THE IMPACT OF FINANCIAL LITERACY ON THE DECISION TO USE A MOBILE PAYMENT APPLICATION: A CASE STUDY OF HOUSEHOLDS THAT HAVE JUST ESCAPED POVERTY IN HANOI

Purpose: To study and quantify the impact of financial literacy on the decision to use a mobile payment application. Case study of households that have just escaped poverty in Hanoi. Theoretical framework: The foundation for building a theoretical framework is the content of personal finance knowledge in this study, including (i) financial knowledge, (ii) financial attitudes, (iii) financial skills and (iv) financial behavior. Financial literacy is assessed using a combination of objective multiple-choice questions and self-assessment. Methods: The study is based on survey data collected through a structured online questionnaire for newly lifted household heads, as well as secondary data collected from respondents' profiles. Research data were collected from 186 responses using convenient non-probability sampling technique and analyzed by descriptive statistics, ANOVA analysis and Binary Logistic regression analysis. ANOVA analysis showed a statistically significant relationship between the groups of subjects in terms of personal financial knowledge (in both self-assessment method and objective test method). Results and conclusions: Binary Logistic regression results show that the main variables that positively affect the probability of a new household head out of poverty include Personal Financial Knowledge, Personal Financial Behavior, Gender, Status Occupation and Education Level. Financial attitudes and Religion have a negative impact on the ability of household heads to decide to use mobile payment applications.


INTRODUCTION
In recent years, there has been an increasing interest in personal financial literacy. It is easier for someone with a good financial understanding to explain economic behavior in all areas of economics and finance. Financial literacy is increasingly influencing an individual's decision-making. Especially when people are more proactive and have a greater sense of responsibility regarding to personal financial planning. That tendency to increase responsibility may stem from each person's need to protect themselves and their families in the face of crises, in which the vulnerable frequently suffer primarily from poetic and inexperienced financial ignorance. Many researchers believe that the global crisis known as the "subprime mortgage crisis" of 2008 brought the importance of financial literacy to the forefront in the past (Mandell  Robb & Woodyard, 2011). However, acquiring modern financial knowledge, skills, attitudes, and behaviors and thus making effective financial decisions has never been easy, and it is especially difficult for the Newly-Escaped-Poverty-List householders (NEPLHs). They are also potentially vulnerable, as they are frequently unable to understand complex financial concepts, preventing them from accessing products that require knowledge and understanding of critical financial concepts and issues. Many people worldwide have seen their earnings and living costs rise due to the COVID-19 epidemic, particularly the impoverished in developing countries (Fuzikawa, E. K., dos Reis, S. A., & Sandi, D. ,2023). The COVID-19 epidemic has had two main economic consequences in Vietnam. On the one hand, the impact of COVID-19 caused significant damage to NEPLHs due to a decrease in income and increase in medical expenses. Due to the tight application of anti-epidemic remedies, supply chain interruptions, and market connectivity disruptions, NEPLHs face even more significant problems in the process of family economic recovery. Mobile payment services, on the other hand, have flourished, allowing all clients to have faster access to personal financial services even during the period of social distancing caused by COVID-19. Financial intermediaries in Vietnam have improved their provision of steady quality financial services while embracing technology to cut costs during the post-COVID-19 economic recovery period. As a result, individual clients will have easier access to mobile payment services. Installing and using mobile payment applications allows NEPLHs in Vietnam to access personal financial services, giving them additional chances to boost their income and reduce the cost of living for themselves and their families. In this context, investigating the impact of personal financial literacy on NEPLHs' decision to use mobile payment services will provide more empirical evidence for governance implications and policy recommendations to improve financial literacy for the NEPLHs, thereby contributing to promoting financial inclusion in Vietnam, particularly for the NEPLHs, and assisting them in maintaining their sustainable economic development.

LITERATURE REVIEW AND THEORETICAL FRAMEWORK
Numerous studies have attempted to explain the definition of financial literacy and personal finance in many countries worldwide, including Vietnam. Personal financial literacy is defined by Servon and Kaestner (2008) as a person's ability to understand and apply financial concepts. According to Huston (2010), Personal finance frequently manifests with two main aspects: Personal financial knowledge and Information literacy skills relating to personal financial knowledge. Personal financial literacy relates to measuring information literacy skills relating to personal financial knowledge in decision-making. Furthermore, over time, aspects of personal financial literacy have expanded to include personal finance attitudes and personal finance behavior, as well as financial knowledge and skills in individual financial decisionmaking.
Although there has been controversy on the definition of personal financial literacy, moststudies agree on the importance of measuring personal financial literacy. According to Huston (2010), many approaches to measure financial literacy. The two most popular approaches are the objective method (such as multiple-choice and quick tests) and the subjective method (such as the self-assessment method). Many studies have concluded that financial literacy can be measured objectively using multiple-choice testing and subjectively using the self-assessment method (Xiao et al., 2014). While objective measures can help check financial literacy performance, subjective measures are essential in explaining a person's confidence or lack of confidence in their financial literacy. The empirical evidence also suggests significant differences between objective and subjective measures of financial literacy (Xiao et  The findings of research on the impact of financial literacy on personal financial behavior and decisions are also varied. According to Peng et al. (2017), the ability to engage in recommended financial activities is dependent on personal financial literacy, and individuals' propensity to invest and save is thought to be influenced by their perceived control over outcomes and financial resources. Simultaneously, financial literacy has been shown to influence personal decisions and intentions regarding asset accumulation, borrowing, and investing.
To sum up,the foundation for developing the personal financial literacy content framework in this study includes (i) financial knowledge, (ii) financial attitude, (iii) financial skills, and (iv) financial behavior. Financial literacy is assessed using a combination of objective multiple-choice questions and self-assessment. Data will be gathered by surveying participants with standardized questionnaires, as has been done in previous studies on similar research topics (Xiao et al., 2014;Peng et al.,2017;Contrigiani, A. C., Marjotta-Maistro, M. C., Montebello, A. E. S., & Sautier, D. ,2023). The findings should contribute to differences in financial literacy and financial behaviour. Financial behaviour in this study is considered the decision of Hanoi householders who have recently escaped poverty to use mobile payment services. We also consider the influence of gender, age, occupation, religion, marital status, family role, and the number of years out of poverty.

RESEARCH DESIGN
To examine the influence of financial literacy on the decision to install and use the mobile payment application, this paper uses the following model: where: is the probability that the NEPLHs installs and uses a mobile payment application P (DMPA = 0) = 1 -P0 is the probability that the NEPLHs does not install and use a mobile payment application Ln The natural logarithm βi Regression coefficient of FLi βj Regression coefficient of CVj

Research Hypothesis
H1: Financial knowledge has a statistically significant positive effect on the probability that the NEPLHs installs and uses a mobile payment application (Po) H2: Financial attitude has a statistically significant positive effect on the probability that the NEPLHs installs and uses a mobile payment application (Po) H3: Financial skills has a statistically significant positive effect on the probability that the NEPLHs installs and uses a mobile payment application (Po) H4: Financial behavior has a statistically significant positive effect on the probability that the NEPLHs installs and uses a mobile payment application (Po) The study uses convenient non-probability sampling technique based on the list of email addresses and phone numbers of the NEPLHs provided by the credit specialists of Vietnam Bank for Social Policies -VBSP. There were 250 survey questionnaires sent, resulting in a sample size of 186 respondents. The results of the descriptive statistical analysis of FLi variables show that most aspects of personal financial literacy have a good average score. The mean score of the Personal Financial Knowledge criterion is the highest (above 3.99) with a standard deviation of 0.88. The criteria for Personal Financial Attitude, Personal Financial Skills, and Personal Financial Behavior received lower mean values, but higher standard deviations, reflecting heterogeneity among respondents. That difference will be further explored and commented on in the analysis of the difference in mean FLi between groups by demographic and sociological criteria (CVj). 6 Table 4 shows the difference in financial literacy between the objective multiple-choice assessment approach and the self-assessment approach based on survey responses. The DFL value is calculated as the difference between the FL and SFL values, where a negative rating (<0) indicates overconfidence and a positive rating (>0) indicates a lack of confidence in their self-assessment of their personal financial literacy. In some observations, the lowest DFL value of -2.48 indicates overconfidence, while the highest DFL value indicates a lack of confidence that is slightly lower than the highest overconfidence. When comparing self-assessment results with objective measures of personal financial literacy, the mean value of DFL at 0.5681 is low. However, the deviation of 0.99844 shows inconsistency among respondents. In fact, given that the survey respondents are all household heads who have just escaped poverty, the disparity in the gap between FL and SFL is not surprising. According to the findings of the analysis, there is a need for more regular training and exchange activities to assist the NEPLHs in more accurately assessing their own financial knowledge, so that they do not lack confidence and do not over-confident in their financial knowledge.

ANOVA Analysis Results
• Differences between groups by sex: Analyzing the difference in FL and SFL between male and female respondents with the One-Way ANOVA technique did not find a statistically significant difference in SFL. However, the results of the analysis show that there is a statistically significant difference in FL. On the mean, female NEPLHs have higher financial literacy (range 3.41 to 4.20) than male NEPLHs (range 2.61 to 3.40). This gap can be explained by the fact that women play a key role in rural Vietnamese families. Female NEPLHs are more responsible than male NEPLHs in enhancing financial literacy.
• Differences between age groups: The results of analyzing the difference in FL and SFL among age groups using the One-Way ANOVA technique did not find a statistically significant difference in self-assessment of financial literacy. The results of the analysis, however, show that there is a statistically significant difference in the results of the overall personal financial literacy measurement (determined by objective multiple-choice measurement), with overall financial literacy being significantly different between the group of NEPLHs under the age of 20 and those between the ages of 20 and 59. The financial literacy of over-60-year-old NEPLHs has fallen slightly below the 2.60 thresholds, owing in part to the effects of old age. This result can be explained by the fact that, as they age, the NEPLHs accumulate better in personal financial knowledge andin attitudes, skills, and personal financial behavior.
• Differences between groups by marital status Unlike the analysis by gender or age group, analyzing the difference in FL and SFL between single and currently married respondents using the One-Way ANOVA technique revealed no statistically significant difference in overall personal financial literacy. The analysis results, however, show a statistically significant difference in the self-assessment of personal financial literacy, with married respondents having higher self-assessment financial literacy than single heads of households who have just escaped poverty. This fact reflects the higher self-confidence of married household heads versus single household heads. This finding has many implications for future research on variables related to household heads' marital and family status who have recently escaped poverty by accessing mobile payment services.
• Differences between groups by role in the Family The study discovered a statistically significant difference in overall personal financial literacy by analyzing the difference in mean values of FL between respondents with different roles in the family using the One-Way ANOVA technique. However, the analysis results show no statistically significant difference in personal financial literacy self-assessment results. Specifically, the overall level of personal financial knowledge varies according to the respondent's important role in terms of income source and contribution to the family. Heads of households who have recently escaped poverty, who are the primary occupation and source of income, have the highest level of personal financial literacy among the three groups. In contrast, financially dependent people have the lowest.
• Differences between groups by career status According to the results of the analysis of the difference in mean values of FL and SFL among respondents in groups by current occupational status, the study discovered a statistically significant difference in overall personal financial literacy as well as individual financial literacy self-assessment results using the One-Way ANOVA technique. When analyzing FL and SFL, the difference in status between the groups was inverse. The mean SFL of the head of household group who has recently escaped poverty without a stable job is lower than that of the other groups, while respondents working in non-agricultural occupations have the highest SFL. The FL of a household head who has recently escaped poverty without a stable job, on the other hand, has a rather high mean value, only slightly lower than that of respondents who are currently employed in non-agricultural occupations. The group of respondents who practiced pure farming had the lowest mean FL and SFL values of the groups. This fact demonstrates that heads of households who practice a combination or non-agricultural practice have more advantages than heads of households who have just escaped pure agricultural poverty; heads of households without stable jobs are motivated to learn and update financial knowledge, skills, attitudes, and behaviors, and have a good overall financial literacy.
• Differences between groups by education level The results of analyzing the difference in mean FL and SFL values among respondents in groups based on educational level revealed that the study found a statistically significant difference in mean SFL but no difference in overall personal financial literacy using the One-Way ANOVA technique. The group of newly out of poverty heads of households has a selfassessment of the mean SFL that rises with educational attainment. In the context of rural research, respondents with a Postgraduate degree were included in the survey sample, but they did not have a higher level of SFL than respondents with a College -University degree. The group with the lowest mean SFL is those who have just finished elementary school, though their overall personal financial literacy is not too low (between 2.61 and 3.40).
• Differences between groups by religion The results of analyzing the difference between the heads of households who have just escaped poverty according to Buddhism are similar to the results of analyzing the difference in mean values of FL and SFL between the respondents belonging to the groups by education level. According to the One-Way ANOVA technique, the study found a statistically significant difference in the mean value of SFL but not in FL among Buddhism and the rest of the survey sample. Given the nature of Buddhists, prudence and modesty may explain why the Buddhism households heads have a lower SFL than the rest of the respondents. These results imply that there is a need for solutions to increase the confidence of householders who have just escaped poverty and are Buddhist disciples.
• Differences between groups by number of years out of poverty In contrast to the analysis of the difference in mean values of FL and SFL among respondents belonging to religious groups, the results of the analysis of the difference between the heads of households who have just escaped poverty by the number of years out of poverty did not find a statistically significant difference in the SFL. However, theydid find a difference in the FL. Heads of households who were removed from the list within the last year had a higher mean SFL value than those who were removed within the last two years. This demonstrates, in part, that not every success in escaping poverty implies that FL has improved in mean value. Over a three-year period, the mean FL of the group of heads of households who had just escaped poverty from the list was the highest, indicating a significant improvement in personal financial literacy. Binary Logistic regression analysis results of the research model with the dependent variable being a binary variable. The analysis process is continued with the elimination of research variables that are not statistically significant to obtain the optimal research model, which can be used in making findings based on the results of the analysis. After removing variables with no statistical significance, the results of the Binary Logistic regression analysis revealed that the independent variables explained 63.6% of the variation in probability in the household head who has just escaped poverty installing and using the mobile payment application. Subsequent analysis shows that the adjusted research model correctly predicts the probability of making a decision to install and use a mobile payment application of the household head who has just escaped poverty with an overall forecast level of about 90.3%. Specifically, the probability of choosing to install and use a mobile payment application of the household head who has just escaped poverty is positively influenced by the factors arranged in order of importance, respectively, Personal financial knowledge, Gender, Personal financial behavior, Career status; Education level; while Personal financial attitudes and Religion have a negative effect on the probability of installing and using mobile payment applications of householders who have just escaped poverty. According to the research findings, if the initial probability of installing and using a mobile payment application by the head of a household who has just escaped poverty is 10%, and personal financial knowledge improves by one unit, the probability of installing and using a mobile payment application by the head of a household who has just escaped poverty will reach 53.18%, an increase of 43.18% over the original probability, while other factors remain constant.

Binary Logistic Regression
The calculation is repeated for the other variables to determine the improvement in the probability of installing and using a mobile payment application by the household who has just escaped poverty when each significant independent variable is improved. Table 18 summarizes the statistical meanings from the adjusted research model.

• Effect of FL2 on P0
According to the research findings, if the Personal Financial Attitude increases by one unit, the probability of the household head who has just escaped poverty installing and using a mobile payment application will reach 2.16%, which is 8.84% lower than the original probability (10%) if all other variables remain constant.
• Effect of FL4 on P0 According to the research findings, if the Personal Finance Behavior increases by one unit, the probability of installing and using a mobile payment application by the household head who has recently escaped poverty will reach 21.29%, an increase of 11.29% over the original probability (10%) if all other variables remain constant.
• Effect of Gen on P0 According to the research findings, a female head of household who has recently escaped poverty has a 31.59% chance of installing and using a mobile payment app, which is 21.59% higher than the 10% chance of a male head of household installing and using a mobile payment app if all other variables remain constant.
• Effect of Car on P0 The study's findings imply that if the household head's employment status is raised by one step, the probability of installing and using a mobile payment application will rise to 18.12%, an increase of 8.12% if all other variables remain constant.
• Effect of Edu on P0 According to the research findings, if the education level of the household head who has just escaped poverty increases by one level, the probability of installing and using their mobile payment application will reach 16.92%, a 6.92% increase over the original probability assuming all other variables remain constant.
• Effect of Rel on P0 According to the findings, a Buddhist householder has a 1.71% chance of installing and using a mobile payment application, which is 8.29% lower than the initial probability for a non-Buddhist householder. religion, as long as all other variables remain constant.
In this study, householders who have recently emerged from poverty in Hanoi, Vietnam, are examined to see how characteristics of personal financial literacy, such as knowledge, attitudes, skills, and behaviors, affect their likelihood of using mobile payment services. The results shown in Table 19. FL2 has a statistically significant positive effect on P0 Approved H3 FL3 has a statistically significant positive effect on P0 N/A H4 FL4 has a statistically significant positive effect on P0 Approved Source: Extracting from survey data processing results

CONCLUSION
Personal Financial Knowledge, Personal Financial Behavior, Gender, Career, and Education level are the key variables that positively affect the probability of choosing to install and use mobile payment applications (P0) of newly escaped poverty householder, according to research findings. Meanwhile, P0 is negatively influenced by Personal Financial Attitude and Religion.
The findings of SFL and FL differences between groups of respondents could be useful to financial intermediary administrators as well as policymakers in increasing knowledge and improving individual financial behavior to encourage newly escaped poverty householders to use mobile payment services, thereby promoting financial inclusion, sustainable growth associated with social security.
The use of an online survey is the most significant limitation of this study. Firstly, the survey questionnaire is primarily limited to fairly simple information, with a structured questionnaire. Secondly, the lack of open-ended questions and opportunities to directly exchange and interview heads of households who have recently escaped poverty can lead to the omission of interesting information, findings related to new factors that have not been mentioned in previous studies. These limitations suggest future research directions by combining multiple data sources and orienting to exploit individual aspects related to financial inclusion of newly escaped poverty household heads in order to have more discoveries and useful implications in the future.