Predictors of solar water heater purchasing intention: an empirical analysis of residential behaviors in Hanoi city, Vietnam

Solar energy is a field that has grown rapidly in the world in recent years to solve the energy crisis and respond to climate change. Vietnam is a country with renewables energy potential including solar power and is pursuing net zero emissions target by 2050. This study explores factors affecting the residents’ intention to buy solar water heater in Hanoi, Vietnam. We used theory of planned behaviors and technology acceptance model to build research model and hypotheses. A community survey was conducted with 450 households in 3 districts using cluster sampling method to collect information. We conducted descriptive statistical analysis, Cronbach’s Alpha test, exploring factor analysis, conformity factor analysis and regression to identify and evaluate the impact of factors on solar water heater consumption intention. The results showed that there were significant factors affecting people’s intention to buy solar water heater including attitude, perceived usefulness, price perception, social influence and financial support in which attitude is the most influential factor. Regarding social characteristic of respondents, we also found positively significant relationship between income, education level and intention to use solar heater. Recommendations for government and business included increasing awareness of the benefits and economic feasibility of renewable energy devices to consumers through social media, implementing financial incentives to support clean energy consumption behavior, and strengtheing the emergence of local solar evangelist groups in the communities. Future research can expand to other cities and explore more indirect factors affecting the green energy consumption behavior of Vietnamese residents.


Introduction
Fossil energy has been playing an important role in the economic and social development of humanity, and still dominates the energy structure of almost every country in the world (Azeez and Atikol 2019, Alipour et al 2020, Dat and Truong 2020).However, this traditional resource is being depleted at a rapid rate due to the exponential increase in consumption demand in recent decades, and is a key factor causing global warming and climate change (Urpelainen and Yoon 2017, Liobikienė et al 2021, Dinh 2022),.The negative impacts of climate change are becoming increasingly serious and affecting the sustainable development of every country and region in the world (Liu et al 2013, Ali et al 2019, Huang 2019, and Duc et al 2022).
Accelerating the development of renewable energy has become a common trend in many countries (Claudy et al 2011, Nyambuu and Semmler 2014, Kumar et al 2022).This step not only helps countries solve the energy supply problem, but also makes an important contribution to efforts to realize climate goals (Pillai and Banerjee 2007, Palm 2018, Opeyemi 2021).In the world, the EU is one of the leading regions in promoting restructuring of the energy industry towards developing clean energy sources.With strong determination to change the direction of the energy industry, the EU aims to increase the proportion of renewable energy sources and bioenergy to 60% by 2030, and increase offshore wind power capacity to 25 times by 2050, to achieve the goal of neutralizing carbon emissions by 2050 (Opeyemi 2021).Besides the EU, the US is one of the world's largest renewable energy producers.According to the US government's direction, most coal-fired power plants 2. Theoretical models and methodology 2.1.Theoretical framework According to Elmustapha et al (2018), if one could identify variables that impact the adoption of a renewable technology such as SWH in a target market and if one could predict adoption with a reasonable certainty, it could dramatically increase the technology adoption by decisive interventions.So far, researchers have used a number of adoption models to explain intentions actual behaviors, including Theory of Reasoned Action (TRA) (Fishbein and Ajzen 1980), Theory of Planned Behavior (TPB) (Ajzen 1991) and Technology Acceptance Model (TAM) (Davis 1989).Essentially, the models attempt to treat the dependent variables of buyers' purchase intent as a function of contextual constructs and develop a predictable understanding of the same for impactful marketing and policy decisions (Gadenne et al 2011, Dima 2013, and Kumar et al 2022).
TPB developed from the theory of reasoned action (TRA), assumes that a behavior can be predicted or explained by behavioral tendencies to perform the behavior (Ajzen 1991).Behavioral tendencies are assumed to include motivational factors that influence behavior, and are defined as the amount of effort people exert to perform that behavior.According to TPB, behavioral tendencies are a function of three factors.First, attitudes are conceptualized as subjectively positive or negative evaluations of the behavior performed.The second factor is social influence which refers to perceived social pressure to perform or not to perform the behavior.Finally, the TPB is built by adding the element of perceived behavioral control to the TRA model (Ajzen 1991, Dima 2013, Bhutto et al 2020).The perceived behavioral control component reflects the ease or difficulty of performing the behavior; this depends on the availability of resources and opportunities to perform the behavior.Ajzen (1991) suggests that behavioral control factors directly affect the tendency to perform a behavior, and if the person is accurate in his or her perception of the level of control, then behavioral control also predicts the behavior (figure 1).
TAM is a theoretical framework gauging technology acceptance and utilization (Davis 1989).It focuses on understanding why users accept or reject specific technology.Within TAM, two main factors determine technology acceptance: (i) perceived usefulness, evaluating users' perceptions of a technology's ability to enhance work efficiency, improve performance, or provide benefits in work or personal life and (ii) perceived ease of use, measuring the user's perception of the ease of using the technology, encompassing their perceived ease in learning and using it.TAM explicates that technology acceptance relies on users' perceptions of these two factors.If users perceive the technology as applicable and easy to use, they are likely accept and use it (Davis 1989, Davis 1993).This model has been widely applied in research and practical contexts to comprehend user behavior towards technology (Venkatesh et al 2012, Li et al 2013, Hsu 2018, Huang 2019) (figure 2).

Hypothesis development
In constructing the research model for this study, the author has summarized research on factors affecting customer intentions in consuming green products in general and renewable energy technologies in particular.The overview also includes identifying theories, empirical models of behavioral intentions, scales of influencing factor variables, and data processing models.The following are some key summaries of the factors identified in the literature.

Intention to buy
A suggested that intention to buy denotes a customer's likelihood to purchase goods or services and is a tool to predict future actual purchase decisions.In turn, customer purchase intention is an individual's willingness to purchase a particular product that is preferred according to some evaluation based on personal experience (Yuan et al 2011, Urpelainen andYoon 2017).It is a self-conscious plan and promise to oneself to buy or not to buy a particular product after considering personal or internal aspects (e.g.personal knowledge, emotions, etc) and external aspects (e.g.related costs and benefits) (Ajzen 1991).Previous studies found that customer behavior is influenced by inventory, brand image, risk, word of mouth, subjective norms, and households' cost-saving experiences Zhang et al 2012, Saleh et al 2014, Sadiq 2018, Elmustapha et al 2018).This is consistent with the concept of utility optimalization, which states that utility plays an important role in determining customer attitudes and behavior (Dima 2013).Customer trends are also important to increase intention to buy.With improved environmental awareness, some people have a positive, careful attitude towards the environment and feel responsible for preserving nature (Chen et al 2016, Kumar et al 2022).And hence, they tend to buy products that are environmentally friendly, do not harm nature, and do not pollute the environment.

Attitude
According to TPB, attitude is considered one of the determinants of behavioral intention (Ajzen 1991).This implies that attitude toward behavior is understood as an individual's favorable or unfavorable evaluative assessment regarding a specific behavior (Ajzen 1991, Kumar et al 2022).In addition, attitude toward using a new system is also defined as an individual's overall emotional reaction toward utilizing that system (Ali et al 2019, Abdullah et al 2021).General attitudes encompass individuals' beliefs that translate into actions related to their concerns.Specifically, when it comes to environmental attitudes, these beliefs often manifest as a willingness to pay more for environmentally friendly products (Chen et al 2016).Liobikienė et al (2021) defined environmental attitude as how people perceive their role in environmental protection, reflecting their awareness of environmental concerns.Wang et al (2019) offered a slightly different perspective, viewing environmental attitudes as individuals' conjectures about what has positive or negative impacts on the environment.Gadenne et al (2011) suggested that individuals with a positive attitude believe in the necessity to protect the environment and preserve nature.In doing so, these customers tend to purchase products that are environmentally friendly and pose no harm to nature in any way (Parkins 2018).Consequently, when they hold a positive attitude towards SWH, it indicates their intention to buy them.Palm (2018) also provided evidence that attitude towards the actual intention of an individual to use a new system or technology must be clearly defined, with a direct and positive impact.
Hypothesis 1 (H1): Environmental attitude has a positive impact on SWH purchase intention.

Product knowledge
Product knowledge is a process of encompassing awareness, collection, management, and analysis of information regarding any product (Huang 2019).According to Hsu (2018), prominent reasons for the low adoption of solar energy products include customers' limited awareness and familiarity with them.Enhancing an individual's product knowledge about the benefits of energy technology facilitates a higher propensity for intention to buy (Parkins 2018).When products require substantial capital for purchase, favorable conditions should be in place to actually lead to the buying behavior.Customers with product knowledge about available support tools find it easier to proceed with subsequent actions (Ali et

Perceived usefulness
Perceived usefulness measures the willingness to adopt something new compared to customers' traditional values (Elmustapha et al 2018).User behavior is shaped by the perception of higher benefits attained through the use of services (Hsu andLin 2008, Claudy et al 2011).Additionally, Caperello and Kurani (2011) argued in their theoretical framework that perceived usefulness can be defined as the extent to which users believe that using products/services will yield significant effectiveness for them.Hhypothesis 5 (H8): Financial support positively influences the households' Intention to buy SWH.

Social influence
Hsu and Lin (2008) defined social influence as the opinions and perspectives of others regarding the approval and usage of new technology.It also asserted that social influence involves mimicking behaviors based on the opinions of individuals' closest and most intimate connections.When consumers observe family members, friends, and colleagues using clean energy systems, they are encouraged and recommended to adopt these systems themselves (Liobikienė et al 2021, Sanguinetti et al 2021).Azeez and Atikol (2019) contend that social influence positively impacts the intention to use products or services.Studies in various countries have affirmed that social influence affects customers' intentions to use services positively (Kumar et al 2022).A positive social influence perception reduces risk aversion, attracts new customers, and fosters revenue, profit growth, and market share for businesses, according to Michelini (2012).

Price perception
Every product is priced at a value that encompasses all costs any individual might incur to purchase it (Claudy et al 2011).The primary constraint for anyone buying SWH is their price (Carperello et al 2011, Danielis et al 2020).Michelini suggested that price is an indispensable part of purchase decisions, especially for items like renewable energy technology, which are infrequently purchased and unlike typical conventional goods.Renewable energy technology requires time for quality assessment, cost evaluation, and comparison between the purchase price and its utility.The higher the cost of a device, the lower the consumer interest, subsequently reducing its adoption rate in the ds market (Batidzirai et al 2009, Nyambuu and Semmler 2014, Opeyemi 2021).
Price perception encapsulates the overall customer evaluation of a product's cost.Price is tied to the trade-off between the price paid and the anticipated benefit.If the price exceeds the customer's acceptable range or implies poor product quality, overall customer perception diminishes.Thus, customers' perception of price can directly influence their intention to buy.Hypothesis 7 (H7): Price perception positively influences the households' intention to buy SWH.
From the above hypotheses, the study model was proposed as follow (figure 3):

Data collection and analysis
Research data collection was divided into two phases.Phase 1, we conducted in-depth interviews with household and focus group discussions (FGD) with market, energy, financial management agencies and businesses selling SWH.The goal of FGD is to identify potential influencing factors and collect information to build constructs for each factor.There were 12 households, 3 SWH suppliers and representatives of 2 management agencies, Hanoi Department of Industry and Trade and Hanoi Department of Finance, participating in FGD.After that, the draft questionnaire was developed and tested with 15 household respondents to validate the questionnaire and design suitable survey strategy.
To measure the main factors and the partial constructs of each factor, we have inherited previous research and applied a 5-point Likert scales ranging from 1 to 5 points' Specifically, 1 -strongly disagree, 2 -disagree, 3neutral, 4 -agree, and 5 -strongly agree for each construct statement (table 1).
In phase 2, we identified the research sample, divided the investigation, performed the official investigation, cleaned and processed the data.The formal investigation and data processing procedures were conducted according to the guidelines of Hair et al (2013).
The study used the following formula to estimate the sample size: in which n is sample size, z is z-score, e is margin of error, p is standard of deviation.Supposed with a 90% confidence level, 50% standard of deviation and a 5% margin of error (for 90% confidence, z-score as 1.65 was used).The sample to ensure reliability was estimated at 438.In fact, the study collected 450 questionnaires from Hanoi respondents through 2 forms: face-to-face interview and online survey.
For face-to-face interviews, the study selected three districts in Hanoi representing different developmental stages to enhance the sample representativeness: developed, developing, and transitioning.The selected districts are Ha Dong (western part of Hanoi), Hoan Kiem (center of Hanoi), and Gia Lam (eastern part of Hanoi).These three areas represent diversity in terms of developmental levels, housing spaces, and income.In each district, the study selected three communes using random method.The sample size in each district was 270, specifically conducting 30 surveys per commune.To select households, we approached the People's Committee of each ward (local government) to collect a list of households in the ward.Each ward includes 10-15 residential groups.We randomly drew 3 residential groups in each ward and selected 10 households in each group using the random method (table 2).
Households were approached in the evening for interviews, which is the time the head of the household is usually available.When a household is absent, the research team will select the neighboring household to interview.Before answering the questions, the interviewee was specifically introduced to the objectives of the interview and the purpose of the research.At the same time, they were asked if they agreed and were willing to participate in the interview.All respondents agreed and were willing to participate and ticked into the form of agreement for interviews.
For the online survey, the survey questionnaire in the form of Google Forms was sent to 2 Facebook pages with a high number of participating residents in Hanoi, including the page of the Vinhomes Royal City urban area, the page of the Hanoi resident group giving exchange information about education and child rearing.The online form includes content about residence to filter out people in Hanoi.We have collected 180 online forms for analysis.The official questionnaire includes 4 main parts.
(1) Introduction to the objectives, content of the interview, respondents' consensus and interview process, (2) Information about people's awareness and attitudes towards environmental and energy issues reproduce (3) People's views and assessments on factors affecting their intention to use SWH and (4) Socio-economic information of households.Official survey was implemented from August to October 2023 in Hanoi.This study was reviewed and approved by the Scientific and Training Committee of the National Economics University, Vietnam which has the responsibility of academic ethics approval.

Characteristics of the survey sample
The below table showed main characteristics of the sample.It illustrates that females (57.11%) participated more in the survey than males (42.89%).The majority of respondents fell within the age range of 40-50 years (42.44%),followed by the 30-39 age brackets (27.33%), and those above 50 years were the least represented (14.22%).Regarding education, participants with a college or university education level (45.78%) dominated, followed by high school or below (31.78%), vocational training (12.22%), and post-graduation (10.22%).Additionally, managers constituted a substantial portion of the survey participants (41.11%).
The data also revealed that the income bracket between 10 to under 20 million VND accounted for a high proportion (47.78%), and a significant majority of survey respondents were married (88.22%).The distribution of participants among different districts was relatively even, with urban residents (52.89%) slightly outnumbering rural residents (47.11%) (table 3).

Cronbach's Alpha analysis
First, the study conducted Cronbach's Alpha analysis to evaluate the reliability of the scale with research factors.Statistical results from measuring 36 observed variables indicated that these variable values were rated within a range from 1 to 5 points.The mean value evaluated by survey respondents for each factor serves as the basis for assessing the importance level of factors influencing the intention to buy SWH.The mean analysis results showed that survey participants rated attitude toward SWH the highest (4.16), followed by product knowledge of SWH (4.05) as the second most important, perceived usefulness (4.03) as the third, and social influence (3.6) as the least important factor (table 4).
The first round of Cronbach's Alpha analysis demonstrated that Cronbach's Alpha coefficients for 8 study factors were greater than 0.6, the two variables SI3 and PP4 had corrected items-total correlation < 0.3, so they were eliminated as weak measuring the related factors (Hair et al 2013).The second round Cronbach's Alpha test showed Cronbach's Alpha coefficients for 8 factors greater than 0.6, and all the observed variables had corrected items-total correlation greater than 0.3.From this round, scales were reliable and well measured related factors; also observed variables were meaningful and valid for explaining dimension of the factors.Hence, from the initial 36 observed variables, Cronbach's Alpha test eliminated 02 observed variables, leaving 34 observed variables for the EFA (figure 4).

Exploratory factor analysis
EFA results using Principal Axis Factoring (PAF) extraction method and Promax rotation revealed KMO measure = 0.843 (> 0.5), Significant Bartlett's test = 0.000 (< 0.05).Thus, the EFA analysis was deemed appropriate for data group.Table 5 illustrated the results of factor rotation, wherein 34 observed variables were categorized into eight groups.The total variance extracted = 75.3%> 50%, indicating that these eight factors explain 75.3% of the data variance.Eigenvalues for each factor are notably high (>1), with the eighth factor having the lowest Eigenvalue at 1.264 > 1. Factor loadings are all above 0.5, and there were no instances where variables load significantly on more than one factor simultaneously.Hence, the factors maintain convergence and distinction in the EFA (Hair et al 2013).Additionally, there was no cross-loading, signifying that the questions within each factor remained distinct from those in other factors.Therefore, following the factor analysis, these factors remained intact, neither added to nor removed (table 5).
Then, the measurements for the eight factors were re-examined through Confirmatory factor analysis (CFA).The initial CFA revealed an issue of item overlap.The covariance of the error terms between variables AT1 and AT2 within the broader variable attitude showed a very high Modification Index (MI) score (at 123.21).These were subsequently combined to reduce MI (Hair et al 2013).The CFA results exhibited standardized loadings above 0.5, all statistically significant (p-values < 0.05), meeting the required model fit indices.Table below demonstrated that the factors' Average variance extracted (AVE) is above 0.5, indicating convergence.The P-values are all < 0.05, signifying that the correlation coefficients for each pair of concepts significantly differ from 1 at a 95% confidence level.Simultaneously, the correlations among the concepts are all below 0.85, ensuring discriminant validity (Hair et al 2013).The Composite Reliability (CR) values for all variables exceed 0.7, ensuring reliability (table 6).

Regression analysis
In the next step, a multiple linear regression analysis was conducted to test the research hypotheses.The result of the model displayed significant value of F test <0.05 showing that the regression model fitted data set (Hair et al 2013).It was shown that adjusted R 2 was equal to 0.703, which means 70.3% of the variation of the dependent variable was explained by the variation of the variables included in the model.The results also indicated 5 factors having significant impacts on intention to buy SWH including attitude, perceived usefulness, financial support, price perception and social influence in which attitude had the strongest influence (β = 0.202).Following this factor was perceived usefulness (β = 0.171) and price perception (β = 0.157).Financial support is the weakest factor affecting purchasing intention (β = 0.139).Product knowledge and perceive behavioral control did not have a meaningful impact on respondents' intention to buy SWH in this model (table 7).The standardized regression equation has the form: Checking the basic assumptions of the regression model showed that all VIF coefficients were less than 2, meaning there was no multicollinearity in the analysis.The Durbin Watson coefficient was 1.867, ranging from 1.5 to 2.5, so autocorrelation did not occur (Hair et al 2013).In addition, the residual distribution graph was symmetrical, meaning there was no uneven variance.The regression model validly explained the variation of intention to buy SWH (figure 5).
Finally, the study examines the differences between SWH consumption intentions according to people's socioeconomic characteristics.The ANOVA test results revealed no significant differences in intention between genders.In fact, there was one insignificant difference as men rated the perceived behavioral control factor higher than women, while women rated the social influence factor higher than men.Additionally, there were no significant differences in evaluating factors between urban and suburban dwellers.The study also did not find a relationship between occupation and intention to use SWH.However, there were differences in behavioral intentions to consume SWH between groups with different educational levels.The ANOVA analysis showed that individuals with higher education (college/university) rate perceive usefulness higher than those with other educational backgrounds.On the contrary, individuals with lower education levels (high school or below) significantly emphasized price perception factor.Finally, there was a significant relationship between income and consumption intention.ANOVA indicated that people in the highest income group have higher intentions than people in the lowest income group in using SWH.

Discussions
This study provided comprehensive insights into the factors consuming consumers' intention to purchase SWH in Hanoi, Vietnam.The results of multiple regression analysis showed that there are 5 factors that significantly influence people's intention to buy SWH including attitude, perceived usefulness, financial support, price perception and social influence.Firstly, attitude emerged as the most influential factor, indicating that consumers having more environmentally conscious are more likely to express intent to purchase SWH.This result is also consistent with the studies of Kuma, Ali et al (2019), Wang et al (2019) and Abdullah et al (2021) on the role of attitude in expected behavior.At the same time, the results also support the TPB (Ajzen 1991) that attitude is one of the most important components of behavioral intention.Attitude refers to an individual's subjective beliefs and views about a certain issue, specifically using SWH to protect the environment and bring convenience to the household.When attitudes improve, the ability to perform behaviors improves (Gadenne et al 2011, Chen et al 2016and Parkins 2018).Hence, management agencies should widely disseminate the benefits of SWH to foster a positive attitude among households, thereby increasing the intention to buy them.This solution was also proposed in research by Wang et al (2019) in China and Abdullah et al (2021) in Malaysia on promoting households' behavior of using solar energy equipment through communication and awareness rising.This finding reinforces the importance of raising awareness and education on environmental issues to promote sustainable consumption patterns.
Secondly, perceived usefulness is the second strongest factor on SWH consumption intention.The more people understand about the potential benefits of the SWH system, the more likely they are to use it.Hsu and Lin  (2008), Claudy et al (2011), Elmustapha et al (2018), Wang et al (2019) studies also had similar results when demonstrating the link between perceived usefulness and expected behavior.In general, when accepting a new technology, individuals must understand the superior aspects of that technology, especially the superiority compared to traditional technology or alternative technologies.In the case of SWH, this technology has outstanding advantages not only from a personal perspective such as convenience, reducing long-term electricity costs, meeting many daily needs but also has other benefits for society such as protecting the environment and reducing greenhouse gas emissions.The more people are aware of these benefits of SWH, the more they intend to consume (Zhang et al 2012, Venkatesh et al 2012, and Tsaur and Lin 2018).Therefore, manufacturers and sellers of SWH should enhance household understanding of these systems through events, direct marketing, personalized sales, or television advertising.Content should emphasize the exceptional features of SWH, such as electricity savings, environmental friendliness, absence of pollution, smoke, and noise, and user safety.
Thirdly, financial support and price perception are also important factors affecting SWH consumer behavior.This result is similar to previous studies of Carley et al (2013), Li et al (2013), Danielis et al (2020), Judson and Irakbash (2022).Households need initial financial leverage to invest in SWH systems.Therefore, they need economic encouragement from the state or equipment supply businesses.Some previous studies also showed that financial support is the most important factor leading to the purchase of solar energy equipment because it creates economic incentives for individuals and households (Hsu and Lin 2008, Claudy et al 2011, Danielis et al 2020, Ali et al 2019).For developing countries, this support is even more meaningful when the majority of people are in the middle income group, and when there is support in price or installation costs, it will be easier for families to new investment in equipment.Therefore, SWH manufacturing businesses can collaborate with banks or financial institutes to implement policies or action programs to encourage the installation of SWH in the future, such as capital policies (providing low-interest loans for purchasing SWH, installment plans, etc.).Additionally, it is essential to enhance households' awareness that 'SWH are a practical and cost-effective choice corresponding to their value' and 'the cost of purchasing and installing SWH is not high.'Furthermore, there is a need to bolster household' perception regarding 'economic considerations when using SWH' and 'the cost-effectiveness of using SWH.Additionally, to enhance household' intention toward installing and using SWH, manufacturers, businesses, or dealers should establish a rigorous after-sales service (customer service) and continuously improve models, varieties, quality, and scale while ensuring competitive pricing.This has also been pointed out in previous studies by Carperello in Italy, Hsu et al in China and Sadig in Pakistan.
Fourthly, social influence was also a factor that positively influences the intention to use SWH.Studies of Hsu and Lin (2008), Michelini (2012), Azeez and Atikol (2019), Kumar et al (2022) had similar results as this study with context in China, Malaysia and some countries around Vietnam.In Vietnam, where social capital and informal information networks are highly developed, this has important implications in promoting SWH consumption.In suburban community settings, the influence of peers, relatives, neighborhood, and social networks significantly impacts consumer intentions more than other factors.Apart from initiatives to increase households' attitude, enterprises selling SWH in Hanoi and Vietnam should put more concentrate on this social influence factors.Strategies might involve organizing events, forums, conferences, and seminars, providing a platform for businesses to present and promote their products, especially through a dense network of civic organizations in communities.
Last but not least, the study also discovered a significant relationship between residents' socioeconomic variables and their intention to purchase SWH, specifically, education level and income have a positive impact on intention to buy.This result is consistent with researches by Chen et al (2016), Wang et al (2019) and Ali et al (2019) on the impact of education and income, especially the difference between the behavioral intentions of the highest and lowest income groups.However, age and gender variables do not have an impact on behavioral intention.This result is contrary to Parkins (2018) and Danielis et al (2020).In Vietnam and Hanoi in particular, men and women of different ages have good access to information through media and social networks, so they can have relative uniformity awareness in consumption trends or more specifically, household appliances such as water heaters.Therefore, there may be no meaningful difference in behavioral intention between men and women as well as different age groups.
Although five factors have been identified that have a significant influence on Hanoi residents' intention to use SWH, there are a number of other factors that may have a potential impact but have not been included in the analytical model, which may include: The first is the variable 'compatibility', defined as the degree to which a given technology or service is perceived as stable, consistent, or consistent with users' existing beliefs and experiences (Tsaur and Lin 2018).Current research on technology acceptance and adoption shows that the level of compatibility is considered an influencer of expected performance, and plays a mediating role in the technology acceptance model as well as regulatory role in various contexts (Saleh et  Second, the variable that can have a potential impact on consumption intention is performance expectancy.This is the extent to which the use of new technology can provide consumers with expected benefits when performing specific activities (Venkatesh et al 2012).In studies related to renewable energy technology, performance expectancy is a factor identified as influencing intention to use (Yuan et al 2011, Sanguinetti et al, Urpelainen andYoon 2017).Huang (2019) suggest that performance expectancy is the best predictor of technology use.Since the higher the efficiency when applying technology, the more consumers' intention to use it increases.Users will use and accept new technology if they believe that the new technology is more useful and valuable in their daily lives (Elmustapha et al 2018).
Third, the innovation variable can also be a potential impact factor.Davis (1989) defined innovativeness as 'the degree to which an individual or organization typically adopts a technology earlier than other members of the same system'.Tsaur and Lin (2018) believed that innovativeness is 'a common and fundamental personal characteristic, which can also be understood as the degree of willingness to innovate'.In the field of technology, Hsu and Lin (2008) emphasized that innovation is a personal quality and has a great influence on intention to use in the model.Besides the direct impact on intention to use, innovativeness also impacts expected performance (Venkatesh et al 2012, Saleh et al 2014).The more innovative users are, the easier it is to access and understand the effectiveness of technology.

Conclusions and implications
Research on the behavior of using renewable energy equipment has not been conducted in detail and widely in Vietnam.To fill this gap, this study identified and evaluated factors affecting the intention to use SWH of Hanoi city residents.Through exploratory factor analysis and regression based on TAM and TPB, we have identified 5 main factors that significantly affect residents' intention to consume SWH.Two factors including product knowledge and perceive behavior control were found to not affect the intention to consume SWH.
People's attitude is the most important variable that determines their intention and behavior to consume green equipment.It is how individuals perceive and approach relevant aspects of SWH.It includes internal thoughts, beliefs and knowledge that determine external actions.When attitudes improve, people will have the intention and tendency to choose green devices.Price is also a factor that can determine SWH usage behavior.Currently, the price of SWH is still higher than traditional alternatives.This is also a barrier to consumers' motivation to use SWH.This needs to be improved soon to promote greener behavior through individual economic incentives.
In addition, consumers also appreciated and expected support from regulatory agencies and businesses in accessing and using this new technology.They also have a clear attitude that consuming renewable energy will contribute to protecting the environment, combating urban air pollution and better responding to the challenges of climate change.Research also shows that social influence plays an important role in shaping consumer behavior.Vietnam is a country with high social capital thanks to the wealth of civil organizations and social relationships.People are influenced on their behavior by the influences of friends, relatives, colleagues and social organizations.
Management implications can come from the government and business sectors to promote SWH consumption.For the state, it is possible to continue developing laws and regulations on renewable energy in general and renewable energy equipment in particular.The government can consider economic incentives such as tax reduction for businesses producing and supplying SWH, and reducing loan interest rates is also a solution applied by many countries.In addition, regulatory agencies can provide support for production infrastructure and promote the environmental benefits of green equipment products in cities, schools and workplaces.The government can also provide price support to consumers such as reducing VAT and consumption tax to encourage their consumption behavior.For businesses, increasing customer awareness and attitudes is a key factor leading to changing SWH consumer behavior.Therefore, businesses need to have long-term and continuous strategies on propagating and promoting the benefits of consuming SWH to people.This can be done through traditional advertising channels but also through traditional advertising channels can be done through social networks.Vietnam has a young population structure, quite fond of technology and popular use of social networks, propaganda about SWH on social networks will be easier to reach people, and more updated to help them change attitudes about green products and SWH.In addition, businesses can also approach civic organizations to propagate and disseminate the benefits, usage and superiority of SWH to potential customers through these organizations.
This study has some limitations that should be explored in further research.Firstly, the sample includes households that are currently using or have not used but intend to use SWH.The small sample size may limit the ability to generalize the results found in the regression model.For this reason, future research could be conducted with a larger sample size and select households that actually use SWH.This allows the assessment of factors affecting actual consumption behavior of green products, rather than just being limited by consumption intention.In addition, the research can be expanded to other cities in Vietnam to confirm these research results as well as find new potential factors.Future research should include some new factors, the expected effect of product usefulness, brand loyalty or the appeal of alternative devices.In addition, the analytical model can be more complex and comprehensive when dividing factors affecting behavior into different layers and considering the impact of each layer of factors on previous layers with the SEM model rather than a regression model.
al 2014, Wang et al 2019, Liobikienė et al 2021).The above studies have basically demonstrated the important role of compatibility level in positively promoting behavioral intention to use technology through improving the fit between technology and related performance of behavior.
al 2019, Danielis et al 2020).If individuals believe they possess the requisite knowledge, skills, and resources, they are more likely to purchase renewable technology and vice versa (Saleh et al 2014, Elmustapha et al 2018).
Hypothesis 2 (H2): Product knowledge positively influences the households' intention to buy SWH.

Table 1 .
Variables and statements for measuring obstructs.

Table 2 .
Direct survey sample distribution.

Table 3 .
Descriptive statistics of the survey sample.

Table 4 .
Results of second round of Cronbach's Alpha analysis.

Table 6 .
The reliability and extracted variance of factors.

Table 7 .
Regression model analysis results.