned for employees and is followed by identification, reliability, competence and openness/honesty. Regarding the customer relationship management, the dominant factor is production/service customization, followed by win - back management and customer information management. The dominant factor of innovation capabilities innovation is process innovation followed by product innovation. Within business performance variable the most dominant factor is financial performance followed by market leadership.

A Comparative study of Taiwan and Bosnia & Herzegovina

By comparing the PLS results from the pilot study and the main study, there are some noteworthy points between the results. Within the transformational leadership, innovation capabilities and business performance, the dominant factor between two the results is different.

The dominant factor of transformational leadership in the pilot study was idealized influence attributes while it is idealized influence behavior in the main study. Regarding the innovation capacities, the dominant factor in the pilot study was product innovation while it is process innovation in the main study. The dominant factor of business performance was

53

market leadership while in the main study it is financial performance. Since these results are not consistent like the other factors, it is important to investigate further to give a better understanding of transformational leadership, innovation capabilities and business performance in this study.

The data gathered has different main characteristics that may explain the differences in results. The pilot study used 33 questionnaires from Taiwan. On the other hand, the total of 225 from the main study was from two location; 102 (45.3%) and 123 (54.7%) were collected from Bosnia & Herzegovina. By separating the data from these two countries, the data was then analyzed again by using the PLS method.

Table 5.7.

Taiwan (N=102) and Bosnia & Herzegovina (N=123) Cronbach’s Alpha results, Composite Reliability Results for All Dimensions

Cronbach’s Alpha Composite Reliability Constructs Number of

items

Note. TL= Transformational Leadership, OT= Organizational Trust, CRM= Customer Relationship Management, IC= Innovation Capabilities, BP= Business Performance.

Table 5.7. shows the R2 and AVE values for both countries. Regarding the results for Taiwan transformational leadership can explain 67% of organizational trust; organizational trust can explain 38% of customer relationship management. Furthermore, customer relationship management can explain 41% of innovation capabilities and innovation capabilities can explain 37% of business performance. Results for Bosnia & Herzegovina are as follows, transformational leadership can explain 49% of organizational trust, and organizational trust can explain 39% of customer relationship management. Moreover, customer relationship management can explain 32% of innovation capabilities and innovation capabilities can explain 26% of business performance. The table shows the results for AVE as well. According to Hair et al. (2011) the values for AVE should be greater than .50. All values are higher than 0.50 for both countries.

54 Table 5.8.

Taiwan (N=102) and Bosnia & Herzegovina (N=123) Measurement Model Results

R2 (%) AVE

Constructs Number of Items

Note. TL= Transformational Leadership, OT= Organizational Trust, CRM= Customer Relationship Management, IC= Innovation Capabilities, BP= Business Performance.

Table 5.8 shows the Cronbach’s Alpha and Composite Reliability results for Taiwan and Bosnia & Herzegovina. Cronbach’s Alpha is used to estimate the internal consistency and reliability of the study. If the Cronbach’s Alpha is higher than 0.7, the results can be accepted.

In this research study, all dimensions are higher than 0.7, thus it can be seen that all dimensions have high levels of internal consistency and reliability. The highest Cronbach’s Alpha value for Taiwan is transformational leadership while for Bosnia & Herzegovina it is organizational trust. Regarding the Composite Reliability the highest value for the Taiwanese sample is business performance while it is organizational trust for Bosnia & Herzegovina.

Tables 5.9. and 5.10. show PLS results the indirect effects for Taiwan and Bosnia &

Herzegovina. The highest indirect effect has transformational leadership on customer relationship management followed by the effect of transformational leadership on innovation capabilities for both countries.

Table 5.9.

PLS results for indirect effects between variables (Taiwan, N=102)

Taiwan

Note. TL= Transformational Leadership, OT= Organizational Trust, CRM= Customer Relationship Management, IC= Innovation Capabilities, BP= Business Performance.

55 Table 5.10.

PLS results for indirect effects between variables (Bosnia & Herzegovina, N=123) Bosnia & Herzegovina

Note. TL= Transformational Leadership, OT= Organizational Trust, CRM= Customer Relationship Management, IC= Innovation Capabilities, BP= Business Performance.

Following table provides the result for PLS testing for main study for both countries.

Taiwan and Bosnia & Herzegovina have some dominant factors in common. The results for Taiwan show that within transformational leadership the dominant factor is idealized influence attributes, while it is idealized influence behavior for Bosnia & Herzegovina. These two countries share the same dominant factor for organizational trust which is concern for employees. Regarding Taiwanese sample the dominant factor for customer relationship management is production/service customization while fit is win-back management for Bosnia & Herzegovina. These countries have similar results for dominant factors regarding innovation capabilities and business performance. The highest value is process innovation while market leadership is dominant factor for business performance.

Table 5.11.

PLS Path Analysis Result for Taiwan (102) and Bosnia and Herzegovina (123) Taiwan

56 Table 5.11. (continued)

Bosnia and Herzegovina (B&H)

Path Hypothesis β path Adj. t-value

Sig. Direction Null Hypothesis

TLOT H1 0.707 29.020 *** + Rejected

OTCRM H2 0.635 13.942 *** + Rejected

CRMIC H3 0.577 8.478 *** + Rejected

ICBP H4 0.520 8.971 *** + Rejected

Note. * p<.01. ** p<.05. *** p<.001.

B&H= Bosnia and Herzegovina, BP= Business Performance, CRM= Customer Relationship Management, IC= Innovation Capabilities, OT= Organizational Trust, TL= Transformational Leadership.

57 Table 5.12.

PLS loadings Taiwan (N=102), Bosnia & Herzegovina (N=123) Loadings

Items Taiwan Bosnia & Herzegovina (B&H)

TLIC 0.843 0.735

TLIIA 0.826 0.794

TLIIB 0.938 0.829

TLIM 0.857 0.797

TLIS 0.903 0.806

OH 0.730 0.761

Cfe 0.862 0.883

Comp 0.778 0.753

Iden 0.771 0.806

Reli 0.806 0.773

WbM1 0.886 0.847

PS 0.909 0.842

CIM 0.824 0.757

PCI 0.909 0.917

PI 0.903 0.876

FP 0.926 0.876

ML 0.937 0.907

Note. TLIC= Intellectual Consideration, TLIIA= Idealized influence attributes, TLIIB=

Idealized influence behavior, TLIM= Intellectual Motivation, TLIS= Intellectual Stimulation, OH= Openness/honesty, Cfe= Concern for employees, Comp= Competence, Iden=

Identification, Reli= Reliability, WbM1= Win-back management, PS= Product/Service customization, CIM= customer information management, PCI= Process Information, PI=

Product Information, FP= Financial Performance, ML= Market Leadership.

58 Figure 5.2. PLS structural model (Taiwan, N=102).

Note. * p<.01. ** p<.05. *** p<.001.

59

Figure 5.3. PLS structural model (Bosnia & Herzegovina, N=123).

Note. * p<.01. ** p<.05. *** p<.001.

60

PLS Results and Discussion

Results obtained from this chapter suggest several important aspects. First, the highest path coefficient in the TOKSIP Model is between transformational leadership and organizational trust (β=0.765). The dominance of this path in the overall model implies that transformational leadership is the key factor to organizational trust, customer relationship management, innovation capabilities and business performance. The more a leader is aware of transformational leadership style and the more he applies the best strategy in terms of idealized influence behavior and intellectual stimulation. According to the result, it will then make a positive contribution to the business performance.

Second, looking only at the transformation leadership variable, the dominant factor is idealized influence behavior. In comparison between Taiwan and Bosnia & Herzegovina the dominant factor remains the same. This suggests that the best way to create effective team is to provide followers with vision and encourage them to question things as well as to pursue their goals. Results also showed that intellectual stimulation has an important role in transformational leadership. It implies that leaders provide new and innovative solutions to their followers in problem solving. The other factors that also affect transformational leadership in descending order are: intellectual stimulation, idealized influence attribute and intellectual consideration.

Third, there is a positive and highly significant relationship between organizational trust and customer relationship management. This indicates that the higher concern for employees in the company, the better the customer relationship management. Looking at the organizational trust dimension, the highest value is the concern for employees followed by identification, reliability, competence and openness/honesty.

Fourth, there is a positive and highly significant relationship between customer relationship management and innovation capabilities. Production/service customization has the highest influence among the three aspects measuring customer relationship management.

This indicated that the more differentiation of the products or services according to customer needs the better customer relationship management will be. The dimension product/service customization is followed by: win-back management and customer information management.

The results showed that Bosnians value more win-back management while Taiwanese value production/service customization more.

Fifth, there is a positive and highly significant relationship between innovation capabilities and business performance. The results showed that process innovation has higher

61

value than product innovation. Furthermore, the results pointed out that process innovation is higher in both Taiwan and Bosnia & Herzegovina. This indicated that process innovation is relevant in a process service reaching the main goal, leading to higher business performance.

Sixth, the results show that financial performance is more important than market leadership. Financial performance measures the company’s capability to use assets in order to generate revenues. The results indicated that in both Taiwan and Bosnia & Herzegovina employees value more financial performance.

Seventh, looking at the comparisons between Taiwan and Bosnia & Herzegovina, the result implied that in both Taiwan and Bosnia & Herzegovina employees value more idealized influence behavior, concern for employees, process innovation and market leadership. The only variable where results are different is customer relationship management. Taiwanese employees evaluate more production/service customization while Bosnians evaluate more win-back management.

In Appendix are provided results for additional TOKSIP model (see Appendix B).

62

相關文件