Factors Affecting the Developing Speed and Sustainability of Night-Time Economic Services in Bac Giang Province

: Until 2030, the province of Bac Giang will focus on promoting the comprehensive development of all economic sectors, with the industrial sector as the main driving force and the service sector as the motivating force. The expansion of concentrated industrial zones, coupled with the formation of new cities, has been a premise that fosters urbanization. These areas attract an abundant workforce of young immigrants and create conditions for the middle class to grow rapidly. That, along with the province’s orientation to develop tourism as an important economic sector, Bac Giang has the advantages and potentials to strongly promote night-time service economy. This article analyzes factors that influence the economic growth of night-time service activities in Bac Giang, establishes a basis for well-planned investment, as well as rapid and sustainable development of the service industry.


I. INTRODUCTION
This research utilizes exploratory factor analysis (EFA) to analyze the factors affecting the development of night-time economic services in Bac Giang through a regression model with the dependent variable being "Development of the night-time economic services in Bac Giang" and independent variables being: Infrastructure and public utility services; employee safety; human resources; policies; the innvolvement of the local community; destination management. Conclusions can be drawn about each factor's level of influence, providing effective and sustainable approaches for developing the night-time service economy in Bac Giang.

II. RESEARCH MODEL AND METHODOLOGY
In this study, the authors conducted a survey through a questionnaire to collect primary data. In addition, the author has researched a number of published scientific articles to serve the research process. For primary data collection, the author used a questionnaire paper and conducted a random survey of people in Bac Giang with the number of questionnaires distributed being 315, the number of valid answers collected being 230. According to Hair et al (2014), the minimum sample size to use EFA is 50, preferably 100 or more. The ratio of the number of samples to one analytic variable is at least 5:1, with the number of variables in the factor analysis model of this study being 35 variables, the number of samples needs to be at least 175, so with a sample of 230 is perfectly suitable for the analysis according to the EFA model.
Observations are included in the questionnaire on a 5-point Likert scale, in which, "1" is "strongly disagree"; "2" means "disagree", "3" means "no opinion", "4" means "agree" and "5" means "strongly agree". The proposed research model includes 6 factors and uses the Likert scale to consider the rating. For the selection of factors, the author uses the expert method, in which the experts have many studies in the fields of economics and services. The research model is shown in Figure 1 below. The dependent variable of the selected model is "Development of night-time economic services" which is measured by 7 observed variables: (1) Ensuring economic efficiency and competitiveness for enterprises; (2). Maximize visitor contributions; (3). Create jobs and raise income levels; (4) Maintain and enhance quality of life; (5) Contributing to preserving cultural values, craft villages, manners and customs of local people; (6) Contribute to the efficient use of resources; (7) Contributing to the conservation of natural resources. The independent variables are: (1) Infrastructure and public utilities (including 12 observed variables); (2) Safety for workers (including 8 observed variables); (3) Human resources (including 5 observed variables); (4) Policy (including 3 observed variables); (5) Participation of local communities (including 3 observed variables); (6) Destination management (including 4 observed variables).

III. REASEARCH RESULT AND DISCUSSIONS 1. Test the reliability of the measure using Cronbach's Alpha coefficient:
Cronbach's Alpha test is a test to analyze and evaluate the reliability of the measure. The purpose of this test is to find out if the observed variables have the same measure for a concept to be measured. The value of more or less contribution is analyzed through the correlation coefficient of the total variable (Corrected Item -Total Correlation), thereby allowing to eliminate inappropriate variables in the research model.
Evaluation criteria: Observable variables with a variable-to-total correlation coefficient (Item -Total Correlation) less than 0.3 will be eliminated and the standard for choosing a measure when Cronbach's Alpha > 0.6 (Hoang Trong & Chu Nguyen Mong Ngoc, 2000).
The results of Cronbach's Alpha analysis for the factors are summarized in the following table:

Source: Research team compiled from survey results
The results of testing the reliability of the factors show that most of the proposed factors have good reliability, Cronbach's Alpha values are all greater than 0.6. Particularly, the factor of the participation of local communities in the development has Cronbach's Alpha coefficient = 0.521, according to Hair et al (2014), although the reliability of this factor is not high, it is acceptable. All variables have Item -Total Correlation > 0.3 so all variables are kept.

EFA exploratory factor analysis for independent variables
The results of exploratory factor analysis EFA show that the test results are relatively good. KMO coefficient = 0.852 > 0.5, so the use of this data set for factor analysis is appropriate (Kaiser, 1974). Bartlett's Test is used to see if the observed variables in the factor are correlated, this value is also achieved when the Sig value of the test = 0.000 < 0.5, so the observed variables are relevant and were eligible for factor analysis by EFA (table 2). The breakpoint of factor analysis is set on the basis of the Eigenvalue coefficient, the maximum number of factors is selected when this coefficient has the smallest value > 1 and the cumulative percentage is greater than 50%. With the above conditions, there are 6 factors extracted at Initial Eigenvalues of 2.069 > 1, the total variance extracted is 62.822% > 50%, showing that 6 factors extracted in EFA reflect 62.8220% of the difference. Variations of all measures are included in the model. The final results of factor analysis exploring the measures of independent variables including 35 observed variables are loaded with 6 factors, the order and names of the factors are summarized in Table 4.

EFA exploratory factor analysis for dependent variable
The results of the first EFA discovery factor analysis for the dependent variable, a quick look at the Rotated Component Matrix table shows that the dependent variable does not converge on one factor. There are 2 factors extracted, in which there is a problem in the observed variables NTED1; NTED2; NTED 7 is uploaded in two groups of variables 1 and 2. At the same time, the variable NTED7 does not download the desired factor at all.
Review the variable-total correlation in the reliability test Cronbach's Alpha coefficient of the measure of economic development = 0.341 > 0.3, however, the coefficient of correlation of this variable -total is quite low compared to other variables in the factor of night-time economic services development. So the author decided to remove the variable NTED7 and run the EFA a second time.
The results of running EFA for the second time after removing the variable NTED7 show that the test results are relatively good. KMO coefficient = 0.89 > 0.5, so it is appropriate to use this data set for factor analysis (Kaiser, 1974). Bartlett's Test also passes when the Sig value of the test = 0.000 < 0.5, so the observed variables are related to each other and are eligible for factor analysis by EFA (table 5). The coefficient Initial Eigenvalues = 5.214 > 1 proves that the extracted factor in Table 5 is representative of the variation with uniform information. The total variance extracted is 58.235% > 50%, which proves that the factor extracted in EFA reflects 58.235% of the variation of all measures included in the model. Combining the results of testing the reliability of the measure (Cronbach's Alpha) and exploratory factor analysis (EFA), the study has identified factors, variables and measures that can be used to evaluate the factors. affecting the rapid and sustainable development of the night-time economy of Bac Giang province.

Regression analysis of factors affecting night-time economic development in Bac Giang province
To test the research hypotheses, the author performed multiple linear regression of 6 variables of 6 factors proposed in the adjusted research model after qualitative research. The selection method is Enter and the regression results are as follows: The result of the regression model synthesis coefficient in Table 2.13 for adjusted R2 (Adjusted R Square) = 0.653 means, with the sample size of the research model of 85, there is 65.3% variation of the dependent boundary ( N ight economy development) is due to the influence of 6 variables included in the above research model. The rest, 34.7% is influenced by factors other than the research model. Statistical value Durbin-Watson = 1.792 < 3, shows that the model has no autocorrelation, therefore, the quality of regression can be considered as good.

Source: Analysis of survey data by the research team
The results of the study in Table 9 show that the acceptance coefficient (Tolerance) is quite high (from 0.621 to 0.849), and at the same time, the variance exaggeration coefficient VIF is low (from 1.021 to 1.223, less than 2). Therefore, it can be concluded that the relationship between these independent variables is not significant, it can be considered that there is no multicollinearity phenomenon in the model.

IV. CONCLUSION
With the aforementioned analysis results, it can be seen that elements having a profound impact on the night-time economic service development in Bac Giang are the "Employee safety" and "Infrastructure and public utility services". Accordingly, the issues that the local administration and people of Bac Giang province need to pay attention to, for the developing speed and sustainability of the night-time service economy here include: Infrastructure such as traffic systems, lighting, sanitation system, information technology system, entertainment system, food chains and shopping facilities; safety for the community such as noise reduction, alcohol use regulations, security and order, air quality, warning of crime risks, conservation and promotion of spiritual life for people. This study is also an important basis for Bac Giang province to implement new policies to support and promote the rapid and sustainable night-time economic service development.