The Determinant Factors of E-Commerce Purchase Decision in Jakarta and Tangerang

This research aimed to identify the factors that influenced the purchase decision in e-commerce. These used factors were brand image, price, and trust which were adopted from the previous research. The research was designed using a quantitative method with accidental sampling. From 200 questionnaires, there were 117 returned questionnaires from respondents who had made online purchases around Jakarta and Tangerang. The data was analyzed using SPSS. The result shows that brand image has positive influences on purchase decision. On the other hand, price and trust do not influence it. Price is the secondary commitment to make a purchase decision. Thus, a lower price does not guarantee the convenience of online shopping. Meanwhile, trust in making online decisions are built from brand image, shop, and strong products in the market place. Those rejected variables are affected by research fields and target samples.


INTRODUCTION
The digital age is commonly hailed as the fourth industrial revolution. This revolution has potentially transformed every daily life behavior from reshaping how people make a decision and enhancing consumers' experiences to creating a new model business to optimize value chain for the unprecedented and unpredicted level of efficiency (Lee et al., 2018). They perceived that the current revolution was the development of wide and rapid internet technology in human activities. With the development of the Internet, firms are increasingly reaching out to the consumers through various channels such as electronic commerce (e-commerce). Online shopping has become increasingly important to consumers. This phenomenon creates a new thing of consumer behavior in online shopping that people can easily buy and sells their product using e-commerce facility, social media, and community (Vasić, Kilibarda, & Kaurin, 2019).
Indonesia is stated as one of the countries with big e-commerce developments in Asia. Several factors support the rapid growth of e-commerce sector in Indonesia. Firstly, smartphone and Internet development are continuously rising. Secondly, Indonesia has a big population with the strengthening purchasing power amid the robust macro-economic growth. Thirdly, Indonesia has a young and tech-savvy population (Asosiasi Penyelenggara Jasa Internet Indonesia, 2017). With 132,7 million active Internet users in 2016, Indonesia can adjust to new technology easily and rapidly (Statista, 2018). Besides that, the support from the government by opening the large scale of investment from domestic or overseas to participate in developing Micro, Small, and Medium Enterprise and e-commerce has made e-commerce growth very fast. This growth gives chances to e-commerce practitioners to grow their business such as Tokopedia, Bukalapak, Elevenia, and others (Harsono, 2016).
However, growing digital business in the developing country does not always run smoothly. Generating revenue through an online store is one of the key issues faced by e-commerce entrepreneurs. Some of them are even out of business like Rakuten or Inapay (Rochmi, 2018). In 2016, a survey conducted by the Australian Government gave information about the penetration of Indonesian e-commerce competing with China in regional characteristic. In the recording data on year growth, the Indonesian e-commerce market was still relatively small compared to China with sales US$692 billion. In sales percentage of total sales, it was amounted to 1,6% compared to 13% in China (Rastogi, 2018). These percentages show that the consumer has a unique way to perform purchase decision and indicates many aspects that are not explored for business to use e-commerce as a mechanism for entering Indonesian markets. According to Das, Gryseels, Sudhir, and Tan (2016), the unique habits of Indonesian consumers are influenced by different geographical and cultural locations. These factors should drive the consumer to be more diverse and dynamic to make their purchase decision.
Purchase decision is a process that encourages consumers to identify needs, generate options, and choose specific products and brands. It is seen as a series of options before making the purchase. Purchase decision has many aspects including where to buy the product, which brand or model, when to make a purchase, how much money to spend, and which payment method. This decision is significantly influenced by product packaging (Salem, 2018). Many factors influence consumers' purchase decision factors. The consumers' condition in decision-making includes external, internal, and marketing factors. Culture, subculture, group, situation, social class, and family are in external factor. For internal factor, there are perception, attitude, knowledge, personality, lifestyle, involvement and role. In marketing factors, they are the product, aesthetic package, promotion, distribution, service, and price (Mutsikiwa & Marumbwa, 2013). According to Karimi, Papamichail, and Holland (2015), the marketing department of marketers investigates consumers' purchasing patterns to know and identify where they buy, what they buy, and why they buy. However, it is not easy to know why consumers buy a particular product.
According to these phenomena, understanding the factors influencing purchase decision is vital for e-commerce practitioners. Rizan, Nauli, and Mukhtar (2017) expressed that the price in Taiwan online decision making dispersed the confusion felt by consumers when they traditionally contacted the suppliers. It appears as the final determinant in online decision making. Other factors also stated by Kusumah (2015), the research involved social media in purchase decision with the demand for purchase decision based on the purchasing power. Thus, the price had an absolute factor because it was the first thing that will be considered. The result also informed that brand image had a positive influence on social media decision making like the second opinion. Bonn, Cronin, and Cho (2016) stated that trust was not only creating consumer and supplier relationship but also the impression after purchasing in the United Kingdom.
Foster (2018) defined brand image as a description of the association and consumers' confidence in a particular brand. Therefore, the brand image of a good product will encourage prospective buyers to buy the product instead of buying the same product from another brand (Rares & Jorie, 2015). According to Zhang (2015), brand image is the consumers' understanding derived from the total set of brand-related activities engaged by the firm.
Moreover, Permana and Ilham (2018) agreed that brand image was the reasoned or emotional perceptions that consumers attached to specific brands. Chakraborty and Bhat (2018) had investigated that brand image was the consumers' perception of a brand as reflected by the quality of products in consumers' memory. The consumer may have several associations about a brand. In the context of products, associations refer to the brand image encoded in the consumer's mental framework. This viewpoint shows that consumers hold the fact based on opinions about the store and feel confident about those opinions (Alamsyah, Trijumansyah, & Hariyanto 2017). Lin and Chen (2006) conducted 369 effective questionnaires in Taipei. It implied that country-oforigin defined as brand image had a positive influence on purchase decision. Therefore, the first hypothesis is: H 1 : Brand image will influence purchase decision positively.
Purchase decision is responsible for price expectation analysis as long as the expected price increases and the purchase decision may also increase (Ozmen, Öner, Khosrowshahi, & Underwood, 2014). Price is the amount of money needed to get some combinations of goods and services (Rares & Jorie, 2015). From the consumer's perspective, price represents the number of money consumers must give up to get the product or service (Lien, Wen, Huang, & Wu, 2015). According to Limpo, Rahim, and Hamzah (2018), price is a factor determined by supply and demand interaction in production. In the online shopping, consumers tend to compare the objective prices (price offered by the current vendor) with reference prices (price offered by other vendors) and form their perceptions of price (Kim, Xu, & Gupta, 2012). From the definition mentioned, it can be seen that the price paid by the consumers includes the services provided by the sellers (Rares & Jorie, 2015). Many companies make approaches to price based on the objectives to be achieved (Nofri & Hafifah, 2018).
It can be concluded that the price is very important in every marketing activity. The price becomes one of the tools needed to get purchase decision of a desired item or service (Rares & Jorie, 2015). Therefore, the second hypothesis is: H 2 : Price will influence purchase decision positively.
The concept of trust can be considered as the expectations held by the consumer that the store, people, and products are dependable and can be relied on to deliver their promises (Alamsyah et al., 2017;Campbell & Fairhurst, 2016). Trust concept is defined as a party willings to permit another party to conduct action. It is based on the expectation that other party will conduct certain important action to the trusted party (Greblikaitė, & Pervazaitė, 2014). Therefore, trust becomes the most important factor in the relationship of buyer and seller especially in e-commerce (Han & Hyun, 2015). It is also considered as one of the reasons for not purchasing online (Bauman & Bachmann, 2017). Thompson, Tuzovic, and Braun (2018) and Kim, Ferrin, and Rao (2008) also said that trust influenced to purchase decision, especially to a purchase decision. According to Ponte, Carvajal-Trujillo, and Escobar-Rodríguez (2015), consumers' trust leads to several benefits. First, the consumers gain satisfaction with seller, company, and transaction. Second, consumers have positive behavior of purchased goods, loyalty, and support to the company. Third, consumers' trust an influences to increase purchase intention. Fourth, consumers will choose to purchase offered commodities from a company they trust.
Trust is a very significant and important factor in the online environment because the consumer has few tangible and verifiable cues regarding the service provider's capabilities and intentions. Other online research and purchase decisions are almost solely by trust (Najafi, 2014). Therefore, the third hypothesis is: H 3 : Trust will influence purchase decision positively.
These variables are used as consumers' benchmarks in making a purchase decision. Furthermore, the e-commerce or market place can use these variables in increasing their value to lead competition and development. The research model for this research is in Figure 1. It shows that brand image, price, and trust affect purchase decision.
The researchers want to reexamine the variables that have been examined separately in the previous research. Moreover, the researchers combine the brand image, price, and trust in a purchase decision in Jakarta and Tangerang. It is to see whether it is still related or in accordance with previous research along with the development of e-commerce technology or the increasing number of new market places. This limitation initiates researchers to focus on limited understanding and to approach the factors influencing consumers' purchase decision in Indonesian e-commerce by adopting variables from previous research.

METHODS
This research is a quantitative research to test the research hypotheses. The sampling technique is accidental sampling. The survey is conducted using Google Form and distributed online by social media in August 2018 until September 2018. The researchers deliver the questionnaire to the online shopper in Jakarta and Tangerang by sending private message through Instagram, Whatsapp, and Line. For this survey, the researchers distribute the questionnaire to 200 people. Only 128 questionnaires return. However, 11 questionnaires are not complete, and 117 questionnaires are used for data analysis. The questionnaire has questions relating to research variables. Based on the existing literature, well-validated measurement items are adopted and included in the questionnaire. Multiitem and five points Likert-type scales from "Strongly disagree" to "Strongly agree" are utilized to measure study variables consistently.
The indicators for price are adopted from Setiawan, Herawati, and Asmarani (2015). Those indicators are in Table 1. Meanwhile, the indicators for trust are from Sfenrianto, Wijaya, and Wang (2018). It can be seen in Table 2. Moreover, Table 3 shows the indicators for brand image from Syahreza (2015). Lastly, in Table 4, the researchers adopt indicators for purchase decision from Hanaysha (2018), Jeong and Crompton (2018), and Marriott and Williams (2018). Moreover, the data calculation is linear regression and performs reliability and validity tests using SPSS. Prices and the quality of products offered are appropriate 3 Product prices are in accordance with perceived benefits 4 Discounts affect you in buying products The seller is reliable because she/he cares about my interests 4 I believe the seller is responsible for every service provided Table 3 Brand Image Indicators   Number  Indicator  1 The seller has a good reputation 2 The seller does the service well 3 The seller provides convenience and safety in shopping 4 The seller provides complete and clear information The purchase decision that I take is based on the good brand image 2 Purchase decisions are taken based on price considerations 3 Purchase decisions are based on trust
The validity test calculates the validity of the indicators. The indicator is valid when it can describe what is measured inside it. In this research, the researchers use a factor analysis of 50% or 0,5. The indicator will be acceptable if the significant value is more than or equal to 0,5. Meanwhile, it will be rejected if it is below 0,5 (Ghozali, 2016). There are three independent variables namely brand image, price, and trust that will be tested. Then, purchase decision is the dependent variable. Each variable has four indicators except purchase decision. The results can be seen in Tables 6-9.     All the results from the validity test are valid except an indicator in Table 7. The fourth indicator (Price_4 = 0,443) is invalid. The researchers assume there is no correlation between indicator compiled with the statement made, so the respondents' answers do not reflect what is measured. It is also because the question is adopted from Setiawan et al. (2015) having a different type of research field. Thus, it is possible to get different validity result. Then, the rest of the indicators get value more than 0,5, and it can be concluded as valid and unambiguous to describe the variables.
Next, the reliability test is performed when the variables are proven as valid with validity test. This test calculates the correlation between indicators and variables (Ghozali, 2016). As a result, the reliabilities of the variables that comprise an interaction should be general high to ensure the suitable indicator measurement with respondents' viewpoint. The indicator can be concluded as reliable if it does not create an unambiguous meaning. It can be seen from the consistency and stability of respondents' answers in a certain time. The researchers use Cronbach Alpha > 0,60 as the scale to measure the reliability of variables. It is adapted from Ghozali (2016). The results are in Table 10. The results of three variables show that the Cronbach's alpha is more than 0,6. It indicates a high level of internal consistency for the scale with the specific sample. Moreover, the hypotheses are tested using a linear regression method analysis with SPSS tools. In the hypotheses, the researchers investigate the influence of brand image, price, and trust on purchase decision. The model summary is in Table 11.  According to Sujarweni (2015), from F-value = 6,078 and F-table = 3,92, it is F-value > F-table as stated in Table 12. It describes the influences of all variables in making purchase decision simultaneously. It also implies that even though the variables model only describes 13,9% of the variances, those variables have a significant effect on a purchase decision. The results can be seen in Table 13. According to Sujarweni (2015), if it is p < 0,05 and t-value > t-table, the hypothesis is accepted. The results are in Table 12. The t-table is 1,65833. H 1 , which is brand image to purchase decision with t = 2,216 and p=0,029, has a significant and positive effect of the purchase decision. Thus, H 1 is accepted.
H 2 regarding price to purchase decision with t = 1,619 and p = 0,108 has a positive effect on the purchase decision. However, it does not have a significant effect of purchase decision at 0,05 level. Thus, it can be concluded that H 2 is rejected.
In H 3 (trust to purchase decision), trust has t = 0,553 and p = 0,58. The result implies that trust has a positive effect on the purchase decision. Similarly, trust also has no significant effect on the purchase decision at 0,05 level. Hence, H 3 is rejected. The detail is shown in Figure 2.

Figure 2 Hypotheses Results
This research is an empirical validation of the existing marketing theory which consumers make purchase decisions in e-commerce. With the advancement of e-commerce, there are many choices of platforms in shopping because e-commerce provides an easy aspect of shopping transactions. Moreover, consumers can choose a product and compare it, and consider several factors from many stores in e-commerce. This is done to attract consumers to buy the goods online.
The researchers recognize that making a purchase decision is not easy unlike shopping directly at the store. In the offline stores, consumers can see the items they want to buy directly. However, in online stores, consumers only see images and product specifications displayed on a market place. From the results of this research, it is found that brand image has a significant and positive effect of the purchase decision. One of the examples is Tokopedia. It builds its brand image for several years of being a preferred market place for consumers to shop. This result is in line with Widyastuti and Said (2017) about the significant and positive impact of brand Image on purchasing decisions. Moreover, Kotler and Keller (2016) stated that creating a brand was to add value to know consumers. It was based on what they thought and felt about the brand.
Next, it is regarding the effect of price on purchase decisions. From the results, the price has a positive effect in influencing the purchase decision. However, it does not have a significant influence on the purchase decision. The consumers feel that the offered price cannot provide comfort for consumers. Several things make price insignificant in the purchase decision. For example, if the price is too cheap in a market place or store, there are concerns about the quality of the goods. In the price factor, it turns out they have a price limit for the purchase decision. This is also in line with Sudjatmika (2017) who agreed that there was a positive impact of price on purchase decisions. However, it did not have a significant effect on the effect on purchase decisions.
Moreover, Sukri and Arisandi (2017) also agreed that price had a positive influence to attract consumers to purchase goods using social media. According to Kotler and Keller (2016), the purchase decision is based on how consumers perceive price and what they consider the current actual price. Consumers also have a lower line price limit and a line on the highest price. It is because price has a signal to or not accept quality, and consumers have different ways to interpret price.
The researchers imply that the average respondents who fill out the questionnaire are 54,6% males with 62,11% between 22 to 30 years in Jakarta. Meanwhile, the previous research does not focus on the e-commerce market and conducts it in a different location. Those factors create different findings.
Moreover, the results of trust have a positive effect on a purchase decision, but it does not have a significant influence on a purchase decision. For example, every consumer has different levels of valuation of trust because trust is built and shaped by the strength of a brand that is known by consumers. The trust of consumers can be seen in consumers' reviews. For example, currently, there are many market places in the Internet world. However, when people enter the market place, they have few reviews from consumers. The confidence in the market place will not automatically be built.
From the results, the researchers notice that most of the respondents' income is above Rp5.000.000,00 per month in 2018 about 34,8%. Meanwhile, respondents' income in the previous research around US$200 to US$300 in 2016 is 23%. Moreover, the current respondents' majority considers buying products online for more than one in a month. This result is similar to Kusumah (2015) about the positive but insignificant impact of trust on purchase decision. According to Kharouf, Lund, and Sekhon (2014), trust is an important factor in online purchase that a market place can provide certainty and reduce consumer risk associated with the purchase decision.

CONCLUSIONS
This research is a collection of answers from respondents to analyze purchase decisions. The researchers find that the brand image has a more significant influence than price and trust. Price is the secondary commitment level to decide to buy. Thus, the lower price does not guarantee the convenience of online shopping. Besides that, trust in making an online decision is not only about the provided information, seller services, same interests, or discount. Trust is about acknowledgment from others and is indirectly built from the market place brand image, shop, and strong products.
From this result, the researchers expect it can be used as the basis for e-commerce to compete in socializing and creating a brand that can be recognized by the public and win the consumers to become the favorite market in shopping online. Moreover, this research has a limitation because the result only covers around Jakarta and Tangerang. Besides that, the variables (brand image, trust, and price) are limited. Therefore, future research can explore the scope by collecting the data at different population and areas in the other country. Then, various fields can be added by collecting more broad data on population and region from other variables such as purchase decisions based on needs or hobby and website security for online transaction in e-commerce.