Results of Reliability Test of Independent Variables Scale

Table 2.11: Results of reliability testing of independent variables scales

Factor

Correlate total variable	Cronbach's coefficient Alpha if variable type
Product Cronbach's Alpha = 0.884
1.1. The product has good, durable fabric and quality printing. Good	0.752	0.857
1.2. Diversity in designs and product types	0.804	0.810
1.3. Products are sewn according to design and size standards. according to customer requirements.	0.772	0.839
Price Cronbach's Alpha = 0.902
2.1. The purchase price is appropriate to the financial capacity of the organization.	0.794	0.869
2.2. Reasonable price compared to product quality.	0.808	0.864
2.3. Competitive price compared to other competitors in the market school.	0.776	0.875
2.4. Have appropriate discount and rebate policies.	0.755	0.886
Brand Cronbach's Alpha = 0.864
4.1. A reputable brand in the market and a supplier Good quality uniform products.	0.714	0.836
4.2. Brand known to many organizations/businesses arrive.	0.727	0.822
4.3 Is the brand that comes to mind first when you have an idea? order uniforms	0.787	0.768
Salesperson Cronbach's Alpha = 0.859
5.1. Sales staff have knowledge and understanding of the product. product	0.774	0.804
5.2. Sales staff are enthusiastic, friendly and happy to explain. answer customer questions	0.800	0.795
5.3. Staff are always ready to serve.	0.798	0.797
5.4. Professional staff.	0.774	0.805
Customer Care Cronbach's Alpha = 0.833
6.1. Good customer support service.	0.629	0.835
6.2. Timely and prompt support.	0.762	0.702
6.3. Warranty and repair services meet requirements.	0.694	0.767
Order Time Cronbach's Alpha = 0.791
7.1. Fast order response time.	0.610	0.742
7.2. Delivery on time as required.	0.705	0.637
7.3. Always update order progress.	0.597	0.766

Maybe you are interested!

(Source: SPSS data processing survey results)

The results of calculating Cronbanch's Alpha coefficient for the research factors show that the Cronbanch's Alpha coefficient of all factors is greater than 0.6.

The correlation coefficients of 23 independent observed variables are all greater than 0.3. Therefore, the scales of "Product", "Price", "Brand", "Sales staff", "Customer care", "Order time" are suitable and reliable. After the Cronbach's Alpha reliability analysis, 23 observed variables of 6 independent variables will be included in the EFA exploratory factor analysis.

Table 2 12: Results of reliability testing of dependent variable scale

Factor

Variable correlation total	Cronbach's coefficient Alpha if variable type
8. Purchase Decision Cronbach's Alpha = 0.911
8.1. You can rest assured when ordering products. uniforms at Lion Uniforms.	0.852	0.857
8.2. You will choose to purchase and use the product Lion Uniform products in the coming time.	0.861	0.840
8.3. You will introduce Lion Uniform for friends/partners in need of uniforms.	0.782	0.912

(Source: SPSS data processing survey results)

The dependent variable “Purchase Decision” consists of 3 observed variables. The result of Cronbach's Alpha coefficient for the factor “Purchase Decision” is 0.911, greater than 0.6, and the observed variables have total correlation coefficients greater than 0.3, so the variable “Purchase Decision” is suitable and reliable for further testing.

2.3.2. Exploratory Factor Analysis (EFA)

Through the reliability coefficient test above, because no variables were eliminated from the research model, the author continued to analyze the EFA for 6 independent variables and 1 dependent variable.

Exploratory factor analysis is used to reduce and summarize research variables into concepts. Through factor analysis, it is aimed at identifying and finding factors that represent observed variables. Exploratory factor analysis is based on standards and reliability.

Extracting factors affecting customers' purchasing decisions for Lion Uniform products is done by KMO coefficient (Kaiser Meyer-Olikin of Sampling Adequacy) and Bartlet's Test in which:

 KMO (Kaiser – Meyer – Olkin) is an index used to examine the appropriateness of EFA, 0.5 ≤ KMO ≤ 1 then factor analysis is appropriate (Hoang Trong & Chu Nguyen Mong Ngoc, Analyzing research data with SPSS, volume 2, page 31, year 2008).

 Bartlett's test of sphericity is a statistical measure used to examine the hypothesis that variables are not correlated in the population. If the Sig. of this test is less than or equal to 0.05, the test is statistically significant and the results of EFA analysis can be used (Hoang Trong & Chu Nguyen Mong Ngoc, Analyzing research data with SPSS, volume 2, page 30, 2008).

The Kaiser criterion is used to determine the number of factors extracted from the scale, to determine which Eigenvalue should be considered. The variance extracted criterion is used to determine whether factor analysis is appropriate.

2.3.2.1. Exploratory factor analysis EFA for independent variables

After testing the reliability of the scale and the appropriateness of the database,

At that time, exploratory factor analysis (EFA) was conducted.

The decision to purchase from customers is affected by many factors, so to research to find out which factors actually affect customers' decision to purchase, it is necessary to put 23 observed variables affecting customers' decision to purchase into EFA factor analysis.

Table 2.13: KMO and Bartlett's test for independent variables

Kaiser-Meyer-Olkin Measure of Sampling Adequacy.

0.826
Bartlett's Test of Sphericity	Approx. Chi-Square	1542,242
	df	190
	Sig.	0.000

(Source: SPSS data processing survey results)

The results of KMO and Bartlett's tests with KMO = 0.826, so factor analysis in this case is appropriate. The Sig. value of Bartlett's test = 0.000 < 0.05. From this, it can be seen that the observed variables are correlated with each other in the population.

Therefore, the data used for factor analysis is completely suitable. We have the summary results in the rotated matrix table below:

Table 2.14: Factor rotation matrix of independent variables

Variable name

Group of factors
1	2	3	4	5	6
NVBH3	0.866
NVBH2	0.854
NVBH1	0.850
NVBH4	0.811
GC2		0.858
GC3		0.827
GC1		0.819
GC4		0.818
SP2			0.877
SP3			0.875
SP1			0.852
TH3				0.878
TH2				0.859
TH1				0.775
CSKH2					0.874
CSKH3					0.778
Customer Service 1					0.764
CEO2						0.856
CEO1						0.801
CEO3						0.779
Eigenvalues	7,017	2,211	2,071	1,683	1,559	1,218
Extracted Variance %	35,087	11,056	10,355	8,415	7,794	6,091
Cumulative variance %	36,595	46,143	56,498	64,913	72,707	78,798

(Source: SPSS data processing survey results)

Eigenvalues represent the variation explained by each factor. Only factors with Eigenvalues greater than 1 are retained in the analysis model, and factors with Eigenvalues less than 1 are eliminated from the research model. This helps improve the reliability and accuracy of the scale.

Through the results of exploratory factor analysis, 6 factors with Eigenvalues = 1.218> 1 (Appendix 3) were extracted, satisfying the conditions. The total variance extracted is

78.798% > 50% (satisfies the condition) this proves that 78.798% of the variation in data is explained by 6 factors. All of the above factors meet the requirements because their loading factors are all greater than 0.5.

The first factor group “Sales staff” (NVBH3, NVBH2, NVBH1, NVBH4): Eigenvalue is 7.017 . This factor consists of 4 observed variables that are closely correlated with each other. This factor includes observed variables related to consultants and sales staff, this is the factor that explains 35.087% of the variation in the survey data.

The second group of factors “Price” (GC2, GC3, GC1, GC4): Eigenvalue equals 2.211, this factor consists of 4 observed variables that are closely correlated with each other. This factor includes observed variables related to price and pricing policy. This factor explains 11.056% of the variation in the survey data.

The third factor group “Product” (SP1, SP2, SP3): Eigenvalue is 2.071, this factor has 3 observed variables that are closely correlated with each other. This factor includes observed variables related to the product, this is the factor that explains 10.355% of the variation in the survey data.

The fourth factor group “Brand” (TH3, TH2, TH1): Eigenvalue is 1.559 , this factor includes 3 observed variables that are closely correlated with each other. This factor includes observed variables related to Brand, this is the factor that explains 7.794% of the variation in survey data.

The fifth factor group “Customer care” (CSKH2, CSKH1, CSKH3): Eigenvalue is 1.218, this factor consists of 3 observed variables that are closely correlated with each other. This factor includes observed variables related to the product, this is the factor that explains 7.794% of the variation in the survey data.

The sixth factor group “Order time” (TGDH2, TGDH1, TGDH3): Eigenvalue is 1.252 , this factor includes 3 closely correlated observed variables. This factor includes observed variables related to the product, this is the factor that explains 6.091% of the variation in the survey data.

2.3.2.2. Exploratory factor analysis EFA for dependent variable

Table 2.15: KMO and Bartllett's test of dependent variable

Kaiser-Meyer-Olkin Measure of Sampling Adequacy.

0.744
Bartlett's Test of Sphericity	Approx. Chi-Square	256,961
	df	3
	Sig.	0.000

(Source: SPSS data processing survey results)

With the Sig. value in Bartlett's test = 0.00 < 0.05, it shows that the observed variables are correlated with each other in the population, the KMO coefficient = 0.744 ≤ 1 meets the conditions, so factor analysis is appropriate for the sample data.

Table 2.16: Results of exploratory factor analysis for dependent variables

Purchase decision

Load factor
QDM1	0.939
QDM2	0.936
QDM3	0.899
Eigenvalues = 2.565 Total variance extracted = 84.496%

(Source: SPSS data processing survey results)

The analysis results show that there is only one extracted factor with Eigenvalues = 2.565 > 1 and the total extracted variance is 84.496%. The loading coefficients of the 3 observed variables are all greater than 0.5, so all variables are kept intact in the research model.

General comments: Through the EFA exploratory factor analysis process, the results show that 6 factors affecting customers' decisions to buy uniforms at Lion Uniform include: "Product", "Price", "Brand", "Sales staff", "Customer care", "Order time". Thus, the research model after EFA exploratory factor analysis is unchanged compared to the initial results, no variables are removed from the model during the process of testing the reliability of the scale and EFA exploratory factor analysis.

2.3.3. Correlation and regression analysis to measure the influence of factors on the decision to purchase uniform products at Lion Group Trading and Service Company Limited

2.3.3.1. Correlation analysis

Test the pair of hypotheses for pairs of independent variables and between independent variables and dependent variables: H 0 : Correlation coefficient is 0

H 1 : Correlation coefficient is different from 0

Table 2.17: Pearson correlation analysis

SP	GC	TH	NVBH	Customer Service	CEO	TH
SP	Correlation coefficient Pearson	1	0.325 **	0.338 **	0.303 **	0.354 **	0.232 *	0.617 * *
	Sig. (2 heads)		0.000	0.000	0.001	0.000	0.011	0.000
	N	120	120	120	120	120	120	120
GC	Correlation coefficient Pearson	0.325 **	1	0.363 **	0.383 **	0.490 **	0.237 **	0.631 * *
	Sig. (2 heads)	0.000		0.000	0.000	0.000	0.009	0.000
	N	120	120	120	120	120	120	120
TH	Correlation coefficient Pearson	0.338 **	0.363 **	1	0.374 **	0.264 **	0.320 **	0.638 * *
	Sig. (2 heads)	,000	,000		,000	,004	,000	,000
	N	120	120	120	120	120	120	120
NVBH	Correlation coefficient Pearson	0.303 **	0.383 **	0.374 **	1	0.408 **	,0375 **	0.559 * *
	Sig. (2 heads)	0.001	0.000	0.000		0.000	0.000	0.000
	N	120	120	120	120	120	120	120
Customer Service	Correlation coefficient Pearson	0.354 **	0.490 **	0.264 **	0.408 **	1	0.245 **	0.551 * *
	Sig. (2 heads)	0.000	0.000	0.004	0.000		0.007	0.000
	N	120	120	120	120	120	120	120
CEO	Correlation coefficient Pearson	0.232 *	0.237 **	0.320 **	0.375 **	0.245 **	1	0.516 * *
	Sig. (2 heads)	0.011	0.009	0.000	0.000	0.007		0.000
	N	120	120	120	120	120	120	120
QDM	Correlation coefficient Pearson	0.617 **	0.631 **	0.638 **	0.559 **	0.551 **	0.516 **	1
	Sig. (2 heads)	0.000	0.000	0.000	0.000	0.000	0.000
	N	120	120	120	120	120	120	120

(Source: SPSS data processing survey results)

Through the correlation test results shown in the table above, we have the following assessment: Hypothesis testing at the 5% significance level, so the Sig. value must be less than 0.05. According to the correlation coefficient matrix, we see that the independent variables " Product ", " Price ", " Brand ", " Sales staff ", " Customer care ", " Order time " all have Sig. values < 0.05 less than the significance level, rejecting the hypothesis H 0 shows that these variables are correlated with the dependent variable " Purchase decision ".

Besides, between independent variables with Sig. < 0.05, the independent variables may not have multicollinearity.

Thus, all independent variables “ Product ”, “ Price ”, “ Brand ”, “ Salesperson ”, “ Customer care ”, “ Order time ” can be included in the model to explain the fluctuations of the variable “ Purchase decision ”. In other words, these independent factors have an impact on customers' purchase decisions for Lion Uniform products.

2.3.3.2. Regression analysis

After completing the steps of exploratory factor analysis and correlation analysis, the next step is to proceed to the regression analysis step. Regression analysis is a statistical analysis to determine how independent variables determine dependent variables. The regression analysis model will describe the form of the relationship and thereby predict the value of the dependent variable when the value of the independent variable is known.

a. Building a regression model

The standardized regression equation for purchasing decisions based on factors has the form

as follows:

QDM = α + β 1 *SP + β 2 *GC + β 3 *TH + β 4 *NVBH+ β 5 *CSKH+ β 6 *TGDH

In there:

QDM: Dependent variable Purchase decision SP: Independent variable product

GC: Independent variable Price

TH: Independent variable Brand NVBH: Independent variable Brand