THE INFLUENCE OF SERVICE QUALITY, PRICE, AND BRAND AMBASSADOR … -

Previous studies reported inconsistent results about the effect of service quality and price on consumers' purchasing decisions. However, no study has been done about the impact of service quality, price, and brand ambassador on purchasing decisions of Tokopedia customers. This study aimed to evaluate the effect of service quality, price, and brand ambassador on buying decisions of Tokopedia customers. This study used a survey method with simple random sampling during the sample determination of Tokopedia customers. The 91 customers were given 44 questions with five alternative answers. The results showed that purchasing decision of Tokopedia customers was positively affected by service quality and brand ambassador. In contrast, the purchasing decision was negatively affected by product price. Therefore, purchasing decision made by Tokopedia customers was affected by service quality, price, and brand ambassador. Further studies are needed to obtain a general conclusion about the shopping behaviour of online shop customers in Indonesia.


INTRODUCTION
There is a lot of innovation in online shops and so-called e-commerce and marketplaces. The community is greatly assisted by an online store that can meet their needs. Interactions between sellers and buyers in online shops are carried out directly without intermediaries via SMS or WhatsApp chats. These market types provide benefits for both consumers and producers. Producers do not need to build physical JIGE 4 (2) (2023) 1011-1019 JURNAL ILMIAH GLOBAL EDUCATION ejournal.nusantaraglobal.ac.id/index.php/jige stores, and the market types could improve time efficiency. Consumers can ask the sellers questions about prices and product types and negotiate selling prices.
Good service helps consumers feel comfortable when shopping online. Online shopping carries a large amount of risk, including product quality. To minimize the risk, consumers must check the service of online shop providers through their Service Quality Score before deciding to buy the products. This check is necessary because consumers can only see products through photos and videos when shopping online. Variable service quality significantly impacts purchasing decisions. A previous study proved that service variability at Supermall Ciledug Tangerang has an average score of 3.408. The value indicates a good criterion for purchasing decisions (Aulia & Hidayat, 2017). In contrast, Noviyanti & Kusudaryanti found that service quality does not affect purchase decisions (p=> 0.05) (Noviyanti, R, D & Kusudaryanti, 2019).
Another factor affecting purchase decisions is the product's price. Previous studies proved that price significantly affects purchase decisions (Amanah et al., 2018). In contrast, Herlina reported that a product's price did not influence the buyer's purchase decision (Herlina, 2018). The third factor that might affect consumers' purchase decisions is a well-known brand ambassador. According to the previous studies, there was an inconsistent impact of service quality (Aulia & Hidayat, 2017;Debby et al., 2022) and product price (Amanah et al., 2018;Herlina, 2018) on purchasing decisions. Nevertheless, no scientific data evaluate the impact of service, price, and brand ambassador on buying decisions in Tokopedia shops. Tokopedia.com is an online shopping centre based in Indonesia with a marketplace business model. Tokopedia is the most viewed e-commerce by visitors (iPrice, 2019). This study aimed to evaluate the influence of service quality, product price, and brand ambassador on the purchasing decision of Tokopedia Online Stores. Buying decision is a step where the buyer makes an actual purchase of a product (Abriyani et al., 2021), while purchasing decision is a consumer activity in determining a purchase when facing two or more options (Firmansyah & Mahardhika, 2018).
Several factors affect the consumer's purchase decision to buy a product, such as service quality, price, and brand ambassador (Venessa & Arifin, 2017;Firmansyah & Mahardhika, 2018). Service quality is a gap between service expectations and the reality of the customer (Indrasari, 2019). Price is an activity in issuing product value to obtain a product or service (Wibowo & Karimah, 2012). A brand ambassador is a person given trust by the producer to sell the product for its reputation (Taft et al., 2013). Some indicators prove service quality. These indicators are reliability, responsiveness, assurance, empathy, and tangible (Kotler & Keller, 2017). Service also has some characteristics, i.e., business access, communication to customers, competence between rivals, politeness in serving customers, product credibility, reliability facing the customer, responsiveness to the customer question, security in product delivery, actual product, and understanding customer needs (Indrasari, 2019).
A price is some currencies that might be expensed to obtain the product or service (Wibowo & Karimah, 2012;Fransiska Vania Sudjatmika, 2017). A previous study reported that price significantly impacts purchase decisions, as evidenced by price discounts and product packages (Fernando & Aksari, 2017). It is vital when sellers and buyers interact (Venessa & Arifin, 2017). A product's price is affected by two factors, namely internal and external factors. Interna factors include production costs, marketing objectives, and sale strategies. External factors included competitors' products, prices, and market demand (Carmelita et al., 2017). Price also has some indicators. The indicators include reference price, market price, and product benefits. Reference prices include selling and buying price. The market price is necessary to provide a reasonable price. Product price must also be determined based on the product value (Wibowo & Karimah, 2012).
Producer often utilizes brand ambassador to communicate their product reputation (Firmansyah & Mahardhika, 2018). Companies commonly hire celebrities as their product ambassadors (Royan, 2005). Some criteria are needed for a good brand ambassador. The requirements include unique charms and benefits for the company. The role of a brand ambassador in communicating the product is to provide a product trial (testimony), offer to interest consumers (affirmations), an actor who appears in the commercial advertisement and who represents it, and become a provider of product directions to the customers (Royan, 2005). A brand ambassador has to consider four product attributes during their communication with the consumers. These attributes are product visibility, credibility, attractiveness, and power over achievement (Samosir et al., 2016). A good brand ambassador has some indicators: transference (openness to the products), congruence, credibility, attractiveness, and power (Lea-Greenwood, 2014).
Previous studies reported different impacts on service quality and product price. However, most of the studies were conducted in the conventional market. Buyers could see the service, product quality, and cost in this market type. Conversely, in the modern marketplace such as Tokopedia, buyers could only see product quality and price from the photo, video, brand ambassador advertisement, and market link. This marketplace carries some risk to the buyers. Therefore, it is hypothesized that service quality, price, and brand ambassador significantly influence the purchase decision made by the buyers on the Tokopedia marketplace.

METHODS
This study utilized Tokopedia online shop customers in Bogor City. The research was carried out from September 2021 to February 2022. The research used a quantitative descriptive method. Ninety-one respondents were selected as the samples using a simple random technique. The research measured service quality, price, and brand ambassador. Service quality was determined based on five parameters, i.e., service reliability, responsiveness, assurance, empathy, and tangibles. The cost was measured according to three parameters, i.e., reference price, price assumption and quality, and final price. The brand ambassador variable consisted of visibility, unity, supplier selection, and power. The respondents were exposed to 44 total questions. The questions had five alternative answers according to the Likert measurement method (Sugiyono, 2014)su Data were analyzed using partial least squares (PLS) using a structural equation model (SEM). The model of each variable was subjected to a validity test using three metrics. The three measures were convergent validity, combined reliability, and discriminant validity.

RESULTS AND DISCUSSION
This study obtained various responses from the respondent about the service quality, prices, and brand ambassadors at Tokopedia online shop. The scale distribution of the reaction is summarized in Table  2. The respondent answer scale was utilized to reconstruct a model using structural equation models (SEM) in the Smart PLS 3.0 program. The load value of the model was above 0.50. The obtained model is presented in Figure 1. We further analyzed the model using the outer loading test to ensure that this study obtained the best model for each variable. The model was evaluated using three metrics. The three measures are convergent validity, combined reliability, and discriminant validity.

Service quality
External stress test as service quality index was carried out using Outer loading validity test. The reflex indicator of respondents for the Service Quality variable (X1) model is presented in Table 3. It can be seen in Table 3 that the reflex indicators KP1, KP10, KP14, KP15, KP2, KP4, and KP8 have a loading factor greater than 0.70. This value means that all hands were valid. Therefore, the service quality variables' structure and results are promising.

Price
Outer loading validity test to evaluate the effect of external stress on product price was carried out using the Haga index PLS variable. The result showed that all reflex indicators had a value above 0.70. The detailed reflex indicator value is presented in Table 4. The data in Table 4 indicated that all configuration indicators were valid. Therefore, these indicators are effective for measuring variables' price structure.

Brand Ambassador
The result of the outer loading test for the brand ambassador model is presented in Table 5. It can be observed in Table 5 that the reflective indicators BA1 to BA9 have a loading factor above 0.70. This value proved that all configuration indicators were valid. Therefore, this metric analysis was practical to measure ambassador configuration variables.

Purchasing decision
Outer loading validity test to evaluate purchasing decision is presented in Table 6. Based on data in Table 6, the reflective indicators KPOT1, KPOT10, KPOT2, KPOT4, KPOT7, and KPOT8 have a loading factor > 0.70. This value means that all configuration indicators are valid. It is stated that the index structure measures the purchasing decision variable, and the results are said to be good.

Composite Reliability Test or Reliability Test
The reliability test is the tool for measuring a questionnaire indicator of a variable or structure. The measuring tool in the form of a questionnaire must provide stable or constant measurement results to prove that the instrument is reliable. Therefore, it is necessary to conduct a reliability test. A survey is declared valid or reliable if the answers from the sample to the statements are consistent and stable. The internal consistency method is used to test reliability. The reliability of the research tools in this study was tested using combined reliability, often referred to as Cronbach's alpha (Table 7).
A structure is reliable if Cronbach's alpha value is > 0.70. The AVE measure is used to measure the reliability of the latent variable component values, and the results are more conservative than the combined reliability. It is recommended that the AVE value is more significant than 0.50 (Fornell and Larker, 1981;Ghozali, 2014)  Tests based on Table 7 stated that the results of the combined reliability and Cronbach's alpha were shown to have good values, namely the value of each variable above the minimum value of 0.70, and the AVE value resulting from all the configurations above was> 0.50. This value means high consistency and stability of the tools used. In other words, service quality, price, brand ambassador, and purchasing decision variables are all excellent measurement tools, and all statements used to measure each component have higher credibility.

Discriminant Validity Test
Discriminant validity rests on the principle that various configuration variables may not correlate with high values. The result of the discriminant validity test in this study is presented in Table 8. The way to test discriminant validity using reflection indicators is to differentiate each AVE square root to the correlation value between structures. Based on Table 4.5 above, the diagonal line is the square root of AVE, and the lowest value is the correlation between constituents. Therefore, this model is valid because we found that the square root of AVE is higher than the correlation value that satisfies discriminant validity. According to Gozari (2015), if the AVE honest root value is greater than the correlation value between constituents, it can be stated that the experiment meets the discriminant validity criteria.

Structural Model Test or Inner Model
Internal models (internal relations, structural models, content theory) describe the relationship between latent variables based on material theory. Structural models are evaluated using the R-squared dependent latent variable. The changes in the R-squared value could be used to assess the effect of certain independent latent variables on the latent dependent variable (Ghozali, 2011). This research obtained an R2 value of the purchase decision variable was 0.891. This value proved a strong correlation between variables and purchasing decisions. The R-squared value of the purchase decision is 0.891 or 89.10%, indicating that the purchase decision variable can be explained by 89.10% by service quality, price, and brand ambassador variables. In comparison, the remaining 10.90% is explained by other variables that are not defined.

Hypothesis Test
The significance of parameter estimation could provide information about the relationship between research variables. The proposed hypothesis was tested using the structural (internal) model by looking at the path coefficient and the statistical significance of t, which indicates the parameter coefficient. The limit of rejection and acceptance of the hypothesis proposed above is the value sig P <0.05. The following table shows the estimation results for testing the structural model. The effect of the intrinsic weighting value of service quality (X1), price (X2), and brand ambassador (X3) on purchasing decisions (Y) was analyzed based on the path coefficient. The result of the hypothesis test in this study is presented in Table 9. The first hypothesis (H1) test The first hypothesis states that the effect of service quality on purchasing decisions is positive and can be ignored. Table 9 shows that the service quality variable has a significance level of 0.937 > 0.05. This data indicates that purchasing decisions have some excellent points but still need to be improved, meaning that H1 is accepted, so service quality has a positive but insignificant effect on purchasing decisions.

Second hypothesis (H2) test
The second hypothesis shows that price significantly negatively affects purchasing decisions. Table 9 shows that the price variable has a significance level of -0.019 < 0.05. This data indicates that the higher the product price, the higher the purchase decision, meaning H2 was rejected, so the price has a significant but negative effect on purchasing decisions.

The third hypothesis (H3) test
The third hypothesis shows a positive and significant influence on purchasing decisions among brand ambassadors. Table 9 shows that the Brand Ambassador variable has a significance level of 0.036 <0.036, 0.05. This value indicates that the more famous a brand ambassador is, the more likely they make purchasing decisions, and people like to shop online at Tokopedia, meaning that if H3 is accepted, then brand ambassadors can have a positive and significant effect on purchasing decisions.

CONCLUSION
Based on the result and discussion, this research concludes that purchasing decision of Tokopedia online shop consumers was significantly affected by service quality, price, and brand ambassador. Service quality and brand ambassadors had a positive effect, while price hurt purchasing decisions. This study only describes the shopping behaviour of Bogor City residents in the Tokopedia online shop. Further study is needed to obtain general shopping behaviour in online shops in other regions and other online shops in Indonesia.