Money are authority and source of influence symbol, which makes people tendent to material life ( Lemrová et al., 2014 ; Taneja, 2012 ). Person gives different meanings of money, so their money attitudes are different. Attitudes towards money affect people’s life areas, such as saving, spending habits, work performances, political actions ( Roberts & Sepulveda, 1982 ; Phau & Woo, 2008 ). Money attitude includes money cognition, money affection and money action ( Li & Zhang, 2002 ; Zhou & Pham, 2004 ). 1) Money cognition :it refers to an individual’s personal understanding and beliefs about money, including their convictions, judgments, and perspectives on money-related matters. 2) Money affection: it involves emotional responses to money, such as feelings of love, anger, sadness, and happiness towards money-related things. 3) Money action: it refers to the inclination towards certain actions related to money, reflecting the individual’s intentions, mental state, and efforts to satisfy wants or needs. Overall, money attitude is defined as a consistent and enduring behavioral tendency towards money-related matters. Prior Scholars developed money attitude scales such as the Money Attitude Scale (MAS) was created by identifying four key factors: 1) power-prestige, where money is seen as a symbol of success and the ability to influence others; 2) retention-time, which involves money behaviors that involve planning and preparing for the future; 3) distrust, which refers to feeling uneasy about spending money; and 4) anxiety, where money is associated with feelings of insecurity and anxiety about finances ( Yamauchi & Templer, 1982 ). The Money Beliefs and Behavior Scale (MBBS) identified six distinct factors: 1) obsession with money: being fixated on money, 2) power/spending: viewing money as a representation of power, 3) retention: focusing on managing money, 4) security/conservative attitude: using money cautiously for security, 5) inadequacy: worrying about not having enough money, and 6) effort/ability: seeing money as a reflection of personal effort and skill ( Furnham, 1984 ). The Money Ethic Scale (MES) was used to analyze the perceptions of money among a group of full-time workers in the United States. Six key factors were discovered: (1) Good: money is seen as good, valuable, and appealing, 2) Evil: money is considered evil, shameful, and unimportant, 3) achievement: money symbolizes one’s accomplishments in society, 4) respect/esteem: money earns individuals respect from others, 5) freedom-power: money provides personal freedom and influence, and 6) budgeting: individual manage their money through budgeting ( Tang, 1992 ). The power-prestige, distrust and anxiety dimensions of money attitude were found closely to be associated with buying behavior as well as credit card usage moderates these relationships ( Roberts & Jones, 2001 ). It suggested that money attitudes toward power and price sensitivity affect buying behavior ( Khare, 2016 ). Money related anxiety increases compulsive buying. Price sensitivity (distrust) decreases buying ( Eroğlu & Kocatürk, 2020 ).

H₁: Power-prestige of money attitude has a positive impact on purchase intention.

H₂: Retention-time of money attitude has a positive impact on purchase intention.

H₃: Distrust of money attitude has a positive impact on purchase intention.

H₄: Anxiety of money attitude has a positive impact on purchase intention.

Consumer conformity is formed from the relationship between social interactions and consumer purchasing decisions for products and services. Conformity can be described as the act of adjusting one’s thoughts and choices to match those of the majority within a group ( Allen, 1965 ). Consumer conformity classifies people into two groups, namely the conformity group and the non-conformity group. The conformity group is further classified into two groups, namely normative conformity and informational conformity. Lascu and Zinkhan (1999) theorized the concept and built a conformity model that provided a framework explaining how an individual conforms to group norms to meet the group’s expectations in consumer behavior. Conformity is the result of two primary factors: normative influence, where an individual aligns their behavior with that of a group in order to gain acceptance and approval, and informational influence, which describes how individuals make decisions based on observing others when they lack sufficient information for independent judgment ( Deutsch & Gerard, 1955 ). Baron, Branscombe & Byrne (2008) , explain that conformity is also a type of social influence in which individuals change attitudes or behaviors to comply with existing social norms. Sarwono & Meinarno (2013) , conformity occurs when a number of people in a group say or do something, causing a tendency for members to say and do the same thing. The model of conforming behavior includes four characteristics (individual characteristics, group characteristics, brand characteristics, task/situation characteristics), two types of norms (extreme, moderate), and two major categories (informational influence, normative influence) ( Lascu & Zinkhan, 1999 ). Conformity has two aspects: informational influence and normative influence. First, informational influence means that individuals tend to align themselves with a certain group because of the influence of information conveyed by the group and the individual beliefs in that group based on their assessment. It can be stated that the views and information from trusted groups help individuals adjust to those groups, leading to the development of conforming behavior. The second component of conformity is normative influence, which refers to an individual’s attempt to alter their perceptions, beliefs, attitudes, and behaviors in order to gain acceptance from specific groups or, in other words, to avoid being excluded. This also causes individuals to take actions that are by the group’s expectations of them. Normative conformity involves individuals adapting their behavior to align with the choices of others in order to gain acceptance or approval from a specific individual or group. In contrast, informational conformity refers to the adjustment of one’s behavior based on relevant information received about the choices made by others. Consumer conformity arises from the connection between social interactions and the choices consumers make when buying products and services ( Princes et al., 2020 ). Park & Yang (2010) found that celebrity conformity has a positive effect on purchase intention. Both normative and informational conformity have been demonstrated to positively influence consumer attitudes, thereby also positively impacting purchase intentions ( Khandelwal et al., 2018 ).

H₅: Normative influence of conformity has a positive impact on purchase intention.

H₆: Informational influence of conformity has a positive impact on purchase intention.

Consumers’ price sensitivity refers to the extent to which consumers respond to variations in the price of a product or service. It assesses the extent to which fluctuations in price affect consumer purchasing behavior ( Dominique-Ferreira et al., 2016 ). Uslu & Huseynli (2018) define price sensitivity in another perspective, which price sensitivity is one of the personalities that would influences individual’s purchasing behavior. The researchers found that personality traits significantly influence price sensitivity, suggesting that individuals’ inherent tendencies towards cost-consciousness directly impact purchasing decisions when faced with price fluctuations. Price is considered as a prime factor in consumer’s evaluation of purchase decisions and it is termed as sum total of money the firm charges its customers with the addition of profitability margin in producing specific goods or services ( Wang et al., 2020 ). Consumers who exhibit price sensitivity tend to prioritize lower prices and are less inclined to make purchases when prices increase. In contrast, consumers who are less sensitive to price fluctuations demonstrate a greater willingness to pay elevated prices for the same products and are more likely to continue purchasing even in the face of price increases ( Foxall & James, 2003 ; Shimp, Dunn, & Klein, 2004 ). Previous research examined price sensitivity as moderating effect on purchase intention. Honkanen et al. (2012) support that price sensitivity influences significantly their positive attitudes and intentions to purchase products. Specifically, when price sensitivity is diminished, the correlation between positive attitudes and purchase intentions becomes more pronounced. In essence, consumers who exhibit lower price sensitivity are more readily swayed by favorable attitudes, thereby increasing the likelihood of their purchasing decisions. Congruence between brand characteristics and consumer’s personal traits plays a crucial role in consumer purchase intentions ( Akbar et al., 2023 ).

H₇: Price sensitivity has a positive impact on purchase intention.

To gain insight into the usage intentions of the social media platform to purchase, this study conducted a qualitative interview method and combined it with relevant literature and expert opinion to develop the questionnaire. After accessing the pre-test, collecting samples and revising the questionnaire, this study used confirmatory factor analysis (CFA) to examine the measurements’ reliability and validity. Regression analysis approach and the SPSS 22.0 software were employed to verify the coefficients and hypotheses to meet the expectations ( Figure 1 ).

This study has targeted the usage intentions of the social media platform to purchase due to the study objective. The participation of the respondents was voluntary and they were treated as politely to fill the questionnaire. All the participants ever bought things on the social media platform. The questionnaire items were identified and designed based on literature reviews, and then modified to make them appropriate for assessing Cambodia consumers’ money attitude, conformity and Price sensitive for using social media platform (Items were measured using a 5-point Likert-type scale (1 = strongly disagree, 5 = strongly agree). The survey was held in Phnom Penh Cambodia. The data were collected over a period of one month. In Cambodia, there are around 18,000 thousand people in 2024. Under the 90% of CI and 3% of statistical error, the sample sizes are at least over 271. In total, 300 questionnaires were administered in the Phnom Penh province of Cambodia (250 questionnaires issued at facing to face, 50 questionnaires in internet) with 283 valid questionnaires returned, indicating 94% return rate. Results of the study are presented on the basis of a frequency distribution, factor and reliability analysis, ANOVA and regression analysis using SPSS ( Norusis, 1999 ).

Of the 283 valid respondents who returned the questionnaires, 37.1% were male and 62.9 percent female. About 56.2% of the respondents were between 21 and 30 years old. According to the results, 70.7% of the respondents have salaries under USD$ 800 dollars per month. Moreover, an analysis of the questionnaire revealed that most participants use social media platform to purchase less than 3 years and less than 15% of their salary to spend in the social media platform each month. Cloth, accessories (glasses, watch, earrings, wristband) and cellphone items are the most to purchase on it. The most social media platform usage are Taobo, Facebook, Instagram, and e-gets.

Principal-component factor analysis was conducted to identify underlying constructs from sets of interrelated items. Factors were extracted and interpreted using varimax rotation method. Table 1 shows the factor analysis and reliability analysis of MAS items ( Yamauchi & Templer, 1982 ), including Power-Prestige (PP), Retention-Time (RT), Distrust (DIS), and Anxiety (ANX). In Table 1 , all factor loadings of variables exceeded 0.5, while Cronbach’s α values exceeded 0.7. Table 2 shows the factor analysis and reliability analysis of conformity ( Bearden et al., 1989 ). Factors such as Information influence (II) and Normative influence (NI) were produced using factor analysis. All factor loadings of variables exceeded 0.5, while the Cronbach’s α values exceeded 0.6. Table 3 shows the factor analysis and reliability analysis of price sensitivity (PS) ( Afendi, 2023 ) and purchase intention (PI) ( Bian et al., 2012 ). All factor loadings of variables exceeded 0.5, while the Cronbach’s α values exceeded 0.8. The four factors of MAS, two factors of conformity and price sensitivity appeared to be appropriate for the next statistical analysis.

To verify the research framework and hypotheses, this study used SPSS statistic software to test multiple regressions. Multiple regressions was conducted to determine the extent to which the independent variables affect the dependent variable is a free .Variable PP, RE, DIS, ANX, II, NI and PS. Thus, multiple linear regression analysis showed the PP, RE, DIS, ANX, II, NI and PS. to the PI. Multiple linear regression equation is as follows:

$Y=\alpha +{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_3{X}_3+{\beta }_4{X}_4+{\beta }_5{X}_5+{\beta }_6{X}_6+{\beta }_7{X}_7+\epsilon$

Information:

Y = User intention (PI);

X₁ = PP, ${\beta }_1$= regression coefficient of PP;

X₂ = RE, ${\beta }_2$ = regression coefficient of RE;

X₃ = DIS, ${\beta }_3$= regression coefficient of DIS;

X₄ =ANX, ${\beta }_4$= regression coefficient of ANX;

X₅ = II, ${\beta }_5$= regression coefficient of II;

X₆ = NI, ${\beta }_6$= regression coefficient of NI;

X₇ = PS, ${\beta }_7$= regression coefficient of PS;

$\epsilon$= Residual.

Table 4 shows the value of tolerance and variation inflation factor (VIF) value for intention to use banking application towards PP, RE, DIS, ANX, II, NI and PS. The result shows that there is no multicollinearity exists in the model since all the tolerance value is greater than 0.1 and VIF value is less than 10 respectively.

After all the assumptions have been fulfilled, the multiple linear regression analysis was performed. The regression analysis yielded a multiple correlation coefficient (R) of 0.857 which means that there was strong relationship between the mean score of PI and the seven of predictor variables (PP, RE, DIS, ANX, II, NI and PS). Based on Table 5 , the value of R² = 0.734 indicates that 73.4% of the variation in PI is explained by PP, RE, DIS, ANX, II, NI and PS. The balance 26.6% is explained by other factors. The overall regression model was significant since the significance value (0.000) and less than 0.05 with F-Ratio of 39.8. The standard error of the estimate is 0.48097 which is low indicates that the error is quite small.

significance level; *: ρ < 0.1,**: ρ < 0.05, ***: ρ < 0.01.