INDONESIAN MILLENIALS ONLINE SHOPPING BEHAVIOR

The aims of this study were to investigate: (1) Perceived risk by consumers influencing on consumer online shopping intention. (2) Consumers trust influencing consumers’ online shopping Intention. (3) The difference between hedonistic shopping motivation and utilitarian shopping motivation as moderation of the influence of perceived risk by consumers in online shopping intentions. Between hedonistic and utilitarian shopping motivation which variable is the stronger? (4) The difference between hedonistic shopping motivation and utilitarian shopping motivation as moderation of the influence of consumer trust in online purchase intentions. Between hedonistic and utilitarian shopping motivation which variable is the stronger? The data in this study is obtained from a questionnare distributed to 210 responden. The method used in sampling is purposive sampling addressed to the whole millennial generation in Indonesia who have had online shopping respondents. The method through the partial least squares (PLS) program, SmartPLS2.0 and t test. The results showed that perceived risk of consumers has a negative effect on shopping intention, and consumer trust has a positive effect on shopping intention. Futher the hedonistic shopping motivation is the stronger variable than utilitarian shopping motivation.


INTRODUCTION
Nowadays, business development is inseparable from the important role of information technology. According to Nisafani et al. (2017) the advancement of information technology has produced many innovations by emerging competitive business environments that once never existed, internet-based online for example. Online shopping or e-commerce is a purchasing process that involves transactions between seller and customer through digital platforms on the internet (websites, applications, etc.) anywhere and anytime (Turban et al., 2008).
Millennial is the youth generation characterized as well-adapted technology users in their daily lives, values, life experiences, motivations, and purchasing behavior in general (Moreno et al., 2017). Indonesian Millennials are now a growing number, currently, at 103 million (Bappenas, 2018). They are looking for the latest products or services through shopping online.
This study use several dependent variables, independent variables and moderating variables using combined theoretical models of theory of reasoned action by Fishbein and Ajzen (1975) and technology acceptance model by Davis (1989) to address online shopping intention variables, then Barnes and Vidgen (2006) model to address trust variable, perceived risk variable using the Bauer (1960) model, and consumer online shopping motivation variables of hedonistic and utilitarian motivations (Beldona et al.,2005).
According to Ajzen and Fishbein (1980) shopping intention is a state before an individual carries out an action which can be used as a basis for predicting the behavior. This statement is supported by the research of Cry (2008), shopping intention is a consumer's desire to buy products or services from a website. The involvement of consumer in online shopping has become an important trend. Thus, it is very important to identify the determinants of consumer's intention in an online purchase, (Ling et al., 2010). This Journal is licensed under a Creative Commons Attribution 4.0 International License According to Pavlou (2003); Forsythe and Shi (2003), online shopping is considered riskier, therefore risk and trust hold an important role in online transactions. The perceived risk by consumers regarding the uncertainty of the consequences of transactions because of unknown factors that lead to unexpected results (Bauer, 1960). According to Kim et al. (2008), the perceived risk by the consumer is the worrisome thought of uncertain potential negative result from online transactions. Some studies Acosta et al. (2014) found that the perceived risk by consumers negatively affects the intention of online shopping.
Consumers' trust fosters intention to buy a product online with trust or expectation that the seller will keep the deal and consumers will not suffer for any loss (Jarvenpaa et al., 2000). Many studies have proven that trust affects online shopping intention (Chin et al., 2018;Batbayar et al., 2018).
Consumers' shopping motivation has two types based on the consumption values which are hedonistic and utilitarian values (Babin et al., 1994). There have been many studies that focused on hedonistic and utilitarian shopping motivation as a determinant factor of intention in online shopping (To et al., 2007). Hedonistic shopping motivation has a positive effect on online shopping intentions as well as utilitarian shopping motivation on shopping intention in online shopping (Childers et al., 2001). However, there are two differences found. First regards on hedonistic and utilitarian shopping motivations as moderating influence of perceived ease of use on user attitude and the effect of perceived ease of use on user attitude, which coming stronger from the hedonistic shopping motivation compared to the utilitarian shopping motivation (Wang et al., 2015).
Millennials are the strong determinant of online shopping behavior and contribute to the highest percentage of online purchases (Dhanapal et al., 2015), so the object of this research was the millennial generation in Indonesia. Today's entrepreneurs must be aware of this influence when designing modern marketing strategies. They also have to offer products and services that are compatible with the millennium segment (Maryoto, 2018). Despite the increasing total of spending on the internet, many websites fail to create positive experiences for customers to buy products (Hausman and Siekpe, 2009).
The topic has been long investigated in many studies with two independent variables which are consumer trust and consumer perceived risk, and one dependent variable which is shopping intention of online shopping consumers. This study also used consumers' shopping motivation (hedonistic and utilitarian shopping motivations). However, this study is quite different from previous research which has not used the consumer shopping motivation variable as a moderation of the influence of consumer trust in consumers' shopping intentions and the effect of consumers' perceived risk of online shopping. This research is important to do because its findings will contribute to the strategy of what entrepreneurs should do to increase demands in online shopping for millennial generation in Indonesia. This increase in sales will also reduce unemployment.

LITERATURE REVIEW
Shopping intention according to Schiffman and Kanuk (2007) is one of the psychological aspects that has a considerable influence on attitudes and behavior. It is the tendency of consumers to buy a brand or take actions related to purchases that are measured by the level of the possibility of consumers making purchases (Assael, 2001).
The intention of online shopping is influenced by the determination of consumers to buy from e-commerce sellers (Salisbury et al., 2001). If consumers are more familiar with e-commerce sellers, they are more likely to access websites with the intention to buy (Forsythe and Shi, 2003). Online shopping intention is a significant predictor for actual purchases and is used to achieve actual purchase goals (Lee and Lee, 2015). Consumers' shopping intention is an online segment that is considered a predictor of purchasing behavior (Pavlou, 2003;Chen et al., 2010;Kim et al., 2008). Shopping intentions in an online segment will influence consumers' desire to purchase from certain websites (Chen et al., 2010). This study used indicators: to be willing to purchase, to be possible to purchase goods in the future, to purchase goods and get as it is expected, to purchase the goods seen on the web in the near future, to purchase goods only if needed, and to be willing to recommend to friends and family (Pavlou, 2003). Schiffman and Kanuk (2008) state that risk is a situation where decision makers have knowledge of the consequences of loss. The theory of risk perception, since 1960 has been applied in decision making to explain consumer behavior (Mitchell, 1999). According to Forsythe and Shi (2003), consumers consider online shopping risk to be higher than offline shopping because the internet is open and complex and technology is beyond the users' control. The perceived risk is consumers' mental thought of a potential negative uncertain result from online transactions (Kim et al., 2008). According to Chang (2008), perceived risk is seen as a level of uncertainty related to conditions when a consumer does online shopping. Consumers feel risky in purchasing goods from the internet (Kuhlmeir and Knigt, 2005;Dennan et al., 2006;Bianchi and Andrews, 2012).
There are three types of dominant risks in online shopping, namely financial risk, product risk and information risk (security and privacy) (Bhatnagar et al., 2000). Indicators of perceived risk in this study are: the decision to download applications at risk, online purchases from websites are at risk because the products/ services provided may fail to meet expectations, risky because the products/services provided may be low quality, risky because they can cause financial losses, risky because it can be time-consuming., high potential risk in installing an application, subjective to product risks, and money loss (Pavlou and Gefen, 2004;Kim et al., 2008). Many empirical studies found negative influence between the perceived risks by consumers and the intention to online shopping (Chang and Chen, 2008;Mohseni et al., 2016;Indiani et al., 2015). This study believes that perceived risk by consumers affect the intention in online shopping, therefore the hypothesis used is as follows: H 1 : Perceived risks by consumers negatively affect consumers' purchase intentions Trust has been identified as the most important supporting factor in e-commerce (Qureshi et al., 2009). It is more important for online than offline retailers because consumers feel more risk in electronic commerce because of their inability to visit physical stores and check the products they intend to buy (Li et al., 2014). Trust encourages upcoming online transaction activities (Shen, 2011). The future of online shopping depends on trust (Wang and Emurian, 2005). The intention to do online shopping depends on trust (Ponte et al., 2015). Indicators of consumers' trust used in this study are the competent website, reliable website, excellentservice-provided web site, stick with the deal/credible website, good-intention and consumers-friendly website, and benevolentintention website (Gefen et al., 2003).
The higher customer trust, the more demands in online shopping (Thamizhvanan and Xavier, 2012). Trust has a positive correlation and has a strong influence on the shopping intention variable (Kim et al., 2008). The results of the empirical study found an influence between consumer trust and intention to online shopping (Harris and Goode 2010;Shankar et al., 2002;Pappas, 2016). This study believes that consumers' trust affects the intention of online shopping, therefore the hypothesis used is as follows: H 2 : Consumer trust has a positive effect on online shopping intention.
According to experts, shopping motivation is driven by hedonistic factors and utilitarian factors (Arnold and Reynold, 2003;Kim, 206;Babin et al., 1994). Motivation is the desire in individuals which forces them to act (Schiffman and Kanuk 2008). Consumer spending behavior is influenced by hedonistic and utilitarian shopping objectives (Chitturi et al., 2008;Kim and Shim 2002). According to Hirschman and Holbrook (1982), the value of hedonistic consumption shows some consumer behavior aspects related to multi-sensory, fantasy and emotional aspects of consumer's experience on the product (Babin et al., 1994). Utilitarian shopping value is usually related to customer needs to complete shopping objectives (Holbrook and Hirschman, 1982).
Hedonic and utilitarian values are the classification of dichotomized consumer values (Childers et al., 2001;Eroglu et al., 2005). The difference in hedonistic/utilitarian systems comes from the concept of the utilitarian and hedonistic product of consumer behavior literature (Hirschman and . Heijden (2004). The study of consumer behavior shows that the utilitarian attitude towards the internet is that users use web application for a certain purpose.
Hedonistic Value Indicators used in this study: shopping for fun, experience, freedom, problem breakout, and the embodiment of fantasy (Babin et al., 1994;Hirchman and Holbrook, 1982). The indicators of utilitarian values are task, cognitive, and efficiency (Babin et al., 1994;Kim, 2006). In internet segment, one web application can be used for both utilitarian or hedonistic purposes and these might moderate user acceptance of this application (Childers et al., 2001).
In the study of the Luk et al. (2013), it was obtained that utilitarian shopping motivation is stronger in the correlation of sacrifice, value, satisfaction, and behavioral intentions. While regarding service quality, value, satisfaction, and intention to behave, hedonistic shopping motivation holds the stronger role. This study model will cover trust, risk, and shopping intention to explore the moderating influence of utilitarian/hedonistic motivation in Web 2.0 applications. There are no studies which use shopping motivation (hedonistic and utilitarian) as a moderating variable from 1 the influence of the risk perceived by consumers on online shopping intention, 2 the influence of consumer trust in online shopping intention. This study argues that the hedonistic and utilitarian shopping motive can be used as moderation, therefore the hypotheses used are as follows: H 3 : There is a difference between the hedonistic and utilitarian shopping motivation as a moderation of the influence of the perceived risk of consumers in online shopping intention. Between hedonistic and utilitarian shopping motives which variable is the stronger? H 4 : There is a difference between the hedonistic and utilitarian shopping motivation as a moderation of the influence of consumer trust in online shopping intention. Between hedonistic and utilitarian shopping motives which variable is the stronger?
Based on previous theoretical studies, the framework for this study can be illustrated in Figure 1 as follows:

Research Design
The study was carried out using a survey method with an onlinedistributed questionnaire.

Sampling Technique
The population in this study is all Indonesian millennial internet users. The method used in sampling is Purposive Sampling addressed to the whole millennial generation in Indonesia who have had online shopping. The questionnaires distributed were 237 from November 23, 2018 to December 9, 2018, but there were only 210 questionnaires used in this study.

Validity test
This study uses convergent validity: a convergent validity test in the PLS with reflective indicators is assessed by the factor loading (correlation between scores item/component score with the contruct score) of indicators that measure the contruct. A factor loading value >±0.30 indicates that it complies with the minimum level, where as a loading factor value of ±0.40 is considered better and in accordance with the rules of thumb used by the researchers. A loading factor of >0.50 is considered significant (Hair et al., 1998).
The rule of thumb used by Chin (1998) for convergent validity is the outer loading of >0.7, communality of >0.5, and average variance extracted (AVE) of >0.5. perceived risk, trust, shopping intention have a factor loading value of more than 0.7, as is seen in Table 1. The value of the AVE and communality of more than 0.5 is also seen in Table 1. Based on Chin (1998),it could be concluded that all of the indicators are valid.

Reliability test
The reliability test in PLS uses 2 methods: The Cronbach's alpha and composite reliability. The composite reliability alpha value should be greater 0.7, although a value of 0.6 is acceptable (Hair et al.,1998). The Cronbach alpha values and composite reliability values for each of the variables in this study are above 0.7, as is seen Table 2. It can therefore be concluded that the instrument used in this study is reliable.

Method of Analysis
Data analysis used was inferential statistical analysis. The one used in this study was structural equation modeling, here in after referred to SEM, which was operated through the PLS program, SmartPLS2.0 and t test

DISCUSSION AND RESULTS
This discussion is focused on four tests of the hypotheses put forward by the objectives of this study, which have been tested through validity and reliability tests. Furthermore, in this study hypothesis testing was done through (1) structural equation modeling (hereinafter referred to as SEM, which was operated through the partial least squares/PLS method, namely SmartPLS 2.0. and 2) differential tests (t-test). The results of data processing for SmartPLS 2.0 are shown in Figure 2 and different tests (t-test) are displayed in Table 3.

The Influence of Perceived Risk by Consumers on Consumer Online Shopping Intention
Hypothesis 1: Perceived risk by consumers has a negative effect on consumer shopping intention. The results obtained by betacoefficient value of −0.116 and t value of −2.375, t table of −1.64 (one-sided hypothesis) and the P = 0.003 indicate the perceived risk of consumers has a negative effect on shopping intention. This finding supports the several studies in prior which claim that the perceived risk negatively affect online shopping intention (Kim and Lenno, 2013;Mohseni et al., 2016). The perceived risk study shows that the main source is uncertainty. According to Chang (2008), perceived risk is seen as a level of uncertainty related to conditions in online shopping.
Consumers develop or adopt strategies to reduce risk (Bauer 1960), thus consumers adopt information handling as a strategy for risk reduction; either they seek new information or they refer and evaluate existing information (Cox, 1967). Vijayasarathy and Jones (2000) state that greater perceived risk by consumers will cause consumers to be reluctant to do online transactions because the perceived risks by the consumer is seen as the potential negative results of online transactions (Peter and Ryan, 1976). Therefore, it can be concluded that online sellers must reduce the risk by providing guarantees, and security so that consumers (Indonesian Millennials) will increase their intention to purchase online.

The Influence of Trust on Consumer Shopping Intention
Hypothesis 2: Consumer trust has a positive effect on consumer shopping intention. The results of the study obtained the betacoefficient value of 0.491 and t value of 6.427, t table of 1.64 (one-sided hypothesis), the P = 0.000 indicating the consumer trust has a positive effect on shopping intention. These findings support the results of several studies of Eastlick et al. (2006) and Yoon (2002). Consumer trust affects shopping demands. According to Jarvenpaa et al. (1999) definition of trust in the context of online shopping is the consumers' willingness to rely on sellers and take action under circumstances where that action could make consumers be vulnerable (sensitive) to the seller.
Trust is considered an important basic element in online trading (Gao et al., 2002). Trust is defined in online transactions as a combination of trust in online trading and trust in online sellers (Brannigan and De Jager (2003). If consumers have trust in online sellers, they spend less effort looking for information about online sellers and carrying out online transactions (Kim et al., 2012).
Customers may not shop online if they do not trust the website they are shopping for (Kim et al., 2011;Jarvenpaa et al., 1999). Trust in online purchases is often interpreted as promises and guarantee of online sellers to provide high-quality products or services to online buyers (Cowcher, 2001). Thus, it can be explained that online sellers must increase trust for consumers (millennial generation) in order to increase online shopping intention.

The Difference between Hedonistic Shopping Motivation and Utilitarian Shopping Motivation as a Moderation of the Influence of Perceived Risk by Consumers in Online Shopping Intentions
Hypothesis 3: There is a difference between hedonistic shopping motivation and utilitarian shopping motivation as a moderation of the influence of perceived risk by consumers in online shopping intentions. The results of the study obtained t-test values of −14.824, t table values of −1.64 (one-sided hypothesis). The results show there are differences in hedonistic shopping motivation with utilitarian shopping motivation as a moderation of the effect of risk perceived by consumers in online shopping intentions. Whereas the strong variable between hedonistic and utilitarian shopping motivations can be seen in Table 3.Since t1 represents hedonistic and shows positive value thus it can be concluded that the stronger motivation is hedonistic shopping motivation. This means that the millennium generations in Indonesia who shop online have hedonistic shopping motivation, shopping for fun, seeking experience, freedom, problems breakout and fantasy embodiment (Babin et al., 1994;Hirchman and Holbrook, 1982).
According to Dowling (1994), perceived risk can be explained as consumer awareness about insecurity and contradictory consequences as a result of buying a product or service. Consumer perceived risk is an important barrier for online consumers who are considering whether to make an online purchase or not. Product risk is associated with the product itself, for example, the product might turn out to be defective. Financial risks including opportunity and time costs are not related to products but with marketing channels (internet), for example, online transactions can be duplicated due to technical errors or double-click unwanted   purchase buttons. Information risk is related to transaction security and privacy; for example, the requirement that consumers send credit card information over the internet can cause concern because of the possibility of credit card fraud (Grady, 1997).

The Difference between Hedonistic Shopping Motivation and Utilitarian Motivation as a Moderation of the Influence of Consumer Trust in Cunsumers Online Shopping Intentions
Hypothesis 4: There is a difference between hedonistic shopping motivation and utilitarian motivation as a moderation of the influence of consumer trust online shopping intentions.
The results of the study obtained the value of t test of 6.163, the value of t table of 1.64 (one-sided hypothesis). This means that the value of t test > t table value, which indicates that there are differences in hedonistic shopping motivation with utilitarian shopping motivation. Whereas the stronger variable between the hedonistic and the utilitarian shopping motivation influences on consumer trust in shopping intention can be seen in Table 3.
Since t1 represents hedonistic and it is positive, it is concluded that the hedonistic is the stronger variable. This means that the Millennials in Indonesia who shop online have hedonistic shopping motivation, which means shopping for fun, seeking experience, freedom, problems breakout and fantasy embodiment (Babin et al., 1994;Hirchman and Holbrook, 1982).
The higher consumer trust will lead to an increase intention to shop online (Thamizhvanan and Xavier, 2012). According to Corbitt et al. (2003). Trust is the most important factor influencing online shopping intentions. Previous studies have recognized that electronic purchasing decisions are inherently risky, and therefore trust can be an important factor in giving consumers the confidence they need to be involved in the transaction (Zhang, 2003). Trust in a website plays an important role in e-commerce, because consumers may not do online shopping if they don't trust the online sellers where they shop (Kim et al., 2011;Ponte et al., 2015;Jarvenpaa et al., 1999).

CONCLUSION
Millennials are the unique generation, tend to spend their income through online shopping in their daily lives. Millennials of Indonesia always develop themselves to anticipate change as a result of technological advances. This is an opportunity for online sellers to increase the online purchase intention of the millennial generation in Indonesia.
The results of this study can be used by online sellers to increase the online purchase intention of Millennials in Indonesia. Thus, it is suggested to online sellers to: (1) The perceived risk by the millennial generation must be reduced. Online sellers must provide an easily-accessible application and the goods/services provided must meet the expectations of the millennial generation. The goods and services should support hedonistic shopping motivation (shopping for fun, seeking experience, freedom, problems breakout and fantasy embodiment), the goods must meet the millennial generation standard and quality, the price must be competitive with good shipping cost and no defects are received by millennial generation as the consumer.
(2) The millennial generation's trust must be increased. Online sellers must keep promises and provide commitments such as providing high-quality goods and services, meeting the needs and demands of Millennials, and uphold honesty.