Journal of Accounting, Management, and Economics Vol. 19, No. 2, 2017, pp. 22-34 Published by Faculty of Economics and Business Universitas Jenderal Soedirman Published online on July 11, 2017 in http://jos.unsoed.ac.id/index.php/jame  Correspondence to : 1YARSI University, Indonesia. E-mail: muslikh@yarsi.ac.id 2YARSI University, Indonesia. E-mail: zainal@yarsi.ac.id 3YARSI University, Indonesia. E-mail: rini.hidayati@yarsi.ac.id Received: May 15, 2017 Revised: May 29, 2017 Accepted: June 12, 2017 INTRODUCTION Internet is a media that can be used to conduct business transactions including making online purchases. The benefits of this online purchase include the convenience of operating for 24 hours and consumers do not need to visit the physical store. But with these advantages, some consumers are still reluctant to make online purchases. The main advantage of buying or shopping online is because of its convenience aspect. Online sellers come to buyers, not just buyers that come to online sellers. Customers can shop by online for 24 hours a day or 365 days per year. Customers do not need to visit physical store by vehicle, no need to park vehicle, no need to worry about traffic jams in the street, because they will not happen if customers buy online. However, online purchases are also at risk of not meeting buyers with sellers directly. Egger (2006), the number of consumers who buy online has grown but in sluggish condition so that trust is urgently needed in order to increase the growth of this online purchase. Ahuja, et al. (2003) in his research on students showed that the issue of secrecy and security is the number one cause of the reluctance of individuals to buy online. Some companies even use certain technologies called cookies to tap the activities that users do when shopping online. There are consumers who argue that shopping traditionally is better than shopping online, because of the factor of trust, satisfaction, and secrecy and security. Therefore, it should be examined how the actual behavior of consumers on online purchase. Data from Media Index Wave-2, 2015, internet users in Indonesia have reached 80 million people. But of the approximately 80 million people, online buying/shopping is not included in the categories of activities done by internet users in Indonesia. Research empirically proved that trust affects the individual’s attitude to purchase online. This study was conducted by Lim, et al. (2006) which stated the effect of trust on attitude to purchase online is significant. These studies used a construct of trust on attitude to purchase online with the same question indicator referring to the study of Jarvenpaa, et al. (2000). The results of research conducted by researcher in 2010 on the behavior of students in Jakarta, products purchased online by students are usually related to fashion products such as clothing and shoes and students rarely buy textbooks online. The latest data in 2015, the number of internet users in Indonesia has reached 88.1 million people. Products purchased by consumers in Indonesia most often are plane tickets (40 percent), followed by books (37 percent), clothing, accessories, and shoes and The Influence of Satisfaction and Trust on Online Purchase Decision (Empirical Study on Students in Jakarta) MUSLIKH1, ZAINAL2, RINI HIDAYATI3 1)2)3)Economics Faculty, YARSI University, Indonesia Abstract The purpose of this study is to determine the effect of consumer satisfaction and trust on buying attitude and its impact on online purchase decision. The study was conducted through survey by distributing questionnaires to 130 students in Jakarta from various faculties. Research sampling used non probability sampling with purposive sampling technique. The results showed that: consumer satisfaction and trust have positive and significant effect on online purchase attitude. Consumer satisfaction and trust have positive and significant effect on online purchase decision, purchase attitude has significant effect on online purchase decision. Keywords Consumer Satisfaction; Trust; Purchase Attitude; Purchase Decision; Online Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 23 electronic goods (21 percent). However, for consumers in Asia Pacific, books are the most frequently purchased item (46 percent), followed by clothing, accessories, and shoes (38 percent), cosmetics and supplements (27 percent), and wholesale goods (26 percent). The development of online purchase actually shows a significant increase in the world, especially in some developed countries. But in reality, this is not the case in Indonesia where the level of online purchase progress is still in a slow stage to grow. Factors that influence it are various starts from characteristic of Indonesian society itself until individual problem such as lack of trust, satisfaction, or security problem. Therefore, it is necessary to do in-depth research including other factors that inhibit the growth of online purchase in Indonesia. Based on the background of the problem, phenomenon gap above, it is necessary to examine how consumer behavior is mainly related to the product online purchase. Research Question 1. How the general picture: customer satisfaction, trust, purchase attitude, and online purchasing decision. 2. How the influence of consumer satisfaction on online purchase attitude. 3. How trust affects online purchase attitude. 4. How the influence of consumer satisfaction on online purchase decision. 5. How trust affects online purchase decision. 6. How purchase attitude affects online purchase decision. Research Objective 1. To find out how the general picture: customer satisfaction, trust, online purchase attitude and purchase decision. 2. To find out how the influence of consumer satisfaction on online purchase attitude. 3. To find out how the influence of trust on online purchase attitude. 4. To find out how the influence of consumer satisfaction on online purchase decision. 5. To find out how the influence of trust on online purchase decision. 6. To find out how the influence of online purchase attitude on online purchase decision. LITERATURE REVIEW Consumer Behavior Consumer behavior is a process that a consumer passes through in seeking, buying, using, evaluating, and acting post- consumption of products, services, and ideas that are expected to meet the needs and desires of consumers (Prasejito, 2005) .The study of consumer behavior is how the decision makers, whether individual, group, or organization in making purchase decision of a product to consume. While, according to Mangkunegara (2004), consumer behavior is action conducted by individual, group, or organization that are related to decision process in getting or using goods that can be influenced by environment. With the development of information technology, will have an impact on consumer behavior in purchase decision. Consumer behavior model in the online era is influenced by internal and external factors as follows: Consumer behavior changes after the technology that we know as e-commerce. The concept of e-commerce is derived from e-Business. Definition of e-commerce is very diverse, depending on the perspective or glasses that use it. Association for Electronic- commerce simply defines e-commerce as “electronically business mechanism”. Commerce Net, an industry consortium, provides a more complete definition of “the use of computer networks (interconnected computers) as a media of business relation creation”. The model of consumer behavior in the digital era is as follows: Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 24 Figure 1. Consumer Behavior Model Based on the above characteristics, it is clear that basically e-commerce is the impact of the development of information and telecommunication technology, thus significantly changes the way people interact with their environment, which in this case is related to the business mechanism. The growing business community that uses the internet in conducting its daily activities indirectly has created a new world domain which is often termed as “cyberspace”. Unlike the real world, cyberspace has unique characteristic whereby a human can easily interact with anyone in this world as far as it is connected to the internet. The loss of the world’s boundaries that enable one to communicate with others efficiently and effectively directly changes the way companies do business with other companies or consumers. That in principle, e-commerce provides the infrastructure for companies to expand internal business processes to the external environment without facing the time and space barriers that have been the main issues. Opportunities to build networks with other institutions must be utilized because today’s real competition lies in how a company can utilize e-commerce to improve performance in its core business. Based on the core business, each company has its own core process sequence, where at various sub-process points there is an interaction between the company and one of the relationships above. If in the past most relationships can only be established “one-to-one relationship” for efficiency reasons, with e-commerce, corporate relationships with other external entities can be made “many-to-many relationship” faster, better, and cheaper. Consumer Satisfaction In the era of globalization, business competition becomes very sharp, both in the domestic (national) market and in international markets. To win the competition, a company should provide satisfaction to customers, for example by providing quality products, marketing systems that follow the development of the times such as online systems, competitive price, faster product distribution or delivery, and better service than its competitors. Products with poor quality, high price, or slow product delivery can make customers dissatisfied, although with different levels. Customers are to be satisfied, because if they are not satisfied they will leave the company and become competitor customers. This will lead to a decrease in sales and in turn will lower profits and even get losses. Therefore, the marketing division measures the level of customer satisfaction to immediately know what attributes of a product that can make customers not satisfied. That customer satisfaction is as a result of customer research on what is expected by buying and consuming a product. That expectation is then compared to the perception of the performance received by consuming the product. If expectation is higher than product performance, consumer will feel dissatisfied. According to the above definition, there are two measures covered. The first measure is customer expectation that serves as a benchmark over the - Need and motivation - Personality - Psychographic - Perception - Learning - Attitude - Need - Choice of product attribution - Attitude - Perception - Lifestyle - Attitude Internal Consideration External Consideration Consumer Seeking Information Purchase Decision Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 25 second measure that is product criteria. The other definition uses only one measure of customer satisfaction after buying and using a product. In this case, product performance score serves as a customer satisfaction score. Kotler (2012) defined “satisfaction” as a person’s happy feeling that comes after comparing perception or impression with the performance (or outcome) of a product and its expectations”. The above definition, satisfaction is a function of perception or impression of performance or expectation. If performance is below the expectation, customers will be dissatisfied and if performance exceeds expectation customers will be very satisfied and happy. Many companies focus on high satisfaction because customers who are satisfied easily change their minds when they get a better offer. Those who are very satisfied are more difficult to change their mind. High satisfaction or pleasure creates emotional attachment to a particular brand, not just rational likes/preferences. The result is if customer loyalty is high. For example, Canon’s senior management believes that for a very satisfied or highly satisfied customer for company is worth 10 times more than a satisfied customer. A very satisfied customer will probably stay with Canon for years and will buy more than a satisfied customer. Tjiptono (2009) stated that customer satisfaction is formed from the perception of the value of supply and expectation. Customers who are very satisfied or happy will have emotional ties and have a high loyalty to the brand. The creation of customer satisfaction can provide several benefits, including the relationship between the company and the customer to be harmonious, providing a good foundation for re-purchase. Customer expectations over time evolve, as more as information received by customers, as well as ever- increasing customer experience that will affect customer satisfaction level. Companies that want to excel in today’s marketplace have to look at customer expectation, company performance, and customer satisfaction. If the performance is below expectation, the customer will be disappointed, if the performance as expected then the customer will be satisfied, and if the performance exceeds expectation then the customer will be very satisfied or happy. Trust Online purchase behavior is based on mutual trust between buyers and sellers. Not infrequently, people who buy online are fooled by the qualifications of the goods offered. Confidentiality of consumer data is an important factor in trust. Consumers expect the data are not given to other parties who are not related. The five conditions of such system of secrecy and security include: Validity that is ensuring that parties involved in transactions through electronic media are true to those parties. Undeniable proof, including those parties involved in the transaction cannot refuse involvement after the transaction. Secrecy guarantee that is limitation of any form of communication that occurs on the parties who transact against other parties involved. Secrecy guarantee is very important in the scope of e-commerce due to the possibility of hackers can obtain personal data information that is sensitive. Secrecy protection that is protection of personal data information about consumers transact through electronic media. Personal data information cannot be provided without the consent of the customer. Another approach used for transactions in e-commerce is Secure Electronic Transaction (SET). The core mechanism of SET is the use of digital certification that enables preventing abuse caused by the prevailing system does not have the authority to own and provide data to credit cardholders and sellers as applicable in processing transactions with credit cards in general. Purchase Attitude Attitude is a positive and negative evaluation that is realized through behavior. Attitude is the implementation of individual behavior of an object. In relation to online purchase, customer attitude is strongly influenced by certain factors. There are four important factors in relation to online purchase attitude: product value, shopping experience, online service quality, and risk perception when buying online. Li and Zhang (2002) that the perceived risk related to buying products online include the problem of knowledge, cost, time, opportunity, and the product itself. While according to Bhatnagar, et al. (2002), besides cost risk, the risk of the product itself, another very important matter is the issue of secrecy and security. This is supported also by the research of George Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 26 (2002) that proved that there is indeed a relationship between consumer secrecy and attitude that leads to purchase online. It further said that attitude is also influenced by the level of trust on the online store concerned so that the perception of risk will be reduced. Koufaris, et al. (2002) found that the factors of control perceptions, pleasant perceptions, and the added value of online purchase are important dimensions of attitude related to online purchase activities. Furthermore, the experience of online purchase that includes evaluation, compatibility, and fun is an important predictor of online purchase attitude. While, Park and Kim (2003) viewed online purchase attitude can be improved if online stores have attributes that include: product characteristic, customer service, promotion, navigation, and ease and importance of security. Purchase Decision Purchase decision is determined by online- based satisfaction and trust. Several studies on purchase decision within the scope of online purchase have been made including the deciding predictor. Shopping or purchasing products online is generally a form of choice, where someone will do the activity or just avoid it. In practice, the process of online purchase is significantly different from offline purchase, even though the product purchased is the same. Generally, convenience is the most important reason a person chooses to make a purchase online. That online purchase is based on time availability. And the most important thing in making an online purchase is the search for certain product information through the internet. Bhatnagar, et al. (2000) stated that consumers turn to the internet with the intent to seek information about products that then make purchase of good and service. Consumers decide to shop online because they have less time. Furthermore, Bhatnagar, et al. (2000) argued that lack of security, technical understanding, and convenience level according to the consumer’s view are the main reasons why consumers choose not to purchase online. Bellman, et al. (1999) has finally concluded that people with high activity who certainly have a little more time tend to search for good or service online. Therefore, several months or several hours each week online consumers access the internet to find information related to the product or service. Related to decide on the purchase online, research results of Ho and Wu (1999) showed between online purchase with logistics support from the online store, product characteristic, website technology characteristic, information characteristic, and web appearance has a positive relationship. Especially for web appearance, it can also show credibility from online store or seller. Bellman, et al. (1999) has finally concluded that people with high activity who certainly have a little more time tend to search for good or service online. In this research, purchase decision is determined by trust. Purchase decision within the scope of online purchase has been made including the deciding predictor. Shopping or purchasing products online is generally a form of choice, where someone will do the activity or just avoid it. The process of shopping or buying online is significantly different from the shopping experience or buying offline even though the product purchased is the same product. Bellman et al. (1999) argue that comfort is the most important reason a person chooses to make an online purchase. It is said further that online purchases are dependent on consumers' willingness to make these online purchases and the availability of time. The most important predictor of making purchases online is searching for specific product information over the internet. Therefore, how many months once or how many hours per week online consumers access the Internet to find information related to the product or service. While related to predictors in deciding to make purchases online, research of Ho and Wu (1999) shows between buying online with logistics support from online stores, product characteristics, website technology characteristics, information characteristics and web appearance has a positive relationship. Theoretical Framework That consumer satisfaction can generate trust. In addition, customer satisfaction also raises consumer attitude toward online purchase. That trust is the most important element in online purchase decision. Trust that customers experience is the overall trust that involves all online purchasing activities. If consumers feel Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 27 satisfied and a sense of trust then they will determine the attitude to buy. Attitude will drive online purchase decision. RESEARCH METHODS Type and Design of Research This research includes survey research conducted by qualitative and quantitative approach. This research is also causal, because it aims to test the hypothesis about the causality relationship of one variable with other variables. This research model is expected to explain the causal relationship between variables in the research as well as generate conclusions that can be used to develop science, especially marketing. Therefore, this research is included in fundamental research. This research will be conducted in one stage, for one year. The research design is as follows: first mapping or descripting: customer satisfaction, trust, online purchase attitude, and purchase decision. Then analyze the variables that influence the online purchase decision. Population, Sample, and Sample Size Determination The study population set or targeted population is all undergraduate students in Jakarta. While, the inaccessible population is all undergraduate students in Central Jakarta. Based on the data obtained in Kopertis 3 Jakarta, there are 350,000 undergraduate students who study in Central Jakarta. In this case, the sample was set at 130 people. Sampling Technique Sampling method is by using purposive sampling method. Sampling method by purposive is sampling method where researcher has criteria or certain purpose to sample to be studied (Indriantoro, 1999). The criteria used in this study is the students have bought online. While, the reason researcher used purposive sampling method in this study was that researcher to really get the required information from the right object. The subject of this research was students in Jakarta. Research Variable and Measurement Variables in this research are: customer satisfaction, trust, purchase attitude, and online purchase decision. Measurements of research variable and indicator used in this study were as follows: (1) satisfaction is the perception of the fulfillment of fun based on transactions conducted. Satisfaction was measured by indicators: conformity with expected, online transaction experience, overall satisfaction. (2) Trust is trustworthiness in online purchase measured by indicators: well-known site, commitment / keeping promise, no problem, trust with information (3) purchase attitude is the implementation of individual behavior in online purchase measured by indicators: fun, cheaper, safer, not found elsewhere. (4) purchase decision is the implementation of consumer behavior in deciding the purchase, measured by: repurchase, make payment, would recommend to others. Data Source and Collection Method Data source in this research consists of primary data and secondary data. Primary data is data obtained directly from respondents through questionnaire answers and through observation. Primary data taken in this research is data about the responses of respondents related to the variables studied. Secondary data is data obtained from literatures related to the problem under study. Secondary data taken in this research is data about previous research related to this research and data about the object under study. Methods of data collection conducted in this study were as follows: 1. Questionnaire that is data collection through the provision of a list of questions filled by respondents. Data obtained through the questionnaire is data about the responses of respondents related to the variables studied. 2. Observation that is data collection through direct observation to the object of research. The data obtained through observation is data about field condition of the object under study. 3. Library study that is collecting data obtained by reading the literature and previous research related to the problem under study. Data Analysis Technique This research used descriptive and inferential statistical analysis techniques. Descriptive analysis to describe all data from research variables in the form of tables, whereas inferential analysis was used to test the hypothesis. Data analysis Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 28 tool used SEM (Structural Equaling Modeling). RESULT AND DISCUSSION Respondent Demography 1) Respondent Characteristic Based on the Faculty The number of questionnaires distributed was 150 copies. Of these, the questionnaires returned of 150 copies. Thus, the response rate is 100%. Nevertheless, there were 20 questionnaires that cannot be analyzed, so the total questionnaires analyzed of 130 copies. In processing the data, respondents were grouped by faculty. To find out more detailed characteristics of respondents based on faculty, can be seen in the following figure: Figure 2. Number of Respondents Based on Faculty Based on the above table, it can be seen that the most respondents were from the Economics Faculty of 25%, then respondents from the Law Faculty of 20%, and the Engineering Faculty of 17%. 2) Respondent Characteristic Based on Money Expended When Buying Products Online Respondent characteristic based on money expended when buying products online is grouped into four groups that are respondents who average spend money < Rp 0.5 million of 12%, spend between 0.5 million to 1 of 85%, and spend 1 million to 1.5 million of 3%. Based on the research results, can be seen that the amount of money that respondents spend when making online purchase is mostly between Rp 0.5 million to Rp 1 million per year. Figure 3. Average Money Spent per Year 3) Respondent Characteristic Based on Consideration when Making Online Purchase Respondent characteristic based on consideration when making online purchase is grouped into four that are based on product/brand quality consideration, based on price consideration, based on security consideration, and based on friend recommendation consideration. Based on the research results, can be seen that most respondents made online purchase on the basis of price consideration with the number of 80 respondents or 62%. The least is the consideration based on friend recommendation of 4%. To find out more detailed characteristics of respondents based on respondents consideration when making an online purchase, can be seen in the following picture: Figure 4. Considered Fartors Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 29 4) Respondent Characteristic Based on the Length of Being Online Store Customer Respondent characteristic based on the length of being online store customer is divided into four groups: < 1 year, > 1-1 year, > 2-3 years, > 3-4years. Based on the research results, can be seen that the majority of respondents in being online store customers is ranged from 2-3 years. Figure 5. Length of Being Online Customer 5) Respondent Characteristic Based on Online Purchase Intensity or Frequency in One Month Based on purchase intensity or frequency, most respondents buy no fewer than once in a month. This is understandable considering the finance held by respondents (students) is still limited. Figure 6. Purchase Intensity/Frequency 6) Respondent Characteristic Based on Products that are Often Purchased Online In buying products online, respondents often consider what products will be purchased. Figure 7. Purchased Product Based on the graph above, products that are often purchased by respondents online are fashion products such as clothes, shoes, and others. 7) Respondent Characteristic Based on E-commerce Frequently Used for Online Purchase Now, a lot of e-commerce products that sell online. Each e-commerce offers specifically products sold such as hijab e-commerce, SMEs product-specific e- commerce, electronics e-commerce, fashion products e-commerce, and so on. The results showed that the most visited e-commerce in online purchase was Lazada as shown in the graph below. Figure 8. E-Commerce in Purchase 8) Respondent Characteristic Based on Interesting Promotion in Online Store Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 30 Figure 9. Interesting Promotion Media Based on the graph above, the most interesting promotion and has an effect in the online purchase is discount. 9) Respondent Characteristic Based on Payment Methods of Online Purchase Many ways are used to pay in online purchase. The results showed that the some respondents in paying online purchase use transfer through ATM of 69%, then followed by COD method of 15%. Figure 10. Paying Methods Descriptive Statistics of Research Data Data description used index numbers. This is done to find out the description of the respondent’s perception level toward the research variables. Based on the descriptive analysis, it shows that almost all variables have mean value of medium, meaning that the students in Jakarta in buying online show: satisfaction in purchase medium, trust in purchase medium, attitude in purchase medium, and speed in purchase decision is pretty good. Table 1. Average Descriptive of Respondents Answers Variable Average Result Description Satisfaction 2.67 Pretty good Trust 2.74 Pretty good Purchase Attitude 2.57 Pretty good Purchase Decision 2.71 Pretty good Analysis Research Result Data The result of confirmatory factor analysis shows that the critical ratio (CR) value of each indicator is greater than 2.0. Therefore, it can be stated that each indicator is proven to meet unidimensionality so that it can be followed by full model SEM analysis. Therefore, the full model can then be analyzed. The results of model testing by looking at the value of absolute fit value indicate that in general the model has goodness of fit that is good. The values of the test of the structural equation can be seen in the figure below. Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 31 Figure 11. Full Model Goodness-of-Fit Test Table 2. Summary of Goodness-of-Fit Test Result Based on goodness-of-fit test, the model is included in good category. Chi square value of 186.664 is smaller than chi square table with probability of 0.109. The determinant value (GFI) of 0.876 or 87.6% indicates the purchase decision (Y) can be explained by the variation of consumer satisfaction and consumer trust. The number of 13.4% is explained by other un- researched variables (citerus paribus assumption). Hypothesis Testing The significance of causal relationship in SEM analysis was tested through the null hypothesis that shows that the coefficient of causal relationship between variables is equal to zero through the t-test commonly used in regression. Based on the output of Goodness-of-fit Index Cut-of-Value Analysis Result Model Evaluation X2Chi-Square Expected small (DF = 166) 186,664 Good Probability > 0.05 0,109 Good CMIN/DF < 3,00 1,138 Marginal GFI > 0,90 0,877 Marginal AGFI > 0,90 0,843 Good TLI > 0,95 0,964 Good CFI > 0,95 0,969 Good RMSEA < 0,08 0,033 Good Satisf action Trust Beha vior Deci sion Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 32 full model SEM analysis in the Appendix, then can be made summary of the causal relationship between variables in this study as listed in Table 3 below. Table 3. Standarized Regression Weight for Hypothesis Testing Estimate CR P Description Z <--- X1 .767 5,131 *** Significantly influence Z <--- X2 .229 2,238 .250 Significantly influence Y <--- X1 .421 2,578 .010 Significantly influence Y <--- X2 .262 2,914 .004 Significantly influence Y <--- Z .379 2,455 .014 Significantly influence Description: X1 : Satisfaction, X2 : Trust, Z : Purchase Attitude, Y : Purchase Decision First Hypothesis: The value of the critical ratio (CR) for the effect of satisfaction on trust is 4.242 greater than the value of ttable (1.960). So it can be concluded that the first hypothesis that express satisfaction has positive and significant effect on trust is accepted. Second Hypothesis: The value of the critical ratio (CR) for the effect of trust on purchase attitude is 4.777 greater than the value of ttable (1.960). So it can be concluded that the first hypothesis that express trust has positive and significant effect on purchase attitude is accepted. Third Hypothesis: The value of the critical ratio (CR) for the effect of trust on purchase decision is 3.058 greater than the value of ttable (1.960). So it can be concluded that the first hypothesis that express trust has positive and significant effect on purchase decision is accepted. Fourth Hypothesis: The value of the critical ratio (CR) for the effect of purchase attitude on purchase decision is 6.328 greater than the value of ttable (1.960). So it can be concluded that the first hypothesis that express purchase attitude has positive and significant effect on purchase decision is accepted. The Influence of Satisfaction on Trust The results showed that satisfaction has positive and significant effect on trust directly of 0.4981. With probability value of 0.000 (p is smaller than 0.05). The one direction mark on the coefficient shows a positive change which means that if the consumer is satisfied then it will increase the trust of online purchase. Online shop sellers strive to satisfy consumers by providing excellent service. This is shown so that consumers feel satisfied. The results of this study are in accordance with research of Lim, et al. (2006) which stated that satisfaction has positive and significant effect on trust. In addition, this study supports the research of Narver and Slater (1994) which stated that consumer satisfaction has significant effect on increasing the trust. The Influence of Satisfaction on Purchase Attitude The results showed that satisfaction has positive and significant effect on purchase attitude directly of 0.643. With probability value of 0.000 (p is smaller than 0.05). The one direction mark on the coefficient shows a positive change which means that if the consumer is satisfied then it will increase purchase attitude. This purchase attitude will have an impact on purchase intention. Online store sellers strive to increase consumer satisfaction by providing the best service. This is shown to make consumers feel satisfied. The results of this study are in accordance with the research of Lim, et al. (2006) which stated that consumer satisfaction has positive influence on the improvement of purchase attitude. In addition, this research supports the research of Narver and Slater (1994) which stated that consumer satisfaction has significant effect on the improvement of online purchase attitude. The Influence of Trust on Purchase Attitude Referring to the results showed that trust has positive and significant effect on purchase attitude directly of 0.310. With probability value of 0.002 (p is smaller than 0.05). The one direction mark on the coefficient shows a positive change which means that if the consumers have trust then it will increase purchase attitude. Online store sellers strive to increase consumer Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 33 trust by passing valid information. This is shown to sell goods sold in the market. The results of this study are in accordance with research of Lim, et al. (2006) which stated that trust has positive effect on the increase of online purchase attitude. In addition, this study supports the research of Cooper, et al. (2004) which stated that trust has significant effect on the increase of online purchase attitude. The Influence of Purchase Attitude on Purchase Decision The research results showed that purchase attitude has positive and significant effect on purchase decision directly of 0.904. With probability value of 0.000 (p is smaller than 0.05). The one direction mark on the coefficient shows a positive change which means that if the purchase attitude increases then it will improve the purchase decision. Online store sellers strive to improve consumer purchase attitude by paying attention to technological development, especially information technology. This is shown to increase purchase decision. The results of this study are in accordance with the research of Lim, et al. (2006) which stated that purchase attitude has positive effect on the increase of purchase decision. Direct, Indirect, and Total Influence To know the direct, indirect, and total influence of the variables, can be shown in Table 4 below. Table 3. Direct, Indirect, and Total Influence Standardized Direct Effect Standardized Indirect Effect Standardized Total Effect X1 X2 Z X1 X2 Z X1 X2 Y Z 0.65 0.23 0.00 0.00 0.00 0.00 0.65 0.23 0.00 Y 0.38 0.28 0.40 0.40 0.262 0.00 0.64 0.37 0.40 Description: X1 = Consumer Satisfaction, X2 = Trust, Z = Purchase Attitude, and Y = Online Purchase Decision Based on the analysis of direct, indirect, and total influence, can be explained as follows: The amount of the direct effect of consumer satisfaction (X1) on the purchase attitude (Z) is 0.65, and the direct effect of trust on purchase attitude (Z) is 0.38. While, the direct effect of satisfaction (X1) on purchase decision (Y) is 0.38. The direct effect of trust (X2) on purchase decision (Y) is 0.28. The amount of indirect effect of satisfaction (X1) on purchase decision (Y) through purchase attitude (Z) is 0.40. The amount of indirect effect of trust (X2) on purchase decision (Y) through purchase attitude (Z) is 0.26. The amount of total effect of satisfaction on purchase attitude is 0.65. While, the total effect of satisfaction on purchase decision is 0.64. The total effect of trust on purchase attitude is 0.23. While, the total effect of trust on purchase decision is 0.37. Based on the results of the mediation test, it can be concluded that there is relationship mediation between satisfaction and trust on purchase decision mediated by purchase attitude, because the value is changed to be significant. CONCLUSION Based on the results of analysis and discussion that has been described can be drawn some conclusions on the hypothesis as follows: The test results of five hypotheses, all are accepted. Consumer satisfaction and trust have positive and significant effect on purchase attitude. Consumer satisfaction and trust have positive and significant effect on online purchase decision. Purchase attitude has positive and significant effect on online purchase decision. Purchase attitude mediates the relationship between consumer satisfaction and trust on online purchase decision. MANAGERIAL IMPLICATION This study provides a managerial implementation for online store managers in the form of strategies to increase online sales. To increase online sales, online store managers need to understand consumer purchase behavior. To improve purchase attitude or behavior of consumers, can be done by increasing customer satisfaction and trust in the management of online stores such as website that attracts and provides valid information. Consumer Journal of Accounting Management and Economics , Vol. 19, No. 2, 2017, pp. 22-34 34 satisfaction can be improved through: quality product, fast transaction, and excellent service. While, trust can be improved through quality information on the website, online store seller commitment to keep promise of both goods quality and delivery time, and providing valid information. REFERENCES Ahuja, M., Gupta, B., & Raman, P. (2003). An empirical investigation of online consumer purchasing behavior. Communications of the ACM, 46(12), 145-151. Bellman, S., Lohse, G. L., & Johnson, E. J. (1999). Predictors of online buying behavior. Communications of the ACM, 42(12), 32-38. Bhatnagar, A., Misra, S., & Rao, H. R. (2000). Online risk, convenience, and internet shopping behavior. Communications of the ACM, 43(11), 98-105. Egger, A. (2006). Intangibility and perceived risk in online environments. Paper presented at the Academy of Marketing. University of Middlesex. George, J. F. (2002). Influences on the internet to make internet purchases. Internet Research, 12(2), 165-180. Ho, C. F., and Wu, W. H. (1999). Antecedents of customer satisfaction on the internet: an empirical study of online shopping. Proceedings of the 32nd Annual Hawaii International Conference on Systems Sciences. Maui. Jarvenpaa, S. L. and Todd, P. A. (1997). Consumer reactions to electronic shopping on the world wide web. International Journal of Electronic Commerce, 1(2), 59-88. Kotler, P. (2012), Marketing Management, (Mellinium Ed.). New York: Northwestern. Koufaris, M., Kambil, A., & LaBarbera, P. A. (2002). Consumer behavior in web-based commerce and empirical study. International Journal of Electronic Commerce, 6(2), 115-138. Li, N., & Zhang, P. (2002). Consumer online shopping attitudes and behavior: an assessment of research. Eight Americas Conference on Information Systems, 508-517. Lim, K. H., Sia, C. L., Lee, M. K., & Benbasat, I. (2006). Do I trust you online, and if so, will I buy? An empirical study of two trust- building strategies. Journal of Management Information Systems, 23(2), 233–266. Park, C. H., & Kim, Y. G. (2003). Identifying key factors affecting consumer purchase behavior in an online shopping context. International Journal of Retail & Distribution Management, 31(1), 16-29. Tjiptono. (2005) Service, Quality, Satisfaction. Yogyakarta: Andi Offset.