Testing Hypothesis with Linear Regression

Một phần của tài liệu (Luận văn) determinants of adidass supply chain performance in asia pacific region 001 (Trang 64 - 77)

Chapter 5: Empirical results of the research

5.4. Testing Hypothesis with Linear Regression

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

Table 5.1 Descriptive statistics results

Variables Mean Std.

Deviation SSP1. Your company shares your strategies to your key

partners by every fiscal year. 4.61 0.575

SSP2. Your company includes your key partners in your

planning and goal-setting activities 4.51 0.576 SSP3. Your company is fully aware of your partner‟s

strategies. 4.58 0.561

SSP4. Your company regularly solves problems jointly

with your partners 4.52 0.558

SSP5. Your company helps your partners to improve their

product quality and price 4.51 0.54

SSP6. Your company considers service, quality and price

as criterion in selecting partners 4.51 0.54

SSP7. Your company share mutual benefits with your

partners 4.51 0.558

SSP8. Your partners‟ strategies and your companys‟

Strategies impact much to your company‟s SCM and finance performance

4.53 0.539 PC1. Your company pro-actively supports your partners

in new product development processes 4.52 0.558 PC2. Your company gets and provides all product

specifications details/requirements from your partners in time

4.38 0.572 PC3. Your company capacity can satisfy the time line of

product creation to your customers 4.26 0.661

PC4. Your company is able to adapt to the changing consumer preferences more quickly, flexibly and efficiently

4.37 0.561 PC5. Your company‟s product creation impacts much to

your SCM and finance performance 4.38 0.527

PSP1. Your company‟s purchasing/supply planning is

more competitive than other competitors 4.53 0.539 PSP2. Your company‟s purchasing/supply planning can

help to reduce inventory management cost, lead time and logistics cost

4.26 0.681 PSP3. Your company‟s purchasing/supply planning plays

a key liaison role to your customers in creating and delivering value

4.37 0.542 PSP4. Your partners highly evaluate your company‟s

purchasing/supply planning 4.41 0.551

PSP5. Your company‟s purchasing/supply planning impacts much to your company SCM and finance performance

4.38 0.545 SP1. Sourcing production is one of your company‟s key

strategies 4.37 0.587

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

SP2. Your company‟s production capability can meet

your customers‟ expectations 4.29 0.563

SP3. Your company‟s sourcing production is aligned with

your partner‟s strategy 4.32 0.617

SP4. Your sourcing production is working effectively in terms of cost reduction, quality, service and delivery improvement and flexibility

4.47 0.549 SP5. Your company‟s sourcing production impacts much

to your company SCM and finance performance 3.97 0.685 ILP1. Your company is effectively implementing lead-

time reduction and wastes elimination program from product development and commercialization to mass production

4.43 0.59

ILP2. Your company is effectively implementing lean

program to eliminating wastes and improving efficiency 4.54 0.583 ILP3. Your company is using less input to produce at a

mass production speed, while offering more variety to the end customers

4.45 0.538 ILP4. Your company strongly pushes partners to shorten

lead-times and improve quality 4.55 0.538

ILP5. Your company‟s Internal Lean Practice impacts

much to your company SCM and finance performance 4.54 0.539 SCP1. Your company delivers products with quality up to

and over expectation of the standard from your customers 4.45 0.556 SCP2. Your company delivers products to customers on

time 4.54 0.538

SCP3. Your company delivers accurate volume of product

to customers 4.48 0.673

SCP4. Your company delivers products to customers at

minimum cost 4.54 0.583

SCP5. Your company achieves good business results this

year 4.57 0.536

SCP6. Your customers are very satisfied with your

services 4.42 0.63

SCP7. Your SC is very flexible and be able to adapt to the

customer various changing needs 4.36 0.611

5.2. Exploration Factor Analysis

Exploration factor analysis (EFA) can be highly useful and powerful multivariate statistical technique for effectively extracting information from large bodies of interrelated data. The primary purpose of exploration factor analysis is keyed to four issues: specifying the unit of analysis; achieving data summarization and/or data reduction; variable selection; and using factor analysis results with other multivariate

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

techniques (Hair et al, 2006). In this research, EFA is used for the purpose of data reduction achieved by identifying representative variables from a much larger set of variables for use in subsequent multivariate analyses. The nature and character of the original variables are retained, but their number is reduced to simplify the subsequence multivariate analysis.

“Principle component analysis” and “Varimax” is used as extraction method and rotation method respectively. “A priori criterion” is used to predetermine the number of factors based on research objective and prior research. It means the five initial factors are retained for the factor analysis without change because the purpose of this research is to study the relationship of these factors but finding new factors. However, numbers of items in each factor are modified during the factor analysis based on two criterions as following:

- Value of factor loading equal or over 0.5 are general considered necessary for practical significance (Hair et al, 2006).

- Variable should generally have communalities of greater than 0.5 (Hair et al, 2006).

Factor analysis attempts to identify underlying variables, or factors, that explain the pattern of correlations within a set of observed variables. Factor analysis is often used in data reduction to identify a small number of factors that explain most of the variance observed in a much larger number of manifest variables. Factor analysis can also be used to generate hypotheses regarding causal mechanisms or to screen variables for subsequent analysis (for example, to identify collinearity prior to performing a linear regression analysis).

Table 5.2 showed the factor analysis result with factor loading values from the Rotated Component matrix as below. Principal component analysis worked on the initial assumption that all variance is common; therefore, before extraction the communalities were all 1.

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

Table 5.2 Factor loading analysis results

Rotated Component Matrix

Component

1 2 3 4 5 6 Item deleted

ILP1 0.73 0.27 0.09 0.33 0.05 0.02 ILP2 0.70 0.49 0.11 0.22 0.27 -0.15 ILP3 0.78 0.26 0.06 0.16 -0.08 0.07 ILP4 0.74 0.26 0.09 0.24 0.22 0.16 ILP5 0.82 0.30 0.05 0.26 0.09 0.07 SCP1 0.61 0.27 0.09 0.25 0.10 0.54 SCP2 0.74 0.28 0.05 0.18 0.13 0.05 SCP3 0.80 0.32 0.09 -0.14 0.31 0.05 SCP4 0.71 0.48 0.11 0.23 0.25 -0.15 SCP5 0.81 0.37 0.06 0.24 0.19 0.09

PC1 0.45 0.59 0.08 0.52 0.06 -0.27 PC1 PC2 0.48 0.58 0.12 0.39 0.26 0.01 PC4 0.42 0.55 0.08 0.45 0.25 -0.06 PC5 0.29 0.91 0.00 0.13 0.12 0.10 PSP1 0.43 0.58 0.04 0.54 0.04 -0.27 PSP1 PSP3 0.26 0.80 0.10 0.16 0.17 0.07 PSP4 0.27 0.81 0.03 0.15 0.09 0.06 PSP5 0.31 0.90 0.05 0.13 0.14 0.09 SCP7 0.35 0.52 0.00 0.32 0.20 -0.19

SP1 0.35 0.88 0.12 0.15 0.15 0.06 SP1 SP4 0.29 0.83 0.01 0.12 0.04 0.15 SP4 SSP1 0.24 -0.01 0.65 0.01 -0.02 -0.22 SSP2 0.07 -0.03 0.89 0.09 0.11 0.08 SSP3 0.03 0.05 0.69 -0.06 -0.12 -0.14 SSP4 0.03 -0.01 0.88 0.04 0.16 0.06 SSP5 0.00 0.04 0.86 0.01 0.00 0.06 SSP6 0.07 0.10 0.85 0.05 0.00 0.07 SSP7 0.00 0.09 0.84 0.06 0.01 -0.07 SSP8 0.04 0.08 0.80 0.01 0.05 0.14 SP2 0.23 0.30 -0.07 0.69 0.06 0.34 SP3 0.29 0.25 0.17 0.67 0.02 -0.09 SP5 0.29 0.16 0.00 0.79 0.14 0.21 PC3 0.37 0.38 0.05 0.15 0.80 0.05 PC4 PSP2 0.38 0.39 0.08 0.15 0.79 0.04 PSP2 SCP6 0.58 0.23 0.02 0.24 0.04 0.59 SCP6

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

Extraction Method: Principal Component Analysis.

Rotation Method: Varimax with Kaiser Normalization.

a. Rotation converged in 7 iterations.

Initial communalities are estimates of the variance in each variable accounted for by all components or factors. Extraction communalities are estimates of the variance in each variable accounted for by the factors (or components) in the factor solution. With the given results that there are 7 items including PC1, PC4, SP1, SP4, PSP1, PSP2, SCP6 should be deleted to increase the reliability and validity of the measures even though these 7 deleted items totally satisfied the two above criterion but due to their complex structure as well as convergent problem. All other items were able to be used for subsequent analysis.

As we know that larger sample size, higher communalities (low communalities are associated with sampling error due to the presence of unique factors that have nonzero correlations with one another and with the common factors), and high overdetermination [each factor having at least three or four high loadings and simple structure (few, nonoverlapping factors)] each increase our chances of faithfully reproducing the population factor pattern. With communalities moderate (about .5) and the factors well-determined, we should have 100 to 200 subjects. That is suitable with the current samples.

5.3. Reliability analysis:

In this stage, reliability analysis is used to assess the degree of internal consistency between multiple measurement items of the constructs. To assess the internal consistency, below criterions are used.

1. The item-to-total correlation exceeds 0.5 and inter-item correlation exceeds 0.3. (Hair et al, 2006).

2. Cronbach‟s Alpha exceeds 0.7. (Hair et al, 2006).

Reliability values indicate the degree to which operational measures are free from random error and measure the construct in a consistent manner. Reliability is typically assessed using Cronbach‟s alpha values. The Cronbach a coefficient (Cronbach, 1951)

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

has been used to evaluate reliability. A scale is found to be reliable if the coefficient is 0.70 or higher (0.80) for confirmatory factor analysis (Nunally, 1978).The reliabilities of the scales are shown in Table 5.3. All values exceed the recommended cutoff point of 0.80.

The numbers of item in each construct are modified if these criterions are not complied. All the constructs got high cronbach‟s alpha values which are above 0.8 (SSP=0.927, PC=0.842, PSP=0.920, SP=0.802, ILP=0.927, SCP=0.914) indicating good reliability and that the items for each construct hang well together as well as satisfactory item-to-total correlations (all>0.5) and inter-item correlations (all>0.3). No modification was required. The retained items and constructs were used for subsequent analysis.

Table 5.3 Results of Exploration Factor Analysis

Construct/Variable

Cronbach’s Alpha

Item-total correlation (>=0.7) (>0.5)

Strategic Supplier Partnership 0.927

Inter-Item Correlation Matrix (>0.3)

SSP1 SSP2 SSP3 SSP4 SSP5 SSP6 SSP7 SSP8

SSP1 1 0.594

SSP2 0.512 1 0.846

SSP3 0.524 0.533 1 0.612

SSP4 0.478 0.922 0.515 1 0.837

SSP5 0.465 0.726 0.535 0.749 1 0.802

SSP6 0.455 0.734 0.526 0.741 0.732 1 0.798

SSP7 0.535 0.679 0.492 0.668 0.707 0.699 1 0.782

SSP8 0.514 0.67 0.454 0.657 0.645 0.654 0.682 1 0.746

Product Creation 0.842

Inter-Item Correlation Matrix (>0.3)

PC2 PC3 PC5

PC2 1 0.766

PC3 0.647 1 0.663

PC5 0.727 0.582 1 0.715

Purchasing/Supply Planning 0.920

Inter-Item Correlation Matrix (>0.3)

PSP3 PSP4 PSP5

PSP3 1 0.806

PSP4 0.715 1 0.810

PSP5 0.829 0.833 1 0.898

Sourcing Production 0.802

Inter-Item Correlation Matrix (>0.3)

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

SP2 SP3 SP5

SP2 1 0.652

SP3 0.417 1 0.545

SP5 0.727 0.581 1 0.772

Internal Lean Practice 0.927

Inter-Item Correlation Matrix (>0.3)

ILP1 ILP2 ILP3 ILP4 ILP5

ILP1 1 0.807

ILP2 0.666 1 0.777

ILP3 0.718 0.63 1 0.782

ILP4 0.667 0.738 0.623 1 0.778

ILP5 0.823 0.754 0.83 0.756 1 0.91

Supply Chain Performance 0.914

Inter-Item Correlation Matrix (>0.3)

SCP1 SCP2 SCP3 SCP4 SCP5 SCP7

SCP1 1 0.668

SCP2 0.559 1 0.757

SCP3 0.588 0.696 1 0.792

SCP4 0.609 0.73 0.796 1 0.913

SCP5 0.71 0.782 0.819 0.835 1 0.876

SCP7 0.422 0.444 0.445 0.764 0.480 1 0.583

Descriptive statistics for each measurement variables which were also reported in Table 5.4 including means and standard deviations, indicating the variable means to be below 5. The standard deviations for the variables range from 0.38 to 0.42 indicating considerable amount of variation in the responses.

Table 5.4 Means and standard deviations Descriptive Statistics

Mean Std. Deviation

SSP 4.25 0.40

PC 3.76 0.39

PSP 3.79 0.38

SP 3.65 0.40

ILP 4.09 0.41

SCP 4.14 0.42

5.4. Testing Hypothesis with Linear Regression

The regression analyses are to determine the strength of the relationship between independent variables and dependent variables. Beside that, in regression analyses an

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

equation can be created. This regression equation allows prediction of values of the dependent from given values of the independent.

In this section, the linear regression analysis was employed to test the hypotheses H1, H2, H3, H4 and H5. In which, dependent variable was SCP and five independent variables were Strategic supplier partnership, Product creation, Purchasing/Supply planning, Sourcing production, Internal Lean practice. In addition, correlation analysis also was conducted to test discriminant validity.

Table 5.5. The correlations of factors of SC

SSP PC PSP SP ILP

SSP

Pearson Correlation Sig. (2-tailed)

1

PC

Pearson Correlation Sig. (2-tailed)

.154* 1

PSP

Pearson Correlation Sig. (2-tailed)

.137 .861** 1

SP

Pearson Correlation Sig. (2-tailed)

.109 .594** .524** 1

ILP

Pearson Correlation Sig. (2-tailed)

.179* .765** .660** .633** 1

The results of the correlations among the factors of SC as presented in Table 5.5 prelimarily allowed us to conclude that these independent variables could be put into the model to explain the dependent variable of SCP. Composite scores for each study variable were calculated by averaging scores across items representing that construct.

Table 5.5 demonstrated the correlation coefficients among study variables. The correlation coefficients ranged from 0.109 to 0.861. None of the correlation coefficients were equal to and/or above 0.90, providing empirical support for discriminant validity.

Based on the results of the multi-linear regression analysis on data of 199 responses, R square was 0.855 and the adjusted R square was 0.851 meaning that the regression line fits the data with no major difference between them (no big shrinkage indications) and can show the observed variation with the close relationship between predictors and dependent variable. The multi linear regression model is fit to population data and usable.

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

Table 5.6. R Square and Adjusted R square Model Summaryb

Model R R Square Adjusted R Square Std. Error of the Estimate

1 .925a .855 .851 .16166

a. Predictors: (Constant), ILP, SSP, SP, PSP, PC b. Dependent Variable: SCP

ANOVAb

Model Sum of Squares df Mean Square F Sig.

1 Regression 29.721 5 5.944 227.449 .000a

Residual 5.044 193 .026

Total 34.765 198

a. Predictors: (Constant), ILP, SSP, SP, PSP, PC b. Dependent Variable: SCP

Table 5.7 was showing the results of linear regression analysis with the following details analysis.

Table 5.7 Results of linear regression analysis Coefficients

Model Unstandardized

Coefficients Standardized

Coefficients t Sig.

B Std.

Error Beta

1 (Constant) 0.275 0.191 3.442 0.0

SSP 0.004 0.029 0.004 0.135 0.893

PC 0.289 0.069 0.266 4.189 0.0

PSP 0.136 0.107 0.136 1.275 0.045

SP -0.042 0.038 -0.039 -1.086 0.279

ILP 0.715 0.047 0.699 15.325 0.0

a. Dependent Variable: SCP

H1: Strategic supplier partnership is not related to Supply Chain Performance.

This hypothesis suggests that Strategic supplier partnership impacts much to SCP.

The results of the linear regression in Table 5.5 showed that the Unstandardized coefficient beta between Strategic supplier partnership and SCP was fairly small and

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

non significant (B = 0.004, t = 0.135, sig. =0.893), demonstrating that hypothesis 1 was not supported by the data. This finding leads to the rejection of hypothesis 1. According to this result, there is no significant relationship between Strategic supplier partnership and SCP, which is an unexpected result, according to the previous literature review on the relationship between Strategic supplier partnership and SCP. It can be explained that adidas‟s T1 and T2 suppliers much more depends on adidas‟s strategy; moreover, there are a wide alternative selection of these suppliers in shoes industry so that from T1 and T2 representatives think that this kind of business relationship is buyer-supplier and the suppliers have to do their best to serve the buyers as much as they can to compete each other and it possibly affect the survey results.

Hypothesis 2: Product creation is positively related to Supply Chain Performance.

Hypothesis 2 posits that product creation is one of the key factors of SCP. From the linear regression analysis results in Table 5.5 indicated that the Unstandardized coefficient beta between product creation and SCP was (B= 289, t =4.189, sig. =0.00), demonstrating that hypothesis 2 was supported by the data. This result confirms a positive relationship between product creation and SCP and indicates that production creation is one of key determinants in whole supply chain. This result will be discussed further in the next final chapter.

Hypothesis 3: Purchasing/Supply planning is positively related to Supply Chain Performance.

Hypothesis 3 predicts that Purchasing/Supply planning is positively related to SCP. From the regression results in table 5.5, the Unstandardized coefficient beta between Purchasing/Supply planning and SCP was (B= 136, t =1.275, sig. =0.045), demonstrating that hypothesis 3 was supported by the data. This result confirms a positive relationship between Purchasing/Supply planning and SCP and indicates that Purchasing/Supply planning is one of key determinants in whole supply chain. This result will be discussed further in the following chapter.

Hypothesis 4: Sourcing production is not related to Supply Chain Performance.

From hypothesis 4, Sourcing production is expected to have positive influence on SCP.

As shown in Table 5.5 that the Unstandardized coefficient beta between Sourcing

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

production and SCP was negative, fairly small and non significant (B=-0.042, t = - 0.1086, sig. =0.279), demonstrating that hypothesis 4 was not supported by the data.

This finding leads to the rejection of hypothesis 4. According to this result, there is no significant relationship between Sourcing production and SCP, which is an unexpected result, according to the previous literature review on the relationship between Sourcing production and SCP.

With the main production sourcing in Asia, it seems that this region will be the main focus area in next several years in terms of very competitive labour cost and other competitive advantages and all T1 and T2 suppliers now tend to expand their bases within this region and it will not affect much to adidas‟s supply chain in this region.

Hypothesis 5: Internal Lean practice is positively related to Supply Chain Performance

According to Hypothesis 5, it suggests that Internal Lean practice is one of the key determinants of SCP. From the linear regression analysis results in Table 5.5 indicated that the Unstandardized coefficient beta between product creation and SCP was (B= 715, t =15.325, sig. =0.000), demonstrating that hypothesis 5 was supported by the data. This result confirms a positive relationship between Internal Lean practice and SCP and indicates that Internal Lean practice is one of key determinants in whole supply chain. This result will be discussed further in chapter 6.

In summary, the hypotheses of this research are stated as below:

Hypothesis 1: H1 is rejected on research sample set. It is concluded that the strategic supplier partnership is not related to SCP.

Hypothesis 2: H2 is accepted on research sample set. It is concluded that production creation is positively related to SCP.

Hypothesis 3: H3 is accepted on research sample set. It is concluded that Purchasing/supply planning is positively related to SCP.

Hypothesis 4: H4 is rejected on research sample set. It is concluded that sourcing production is not related to SCP.

Hypothesis 5: H5 is accepted on research sample set. It is concluded that internal lean practices is positively related to SCP.

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

Figure 5.2.The revised model of determinants of Supply chain performance.

Product creation

Purchasing/Supply planning

Internal Lean practice

Supply Chain Performance

H2 H3

H5

(B 71 5, t

=1 5.3 25, sig .

=0.

00 0),

tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg

CHAPTER 6

Một phần của tài liệu (Luận văn) determinants of adidass supply chain performance in asia pacific region 001 (Trang 64 - 77)

Tải bản đầy đủ (PDF)

(95 trang)