THE INTENTION TO USE E-COMMERCE TO PURCHASE GREEN COSMETICS WITH A MODIFIED UTAUT2 APPROACH

: The quantitative research examines factors affecting the intention to use e-com-merce by employing a modified UTAUT2 model. The sample is 263 e-commerce users in Surabaya, Malang, and Kediri who have e-commerce account and know about Sensatia Botanicals. The Data was analyzed using PLS (Partial Least Square) in SmartPLS 3.0. The results of the analysis indicate that performance expectancy, facilitating conditions, social influence, habit, perceived playfulness and price value have a significant positive influence on the behavior intention to use e-commerce, and that effort expectancy and hedonic motivation do not have any significant positive influence. Gender difference in this study can increase the intention of individuals to use e-commerce on the price value factors, while age and income difference in this study neither increase nor decrease the intention of individuals to use e-commerce to purchase the products of Sensatia Botanicals.

People can shop online through websites or applications that can be accessed with a smartphone with technology development in industrial era 4.0. According to the statistical data throughout 2017 in Indonesia, IDR 34.5 billion was spent buying beauty and fashion products online (Dinisari, 2018). Cosmetics have become a need for almost every woman to support her appear ance (Fauzi et al., 2018). How-ever, as time goes by, cosmetics needs are not only for women but also for men. In the middle of environmental damage issues, green cosmetics of environmentally friendly begin to appear. The emergence of green cosmetics is based on public awareness of the environment and the importance of healthy cosmetics to not cause skin problems when the user applied (Sindo, 2012). Sensatia Botanicals is a brand of green cosmetics products from Bali that got an award for "The best eco-Beauty Brand in 2019" because the products are eco-friendly, get certified halal, and certified GMP (Green Manufacturing Product), and it has to benchmark on 40 countries. E-commerce is defined as a place where there is an opportunity to buy or sell products or services (Kozinets et al., 2010). E-commerce is expected to increase sales of Sensatia Botanicals. It eases customers to shop and use products because they are located far from the store, considering that Sensatia Botanicals still has 16 (sixteen) official stores located in Jakarta, Bandung, and Bali. Purchasing Sensatia Botanicals products can be conducted by accessing the website (sensatia.com) and several other e-commerce sites, such as blibli.com, JD.id, Lazada, Lemonilo, Sephora, Shopee, Sociolla, Tokopedia, and Zalora.
The success and failure of e-commerce implementation are very dependent on user acceptance, which in turn affects the improvement in service quality and sales results of Sensatia Botanicals. Some people tend not to adopt online shopping behavior, as evidenced by the Indonesian Internet Service Provider Association Survey (APJII), there are 12.2% of internet users who do not want to shop online because they have not been able to use the application, 9.5% of them worry that goods will not arrive, and 9% others find it too complicated to shop online if they have to transfer (Annur, 2019). However, many people also choose to shop through online shopping applications (e-commerce) to take advantage of attractive offers obtained from product purchases through the application. Therefore, identifying factors that influence consumers in using e-commerce to purchase a product becomes necessary and very important.
Even though people have begun to be aware of the importance of protecting the environment, consumption patterns do not necessarily change using green products. Furthermore, Indonesia is a potential market for cosmetics, but green cosmetics have not been widely used because of the lack of consumer knowledge about environmentally friendly cosmetics (Febrya, 2016). Awareness and concern for the environment make consumers begin to be intentioned in using environmentally friendly products. Consumer intention in a product can encourage the purchase of these products. Many factors can influence consumer behavior while buying a product, both internal and external factors (Kotler and Keller, 2012).
In this study, researchers used a modified Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) approach with adding perceived playfulness. The intention in using technology is also influenced by perceived playfulness which can arise when conducting product purchase transactions through e-commerce. This study aims to determine the factors affecting the formation of an intention to buy green cosmetics (Sensatia Botanicals) through e-commerce using a modified model of UTAUT2.

LITERATURE REVIEW Green Cosmetic
A green product is defined as a product that must consider environmental aspects in the product life cycle to minimize negative impacts on nature (Junaedi in Shaputra, 2013). It means that green cosmetics are made from ingredients that are not harmful in both production and consumption to minimize negative impacts on the environmental environment.

E-Commerce
Electronic commerce (e-commerce), based on Laudon and Laudon (2009), is a process of buying and selling products electronically by consumers and from company to company with the internet as an intermediary for business transactions. According to Hoffman and Fodor (2010), e-commerce can work well if run based on the 4C principle: connection, creation, consumption, and control.

Unified Theory of Acceptance and Use of Technology (UTAUT)
The research model of UTAUT was developed by Venkatesh et al. (2003). It combines eight theories of acceptance of information technology, which are Theory of Reasoned Action ( UTAUT2 model is the development of UTAUT, which learns about acceptance and use of technology in the customer's context. The UTAUT model has four constructs in determining user behavior intentions, including performance expectancy, effort expectancy, social influence, and facilitating conditions (Venkatesh et al., 2003). In contrast, UTAUT2 has added three new constructs: hedonic motivation, price value, and habit. In UTAUT2, individual differences such as age, gender, and experience are used as moderation variables in influencing behavioral intentions using technology.

The Effect of Performance expectancy (PE) on behavior intention (BI)
Performance expectancy is the extent to which the use of technology can benefit consumers in conducting activities (Venkatesh et al., 2003).  found a positive relationship between performance expectancy and intention for M-banking. Other research conducted by Sundaravej (2010), Ghalandri (2012), Borrero et al. (2014), Escobar-Rodríguez and Carvajal-Truzillo (2014), and Osakwe et al. (2017) also found significant results in the relationships of performance expectancy and behavior intention. Venkatesh et al. (2003) argued that the relationship between the performance expectancy and the behavior intention is moderated by gender and age, where the influence is stronger in males and younger individuals (Venkatesh et al., 2003). The research results by Rezvani et al. (2013) showed that income positively affects the intention to Herbalife products. Based on some related research, the H1 can be concluded in this study: H1a : Performance expectancy (PE) has a significant and positive effect on behavior intention (BI). H1b : Performance expectancy (PE) affect behavior intention (BI) and it is moderated by gender, age, and income Venkatesh et al. (2003), Effort expectancy is the level of ease the use of technology can reduce efforts (labor and time). Leong et al. (2013) argue that users become more skilled and adopt technologies such as shopping applications when easier to use. The positive relationship between effort expectancy and behavior intention has been confirmed by Escobar-Rodríguez and Carvajal-Truzillo (2014) Mahendra and Ardani (2015) shows that demographic factors consisting of age, education, and income have a positive and significant effect on buying cosmetic products of the body shop in Denpasar. Thus H2 hypotheses in this study are: H2a : Effort expectancy (EE) has a significant and positive effect on behavior intention (BI) H2b : Effort expectancy (EE) affect behavior intention (BI) and it is moderated by gender, age, and income

The Effect of Social influence (SI) on behavior intention (BI)
Social influence is defined as the personal welcome of a consumer who considers that another person is trustworthy and influential in using a new system (Venkatesh et al., 2003). The research on marketplace adoption shows that users will use the application if they are influenced by environmental factors such as friends, relatives, or opinions . Social influence is believed to have a significant impact on intention (Borrero et al., 2014;Escobar-Rodríguez and Carvajal-Truzillo, 2014;Osakwe et al., 2017;Phang et al., 2018). Social influence influences can vary between age and gender. Thus the H3 of this study are: H3a : Social influence (SI) has a significant and positive effect on behavior intention (BI). H3b : Social influence (SI) affects behavior intention (BI) and it is moderated by gender, age, and income.

The Effect of Facilitating conditions (FC) on behavior intention (BI)
Facilitating conditions are important for consumers in conducting transactions using e-commerce. In research conducted by Sedana and Wijaya (2010), Ghalandri (2012), Escobar-Rodríguez and Carvajal-Truzillo (2014), and Shang et al. (2017), facilitating conditions had a positive and significant correlation to behavior intention. Curtis et al. (2010) found that the influence of facilitating conditions on behavior intention was greater than that of women compared to female males. The relationship facilitating conditions to behavior intention is more necessary in older people because older age is harder to process new or complicated information when studying new technologies (Venkatesh et al., 2003). Thus the H4 of the study are: H4a : Facilitating conditions (FC) have a significant and positive effect on behavior intention (BI). H4b : Facilitating conditions (FC) affect behavior intention (BI) moderated by gender, age, and income.

The Effect of Habit (H) on behavior intention (BI)
Habit can determine the intent of user behavior in using technology (Venkatesh et al., 2012). Along with the increased experience in using technology, users began to use the technology with the usual (Venkatesh et al., 2012). Research by Escobar-Rodríguez and Carvajal-Truzillo (2014), Nuriska et al. (2018), and Yusuf and Indrawati (2019) stated that habit has a significant and positive influence on the behavior intention to use technology. Venkatesh et al. (2012) argued that the influence of habit differs among users of different ages and genders. According to Putranto and Yudi (2015), in their research, the variable income moderation influences the habit related to the behavior intention Wifi service. Based on several related studies, the researchers formulated the H5 hypothesis from the study: H5a : Habit (H) has a significant and positive effect on behavior intention (BI). H5b : Habit (H) affect behavior intention (BI) and it is moderated by gender, age, and income. Wang et al. (2008) describe playfulness as a state of mind. Wang et al. (2008) showed that perceived playfulness had a significant influence on the intention of students to use learning applications. Another study showing a significant influence of perceived playfulness on behavior intention by Moghavvemi et al. (2017) and Moridis et al. (2017). According to Terzis and Economides (2011), perceived playfulness can be felt from various age groups, where users (both young and old) will use a website-based system when they have fun and have clear content. Other researchers argue that perceived playfulness effects differ by gender. Papastergiou and Solomonidou (2005), in Moghavvemi et al. (2017), argue that women than men more perceive the use of the internet for fun or entertainment. Thus the H6 of this research are: H6a: Perceived playfulness has a significant and positive (PP) effect on behavior intention (BI) H6b: Perceived playfulness (PP) affect behavior intention (BI) and it is moderated by gender, age, and income.

The Effect of Hedonic motivation (HM) on behavior intention (BI)
Hedonic consumption is symbolized as a behavioral aspect closely related to consumption's multisensory, fantasy, and emotive aspects. As conducted by Van der Heijden (2004), some research found that hedonic motivation (conceptualized as a perceived enjoyment) influenced the intention of using technology. The research by Escobar-Rodríguez and Carvajal-Truzillo (2014), Moghavvemi et al. (2017), and Phang et al. (2018), also showed similar results. Venkatesh et al. (2012) argued that younger males tend to demonstrate a greater tendency to seek novelty and innovation during the early stages of using new technologies that enhance the relative importance of Hedonic motivation. Thus the H7 of this study are: H7a : Hedonic motivation (HM) effect behavior intention (BI) H7b : Hedonic motivation (HM) affect behavior intention (BI) and it is moderated by gender, age, and income.

The Effect of Price value (PV) on behavior intention (BI)
A positive price value occurs when the benefit of using technology is greater than the cost incurred, and the value of the price positively impacts the behavior intention (Venkatesh et al., 2012). In the research by Nuriska et al. (2018), Indrawati and Haryoto (2015) proved that the price value is important to the interest in using technology. The re-search of Mahendra and Ardani (2015) showed that demographic factors consisting of age, education, and income have a positive and significant impact on buying consumers in the Body Shop cosmetics products in Denpasar. According to Venkatesh et al. (2012), female respondents were stronger in influence than men at the price value related to the behavioral intention. The influence of age proved to be moderation is found that the older is stronger in influencing facilitating conditions, price value, and habit of behavior intention (Venkatesh et al., 2012). Thus the H8 of this study are H8a : Price value (PV) has a significant and positive effect on behavior intention (BI). H8b : Price value (PV) affects behavior intention (BI) moderated by gender, age, and income.

Conceptual Framework of The Research
This study adopted seven constructs in the UTAUT2 model (performance expectancy, effort expectancy, social influence, facilitating condition, hedonic motivation, price value dan habit) and added perceived playfulness. Perceived playfulness added to the model was adapted from the research conducted by Moghavvemi et al. (2017). Haikal Bekti as chairperson of Traffic Management Lazada, there are two main reasons for online shopping trends to develop: promotion and pleasure (Sopia, 2018). Sulaiman (2016) argues that online shopping behavior is a fun activity because it can release emotions. The UTAUT2 model has three moderator variables which are age, gender, and experience. However, the researcher removed the experience variable because it required periodical measurements by this research. Besides, the researchers added income variables adapted from Indrawati and Haryoto (2015). This research developed the framework of the research concept in Figure 1 below.

METHOD
In this research, researchers used quantitative research type. The type of research used was explanatory. The population in this research is infinite. The population is all e-commerce users located in East Java. It is assumed that e-commerce users need information technology in the transaction pro- cess of Sensatia botanicals. The researchers took 263 samples using a questionnaire in 3 cities in East Java: Surabaya, Malang, and Kediri because these three cities have the largest population in East Java. The sample selection in this research used a purposive sampling technique, considering a minimum age of 15 years, having an e-commerce account, and knowing the Sensatia Botanicals products. Data analysis techniques in this research used SmartPLS 3.0. This research divided the income level into two categories which were  IDR4,300,000, and > IDR4,300,000, divided the age into two categories which were  24 years and >25 years and divided gender into two categories: man or female.

Characteristic of respondent
From 263 respondents, percentage women dominated the research of 192 (73%), the age category of 20-24 years dominated research of 159 (60.5%), the type of work was dominated by students of 96 (36.5%), the income of <IDR 1,700,000 dominated research of 94 (35.7%), and most respondents had been using e-commerce for more than two years of 152 (57.8%).

Evaluation of Measurement Model (Outer Model)
From the whole test, all item had a loading factor value > 0.6, composite output reliability value > 0.7, and Conbrach's Alpha> 0.6, AVE value > 0.5.
Thus, it can be concluded that all items of this research are valid and reliable. In contrast, the discriminant validity evaluation was conducted by looking at the cross-loading value and showing that it had good discriminant validity.

Evaluation of Structural Model (Inner Model)
This research showed that Performance expectancy, effort expectancy, social influence, facilitating conditions, habits, perceived playfulness, hedonic motivation, the price value of 59.9%. In comparison, the remaining 40.1% were contributions of other variables which were not discussed in this research.

Hypothesis Test
Based on Ghozali and Latan (2015), the test criteria is if the t-statistic  t-table (1.96) or the probability value of the level of significance ( = 5%). Furthermore, it can be stated that there is a significant effect of the independent variable toward the dependent variable.

Hypothesis Test Moderation Effect
The moderation effect is generally used to show an interaction between exogenous variables (predictors) and moderator variables affecting endogenous variables (Ghozali and Latan, 2015). Because the construct used in this research was in the form of a category, the moderation effect test was carried out using a multigroup analysis.

DISCUSSION The Effect of Performance Expectancy on Behavior Intention
The result of data analysis and hypotheses test showed that performance expectancy consisting of 4 (four) indicators, perceived usefulness, extrinsic motivation, job fit, relative advantage, positively affected behavior intention to use e-commerce to purchase Sensatia Botanicals. This is in line with the research by Venkatesh et al. (2003), Ghalandri (2012), Borrero et al. (2014), Escobar-Rodríguez and Carvajal-Truzillo (2014), and Osakwe et al. (2017). These results indicate that users find e-commerce applications useful in everyday life, increasing opportunities to achieve goals and helping meet needs faster. Thus, it fosters intention in using the application to shop for Sensatia Botanicals.

The Effect of Effort Expectancy on Behavior Intention
The result of the research showed that effort expectancy didn't affect behavior intention. The results of this research was not in line with the research conducted by Ghalandri (2012), Escobar-Rodríguez and Carvajal-Truzillo (2014), Osakwe et al. (2017). However, this research was in line with research carried out by Sedana and Wijaya (2010), Phang et al. (2018), and Shaw and Sergueeva (2019). It states that that effort expectancy did not affect behavioral intention. Based on interviews with several respondents, it was known that they considered that the use of e-commerce was easy, but beauty products will be more liked when buying offline because they can try a tester. Besides, the ease of use of e-commerce could be felt when the application runs smoothly and doesn't often have problems (trouble). In this case, some applications went down when many people had access.

The Effect of Social Influence on Behavior Intention
The analysis result showed that the social influence variable had a positive effect on behavior intention. It showed that the greater the effect of an individual's social environment, the greater intention in using e-commerce when people wanted to find a product or buy the product they wanted. This research was in line with the research by Escobar-Rodríguez and Carvajal-Truzillo (2014) and Phang et al. (2018).

The Effect of Facilitating Conditions on Behavior Intention
The result of this research showed that facilitating conditions affect behavior intention. The results of the research were in line with the research by Ghalandri (2012), Oliveira et al. (2014), Escobar-Rodríguez and Carvajal-Truzillo (2014), and Shang et al. (2017. Users felt that facilitating conditions such as mobile phones/laptops, shopping applications, sufficient knowledge, and assistance from others. At the same time, housing e-commerce encourages their intention in using e-commerce to shop for Sensatia Botanicals.

The Effect of Habit on Behavior Intention
The result of the research revealed that habit was a predictor of behavior intention using e-commerce and had a positive effect. That was supported by research conducted by Venkatesh et al. (2012), Escobar-Rodriguez andCarvajal-Truzillo (2014), Nuriska et al. (2018), and Yusuf and Indrawati (2019). The public had become accustomed to using e-commerce as a shopping medium.

The Effect of Perceived Playfulness on Behavior Intention
The result of the research showed that perceived playfulness affects behavior intention. This research was in line with Lu et al. (2016) and Shang et al. (2017). It can be concluded shows that users feel pleasure while using e-commerce.

The Effect of Hedonic Motivation on Behavior Intention
In this research, hedonic motivation using ecommerce didn't affect significantly on behavior intention. The results of this research contradicted the research of Escobar-Rodriguez and Carvajal-Truzillo (2014), Moghavvemi et al. (2017), Phang et al. (2018). however, it was in line with the research by Nuriska et al. (2018) and Yusuf and Indrawati (2019). In this case, it can be concluded that people intend to use e-commerce to buy Sensatia Botanicals based on needs (utilitarian value).

The Effect of Price Value on Behavior Intention
Price value consisting of 3 (three) indicators according to Venkatesh et al. (2012), namely reasoned price, good value of money, and good value at the current price positively affected the intention in using e-commerce behaviors by people in East Java. This research was supported by Nuriska et al. (2018) and Indrawati and Haryoto (2015). Respondents felt that the cost of using e-commerce was balanced with the ease received while conducting transactions.

Gender, Age and Income as Moderation Variable
In this research, gender didn't significantly strengthen the relationship between performance expectancy, social influence, facilitating condition, habits, perceived playfulness with behavioral intention. Still, gender affects the relationship between price value on behavior intention. This research's results were following previous research by Tavares et al. (2018) and Nuriska et al. (2018). Almost all people in Indonesia have been connected to the internet in this digital era and 86% of internet users have conducted online shopping. Age moderation testing using the PLS-MGA method showed that age groups of 24 years and> 25 years didn't have different intentions in using e-commerce to purchase Sensatia Botanicals. These results were in line with research by Nuriska et al. (2018). The results of this result confirmed that there was no effect of age in this research. The characteristics of the millennial generation and generation Z are closely related to technology, and almost all information about e-commerce can be accessed through digital media. The PLS-MGA method's income moderation test showed groups with income d" IDR.400,000 and groups with income> IDR 4,300,000. There is no difference in e-commerce behavioral intention to purchase Sensatia Botanicals. This result was in line with the research by Nuriska et al. (2018) and Yusuf and Indrawati (2019). This research confirmed that there was no effect toward income because basically e-commerce could be accessed free of charge by all consumers through smartphones connected to the internet network. Besides, all consumers from all classes of life with diverse incomes can buy Sensatia Botanicals products

CONCLUSIONS
The result of the analysis indicates that performance expectancy, social influence, facilitating conditions, habit, perceived playfulness, and price value significantly influence behavior intention to use ecommerce and that effort expectancy and hedonic motivation do not have any significant positive influence.
Gender differences can increase the intention of individuals to use e-commerce to purchase Sensatia Botanicals on the price value factor, while age and income differences in this study neither increase nor decrease the intention of individuals to use e-commerce to purchase Sensatia Botanicals.
Based on the results in this study, the model obtains an R-Square value on the behavior intention variable of 59.9%.

LIMITATIONS
The limitation of this study is using all types of e-commerce and use one brand Sensatia Botanicals. This research could not conclude the behavior intention to use e-commerce for green cosmetics products in general.

RECOMMENDATIONS
Further research is expected to examine the intention of using other platforms used to do online shopping, such as Instagram, social media, or Facebook. Thus, it obtains the results of research from different sides of this research. Further research should use populations from other types of businesses other than green cosmetic Sensatia Botanicals because the results obtained are not necessarily the same as this research. Further research can be carried out by using the complete UTAUT2 concept and the moderation effect following the initial UTAUT2 concept by adding other variables to the development of models that have not been used in this research. Such as variables of utilitarian motivation, location, trust, perceived risk, system quality. That is expected to prove further the effect on customer intention in the use of e-commerce. The Company of Sensatia Botanicals can implement marketing strategies such as free shipping or shopping vouchers with the minimum purchase, increasing consumers' intention in buying Sensatia Botanicals products through e-commerce.