Building a Research Model on Factors Affecting the Effectiveness of Internal Control System in Social Insurance Collection

Qualitative research

Maybe you are interested!

• Building a Research Model of Factors Affecting Agribank's Brand Value
• Building a Scale and Research Model of Factors Affecting Customers' Decision to Choose a Bank to Deposit Savings at
• Initial Research Model on Factors Affecting the Internal Control System on Social Insurance Collection
• Factors affecting the effectiveness of internal control system in social insurance collection at Binh Duong Provincial Social Insurance - 19
• Factors Affecting the Management of Compulsory Social Insurance Collection

-Systems approach

- Method of synthesis, analysis and comparison

-Expert Interview

Draft scale

Preliminary research Group discussion Expert consultation

Mock interview

Quantitative research

Official scale

Calibrate the scale

Scale verification

Cronbach's Alpha Test

Factor analysis

EFA

Check Variance Extraction Type of variables with small EFA weights

Check the extracted factors

Adjust the model

Check the suitability of the research model

Test the assumptions

study

Regression analysis

multivariate

Propose solutions

Figure 3.1 : Research design diagram

3.2. Research method

3.2.1. Qualitative research methods

Systemic approach: this method is used to study domestic and foreign research, documents, research works published in the form of books, theses, dissertations, articles published in scientific journals, etc. in the last 20 years.

Method of synthesis, analysis and comparison: The author will synthesize the factors studied in domestic and foreign documents. From there, proceed to analyze and compare the differences.

Expert interviews: The author conducted in-depth interviews with 05 experts (including directors, deputy directors and department heads) with experience in the field of internal control to explore specific topics revolving around identifying factors affecting the internal control system on social insurance collection. From there, the author proceeded to collect as much information as possible on the topic under study.

The information collection is carried out through a questionnaire prepared by the author in advance to consult experts on the factors affecting the effectiveness of the internal control system in the collection of social insurance. After synthesizing expert opinions, the author will discuss with the experts to reach a high consensus on the addition, adjustment and retention of factors that are considered most suitable for the specific characteristics of Binh Duong. The adjusted questionnaire will be emailed to the experts for consultation once again to check and agree on the language of presentation before sending to the units for survey. All observed variables in the questionnaire (scale components) use a 5-point Likert scale with the choice of number 1 being completely disagree with the statement and the choice of number 5 being completely agree with the statement.

3.2.2. Quantitative research methods

a. Building a scale

The scales used in this study are based on previous studies and existing theories in the country and the world. These scales are adjusted and supplemented to suit the actual characteristics of Binh Duong province. Based on qualitative results with expert opinion techniques, there are 7 scales in this study, respectively as follows:

- Control environment scale

The control environment is denoted as MTKS and is measured by the following 7 observed variables:

MTKS 1: Honesty, ethical values ​​and working capacity of social insurance officers and employees

MTKS 2: Management capacity and operating style, ethics, behavior and work efficiency of leaders

MTKS 3: Binh Duong Provincial Social Insurance handles records within the prescribed time limit MTKS 4: How to establish authority and responsibility as well as organizational capacity

organization and development of work in the unit

MTKS 5: Directions and instructions of the leadership in performing the duties of cadres and civil servants

MTKS 6: Transparency in activities regarding the control of revenue and expenditure of social insurance regimes MTKS 7: Personnel recruitment policy focuses on expertise and ethics

labor.

- Risk assessment scale

Risk assessment is denoted as DGRR and is measured by 8 observed variables.

after:

DGRR 1: Personnel arrangement of the Board of Directors in professional and technical matters DGRR 2: The Board of Directors' prudence in considering and resolving issues

potential risks at the unit

DGRR 3: Measures and solutions of the Board of Directors when potential risks occur at the unit

DGRR 4: Warning from the Board of Directors of the Social Insurance Agency to its affiliated units to prevent risks

DGRR 5: Periodic reporting regime on risk assessment of units under the Social Insurance agency

DGRR 6: Develop risk assessment processes at the unit DGRR 7: Have risk assessment methods

DGRR 8: Propose measures to manage risks.

- Control activity scale

Control activities are denoted as HDKS and are measured by 7 important variables.

close:

HDKS 1: To control each other, divide responsibilities clearly and fully

HDKS 2: Implementation of social insurance policies and laws in management activities

social insurance fund

HDKS 3: Understand the professional regulations on social insurance law of the Board of Directors.

HDKS 4: Approval of decisions and guiding documents on social insurance law by leaders at all levels

HDKS 5: Application of information technology in social insurance fund management activities

HDKS 6: Annual inspection and examination plan on the implementation of regulations on social insurance, health insurance, and unemployment insurance

HDKS 7: Plan for rotating ministries and officials in the Social Security agency.

- Information and Communication Scale

The Information and Communication Scale is denoted as TTTT and consists of 7 important variables.

close:

time.

TTTT 1: Update information accurately and promptly

TTTT 2: Information is provided in the right place, with enough details, presented appropriately and promptly.

Information 3: Receiving complete and accurate information from superiors to subordinates Information 4: The Board of Directors' interest in developing the information system Information 5: Information from outside must be received completely and honestly

TTTT 6: Ensure communication between functional departments and direct units

Binh Duong Social Insurance

TTTT 7: Monitor information feedback.

- Monitoring scale

The Monitoring Scale is denoted as GS and consists of 6 observed variables:

GS 1: Results of inspection and examination of social insurance, health insurance, unemployment insurance for enterprises participating in social insurance

GS 2: Management's supervision of the internal control department's performance and situation

GS 3: Remedial measures when detecting weaknesses and limitations during the inspection and examination process

GS 4: Cooperation between agencies and sectors in inspection and examination of the implementation of social insurance laws

GS 5: Measures to handle when there are conclusions about signs of violations during inspection, examination and supervision

GS 6: Consent of organizations and individuals in the implementation of inspection, examination and supervision.

- Scale of awareness of participating in social insurance

The Social Insurance Participation Awareness Scale is denoted as YTTG and includes 4 observed variables:

YTTG 1: Employers are aware of the importance of social insurance participation for employees

YTTG 2: Workers are aware of the importance of participating in social insurance for their own benefit.

YTTG 3: Does the Social Insurance Agency inspect and check the participation of employees in social insurance?

YTTG 4: Does the Social Insurance Agency regularly disseminate to employers and employees the importance of participating in social insurance?

- Scale of income control ability

The Control Ability Scale is denoted as KSNB and is measured by the following four observed variables:

KSNB 1: Organizing the internal control system of Binh Duong Social Insurance always has greater benefits than costs.

KSNB 2: Binh Duong Social Insurance's internal control system satisfies all implementers

KSNB 3: The internal control system of Binh Duong Social Insurance has achieved the goals set by the collective.

KSNB4: The internal control system of Binh Duong Social Insurance has minimized management risks.

b. Methods of measuring and calculating data

Quantitative research method is applied in the process of synthesizing observed variables and a group of factors as well as the relationship between each group of factors affecting the effectiveness of the internal control system in the collection of social insurance of Binh Duong Provincial Social Insurance.

Descriptive statistical method: This method helps the author to make objective assessments. The author builds a survey questionnaire (based on the research process combined with opinions from experts) and sends it to collection specialists and social insurance officers and employees to conduct the survey. From the survey results, conduct statistics and consolidate his assessment views.

Scale quality testing : use Cronbach's Alpha test to determine the quality of the constructed scale. According to Nunnally and Bernstein (1994), a scale has good reliability when it varies in the range [0.70-0.80]. If Cronbach's Alpha

≥ 0.60 is an acceptable scale in terms of reliability. And a measurement variable with a total item correlation coefficient (adjusted) ≥ 0.30 meets the requirements. (Quoted by Nguyen Dinh Tho, 2011).

Exploratory factor analysis EFA: using KMO, Bartlett and variance extracted tests to determine representative scale system.

- Factor loading represents the correlation between the original variable (items) and the factor. To determine the factor loading, according to Hair et al (2009), it is determined by the sample size and the smaller the loading factor, the larger the research sample size. Factor loading > 0.5 is practically meaningful. (quoted from Le Van Huy and Truong Tran Tram Anh, 2012). Achieving convergence value if any variable has a factor loading factor less than 0.5 will be eliminated, the largest factor loading factor in any column belongs to that factor. Achieving discrimination value if the factor loading factor appears 2 factors, the distance between the 2 factors must be greater than 0.3, then take the observed variable with the larger factor loading factor.

- Using the principal axis factoring method with Promax rotation (oblique rotation) (According to Anderson and Gerbing (1988) the method

This will reflect the data structure more accurately than the principal components extraction method with varimax rotation (orthogonal rotation).

- Bartlett test: This test is used to examine whether the correlation matrix is ​​an identity matrix I, which is a matrix whose components (correlation coefficients between variables) are zero and whose diagonal (correlation coefficient with itself) is 1. If the Bartlett test has p < 5%, we reject the hypothesis H 0 , meaning that the variables are related to each other. (Nguyen Dinh Tho, 2011).

- KMO test: According to Norusis (1994), KMO test is an index used to compare the magnitude of the correlation coefficient between two variables Xi and Xj with the magnitude of their partial correlation coefficients. To use EFA, KMO must be greater than 0.50. (Quoted by Nguyen Dinh Tho, 2011).

- Total variance extracted: When evaluating EFA results, we need to consider the total variance extracted. This total shows how many percent of the measured variables are extracted. This total must be 50% or more. (Nguyen Dinh Tho, 2011).

- Eigenvalue criterion: is a commonly used criterion in determining the number of factors in EFA analysis. With this criterion, the number of factors is determined at the factor (used at the factor) with eigenvalue ≥ 1. (Nguyen Dinh Tho, 2011).

Testing methods used in quantitative analysis:

ANOVA analysis of variance: In the scope of the thesis, the author compares the mean values ​​of 3 or more groups. Therefore, the author will use analysis of variance (Analysis of variance - ANOVA). This technique is based on calculating the level of variation within groups and the variation between group means. Based on these two estimates of the level of variation, we can draw conclusions about the level of difference between group means. (Hoang Trong and Chu Nguyen Mong Ngoc, 2008). The author uses one-way ANOVA analysis of variance. Because only 1 factor variable is used to classify observations into different groups. If sig. ≤ 𝛼 , the conclusion is sufficient to reject the hypothesis and accept the alternative hypothesis. If sig. > 𝛼 , there is not enough basis to reject the hypothesis, that is, there is not enough basis to accept the alternative hypothesis. (Le Van Huy and Truong Tran Tram Anh, 2012).

3.3. Collection and processing of research data

Data collection method : conducting direct interview techniques combined with sending questionnaires via email. Sampling was conducted using a non-probability convenience method, that is, selecting accessible subjects who are knowledgeable and involved in implementing internal control on social insurance collection.

This sampling method can select subjects that are accessible to them. This method has the advantage of being easy to access to research subjects and is often used when time and budget are limited. The disadvantage of the method is that it does not generalize to the population.

Sample size depends on many factors such as processing method, regression model. The sample size must be large and meet the sampling criteria using the above-mentioned analysis tools. To use EFA, a large sample size is required. According to Hair et al (2009), in EFA, sample size is often determined based on the minimum size and number of measured variables. The minimum sample size used is 50, preferably 100, and the ratio between observations and measured variables is 5:1, preferably 10:1 or more.

Therefore, the thesis chose to approach the sample size with a minimum number of 100 and the observation ratio on the measured variable is 5:1. However, in order to increase the reliability of the topic, the author conducted a survey on 220 subjects.

Survey subjects : Survey subjects are leaders, social insurance officers and officers working in social insurance collection management in Binh Duong province.

Tools: After collecting, the survey forms will be reviewed to eliminate surveys that do not meet the requirements. Next, the data will be coded, entered into the computer, and cleaned using SPSS 20.0 and Excel software.

3.4. Research model and hypothesis

3.4.1. Building a research model on factors affecting the effectiveness of the internal control system in social insurance collection

From the study of theoretical foundations of INTOSAI 2013, theories of internal inspection and risk control activities in the social insurance industry and previous studies at home and abroad on factors affecting the effectiveness of the internal control system in social insurance collection and experts' opinions, the author proposes a research model including 6 factors affecting the internal control system on social insurance collection as follows: control environment factor, risk assessment factor, control activity factor, information and communication factor, monitoring factor and awareness factor of social insurance participation.

+

+

Control activities

+

Revenue control capabilities

+

+

+

Control environment

Risk Assessment

Information and communication

Monitor

Awareness of participating in social insurance

Source: Author's synthesis

Figure 3.2 . Initial research model on factors affecting the internal control system on social insurance collection.

3.4.2. Building research hypotheses

From the above research model, combined with the discussion results of experts, the author proposes 6 hypotheses that need to be tested, including:

Hypothesis (H1): Control environment has a positive impact on revenue control ability.

Hypothesis 2 (H2): Risk assessment has a positive impact on revenue control ability.

Hypothesis 3 (H3): Control activities have a positive impact on revenue control ability.

Hypothesis 4 (H4): Information and communication have a positive impact on revenue control ability.

Hypothesis 5 (H5): Supervision has a positive impact on revenue control ability.

Hypothesis 6 (H6): Awareness of participating in social insurance has a positive impact on the ability to control revenue.