THE EFFECT OF SERVICE QUALITY ON CUSTOMER SATISFACTION IN SHAPING CUSTOMER LOYALTY AT MCDONALD'S FAST FOOD OUTLETS IN THE JABODETABEK AREA

The aim of this research is to analyze and know the impact of service quality in shaping customer’s loyalty through customer’s satisfaction. The research uses non-probability purposive sampling technique and the research is quantitive research. The research uses questionnaires as the instrument to collect the data and analyzed with SEM. The total collected data are about 150 respondents. The result of the data research shows that, has five dimensions affect the service quality positively and the other results tells that customer’s satisfaction is a major factor to customer’s loyalty. The result of the research shows that there is no any direct relation between service quality and customer’s loyalty, so the direct effect of service quality on customer’s loyalty does not sustain on the research.


INTRODUCTION
McDonald's Corporation or commonly referred to as McD is one of the largest fast food or fast food restaurant franchises globally with the main dish at this restaurant is Burger. In addition to burgers, McD also provides other snacks, namely fries that are a pairing of burgers, local dishes from where the franchise of a restaurant is located, and soft drinks or carbonated drinks.  (Angga, 2017).
At this time, those in the world of trade feel more and more competing with each other businesspeople, and on the other hand, there are changes in competition that are increasingly felt in the world of trade making businesspeople in Indonesia increasingly improve the quality of their services to be able to satisfy customers with the aim of increasing customer loyalty to a particular brand. According to Dick and Basu (1994), one of the main parts of activities in marketing strategies is seen from various certain factors, namely the achievement of loyalty to customers through several marketing strategies (Siregar, 2004). According to Kotler (2019), service quality is one sense of assessment of consumers with the level of service that has been received with the level of service that is in accordance with expectations.
According to experts Castro and Wardrobe (1999), there is loyalty to customers which in addition to increasing business value, but there are also some things that will attract new customers (Beerli et al., 2004). In a short period of time, companies that can improve customer retention will see the greatest revenue gains. Profit is one of the most important avenues for consistency in business, because with the profits that can be generated, the business itself revolves around various products and services that are sold and offered (Soeling, 2007). In the long run, companies that increase service loyalty will be able to generate more targeted profits because the loyalty provided can make customers willing to pay higher prices than before.
Customer satisfaction is a way to create customer loyalty. There are various benefits obtained from customer satisfaction that will make a company to achieve high customer satisfaction targets. In addition, with the existence of customer satisfaction in the company that can improve performance on customer loyalty to achieve profits and customer satisfaction can also create a company for a turnover in customers, to reduce customer sensitivity to a price, can reduce additional costs if there is a problem with marketing, reduce in terms of costs that will be caused by remembering the amount to customers, make use of advertising to achieve success and can increase excellence in business (Fornell, 1992).
Companies that take the opportunity to carry out several activities such as improving in terms of effective service are one of the foundations for determining and making decisions in providing complaints against some consumer sampling from a learning on failure that will ultimately make consumer loyalty. Customer satisfaction is the most important reason for a company or business to continue to run smoothly.

Figure 2. Number of Charts on Indonesian Community Options
Source: cnnindonesia.com (2020) Growth in the food industry with the fast food restaurant sector in Indonesia shows that there is rapid progress , and the conclusion of figure 2 shows that fast food restaurants are more interested at 80% than food courts/hawker centers at 61% and in 3 rd place there are middle class restaurants/cafes at 22% and fine dining 1%. This shows that Indonesian people are more interested in eating at fast food restaurants.
Looking at this percentage figure, it can be concluded that there is a great opportunity for the fast-food restaurant sector in business growth. Various kinds of fast-food restaurants are also currently there are many companies that are increasingly competing. Increasing competition between fast food brands, makes every brand in Indonesia must be able to make their customers feel impressed when buying products that will make them differentiate themselves to the fast-food brand.
The more levels of competition, it will cause customers to pass through a phase where more choices in product, quality, and price are diverse, so that in the end customers will continue look for a value that they think can feel most satisfactory from several available product options (Kotler, 2005).
The quality of a product that is not satisfactory can cause a sense of dissatisfaction with customers, and it does not only happen to customers who have bought products at the fast food restaurant, but can also have an impact on others. Because if customers feel disappointed with the product, they will tell and share their experience with the environment around them about the fast-food restaurant (Lupiyoadi and Hamdani, 2006). With the number of fast food restaurants spread throughout Indonesia, there are cities that have the highest number of fast food restaurants, including in the Greater Jakarta area. With the Jabodetabek area which is an area with a dense population of approximately 21 million people and is also a city that has dense activities. The population who visits fast food restaurants in the Jabodetabek area is 12 million people from the total population of 21 million people in the Greater Jakarta area (Pratomo, 2018 In keeping up with the times and remaining a trend, McDonald's fast-food restaurants create fast food restaurants with a modern lifestyle. In addition, McDonald's has standardized globally in 20 global markets (McDonald's, 2018). Standardization itself is one of the main factors to support the success and progress of fast-food restaurants (Lee, 1987). With McDonald's vision to become a fast-food restaurant that has the best service globally, assurance and standardization are carried out through product quality, satisfactory service from employees to customers, cleanliness that is always maintained by McDonald's, and the safety of the products produced by them (McDonald's, n.d.).

Fast food restaurants
PT. Rekso Nasional Food, as the founder of McDonald's in Indonesia has carried out and supervised their standardization. With the standardization carried out by them, connect with customer is to recognize customers well so that customers feel liked and respected, then be authentic which is service and polite behavior to customers. Put get first, prioritizing customers when customers need help will be related to service quality and will affect customer satisfaction. McDonald's last policy to standardize is delicious food, which is ensuring that the food products produced are delicious, healthy, and also meet good food standards.
With the population density in the Jabodetabek area that makes fast food restaurants as their choice of place to eat, besides that the reason for making fast food restaurants as a place to eat is to be a place to meet some of their friends, socialize, and even make the fast-food restaurant a place to work. From this lifestyle, the Jabodetabek area was chosen as the focus of this study and from the focus of the study, the purpose of this study was to determine the correlation or on the quality of McDonald's services in the Jabodetabek area to McDonald's customer loyalty in the Jabodetabek area through McDonald's customer satisfaction in the Jabodetabek area that has been carried out by McDonald's restaurants.

Research Methodology
Research uses a quantitative approach motde, namely research is carried out based on the philosophy of positivism which aims to examine a certain population or sample randomly using data collection techniques using an instrument and data analysis using statistical analysis (Sugiyono, 2015). With the use of two types of data, namely primary data obtained through surveys and secondary data obtained by the non-probability sampling method, namely data collection techniques to the point that all data can have the possibility of being selected as data or samples that are not the same.
Sampling by means of non-probability sampling will use purposive techniques, which is when researchers select samples based on several criteria. Testing for sample criteria is variable loyalty, so to test this, the selection of respondents must be respondents who have visited and eaten at McDonald's outlets in the Jabodetabek area.
The target population in this study is regular undergraduate students who have visited and eaten at McDonald's outlets in the Jabodetabek area. The population is unlikely to be studied due to cost constraints, time constraints, and energy limitations. With the selection and sampling it can represent a population. The study also uses the application of Structural Equation Modeling or SEM which aims to find out and adjust the size of the sample to be studied. SEM or structural equation modeling itself uses a model to estimate samples, namely MLE or maximum likelihood estimation of 100 -200 samples. With the use of a research questionnaire that has 25 questions, the minimum sample size in the study must have as much as five times the number of questions. Therefore, the minimum number of respondents needed is 125 respondents. Therefore, to meet the use of respondents in this study, 150 respondents were used.

Figure 3. The Analyst Modelis
Source: Zeithaml et al. (1996); Schnaars (1991); Oliver (1997) With this analysis model, independent variables are obtained which are variables that affect variables or variables that are the cause of changes in dependent variables (Sugiyono, 2015), then dependent variables are variables that are influenced or variables that become a result due to independent variables (Sugiyono, 2015), and moderating variables or intermediate variables, namely variables that affect a relationship between variables that are being studied but cannot be measured, seen, and manipulated, with currents that must be inferred from the independent variables being studied (Sarwono, 2006). In this type of quantitative research, the use of two types of hypotheses is emphasized, the first is a hypothesis with one variable and a hypothesis with two or more variables (causal hypothesis) (Cooper and Schindler, 2006). For hypotheses that can be formulated are as follows.
H1 that is, there is an influence on the quality of McDonald's service in the Jabodetabek area to McDonald's customer satisfaction in the Jabodetabek area. There is a strong relationship between service quality and customer satisfaction (Fullerton and Taylor, 2002).
H2 that is, there is an influence on the quality of McDonald's service in the Jabodetabek area to the loyalty of McDonald's customers in the Jabodetabek area. The value of service quality is an aspect that influences the value of loyalty (Zeithaml, Parasuraman, and Berry;. H3 that is, there is an influence on McDonald's customer satisfaction in the Jabodetabek area to McDonald's customer loyalty in the Jabodetabek area. The value of satisfaction has benefits with a positive relationship with the value of customer loyalty (Caruana, 2002).

Results of Discussion
The model conducted for this study used the Maximum Likelihood Estimation (MLE) method. MLE itself is a method of estimating a parameter from several data by following a certain distribution distribution. With this model, the calculation will use the SEM method because the use of the SEM method is considered more efficient and easier to use. SEM is also efficient when the assumption of multivariate normality is met. Then discussion and testing is carried out on the model (model fit) with existing data and testing existing relationships.

Test Measurement Model
After making the Measurement model that can be seen in Figure 4, data is collected for testing and input. The measurement model itself is one part of the SEM model which consists of latent variables, namely in constructs and manifest variables, namely indicators. The purpose of testing is to be able to determine the accuracy value of manifest variables that can explain the existence of latent variables.
In the first testing section to assess the Goodness of Fit or GOF. The GOF calculation results are obtained from calculations in the SEM model which can be seen in table 3.  (Ghozali, 2009). If there is one or more parameter values that have been fitted, the model is declared fit or feasible (Solimun, 2005).
After that there is a validity test. Validity testing is carried out when the model obtained is assessed fit, after that a construct validity test is carried out which is an assessment of indicators against constructs that can explain a construct. The test is carried out through a convergent validity test which is an assessment of the indicators that make up a construct and the construct has a value on the loading factor. The value of the loading factor can be seen in the following table. From the results of the table research above, all values on the loading factor have a value of > 0.5. Thus, these indicators meet the requirements in the model in forming construct indicators on service quality, customer satisfaction, and customer loyalty. As for the value of AVE or Variance Extracted, there are those who show a value of more than 0.5 which means that there is good convergent validity. This value shows that the indicators above can represent variables that are being developed.
It can be seen from table 5, from five dimensions can be sorted the strongest dimensions consecutively, namely, Empathy, Responsiveness, Tangible, Assurance, and Reliability. There is a quality determining dimension of the importance value according to the value of the customer. With a very critical Empathy value and followed by the dimensions of Responsiveness, Tangible, Assurance, and Reliability.
Fromthe calculation above, the value of p is 0.000 so that it can be concluded from all five dimensions above that affect significantly. With the relationship of these five dimensions to service quality, including Tangible which has an estimated parameter of 0.733, Empathy has an estimated parameter of 0.872, Reliability has an estimated parameter of 0.690, Responsiveness has an estimated parameter of 0.849, and Assurance has an estimated parameter of 0.696. It can be concluded that the five dimensions above affect the construct of service quality.  After that, there is a reliability test on the construct. Reliability is a measure of the internal consistency indicator of a construct. For results, a large reliability value will provide an assessment that means an individual's indicators will not change or be consistent with their measurements. For the calculation of construct reliability can be seen in table 6. An assessment of construct reliability above 0.70 means that the reliability value is good and table 6 shows that the construct reliability value obtained is feasible for use in the model.
For the P value, table 6 means a significant level of 5% because it is close to 0.000 and has a relationship between customer satisfaction and service quality, customer satisfaction and customer loyalty, customer loyalty and customer satisfaction. For the standardization process shown in table 7, a high correlation value of >0.5 indicates a positive and one-tailed relationship. From this it can be concluded that the greater the value on the quality of service that has been provided by McDonald's, the value given by customers will be more satisfied with McDonald's, including for the other two constructs. From table 7, the closest relationship is the relationship between customer satisfaction and customer loyalty. For

Structural Model Test
After testing the measurement model, testing was carried out on the structural model. The difference in this structural model is that there will be independent variables and dependent variables because the understanding of the structural model itself is a construct relationship that has cause and effect. For this model, the independent variable is service quality and for the dependent variable is customer satisfaction and customer loyalty. For the calculation results, the calculation formula or equation is obtained as follows. Assurance = 1.000 KL, Responsiveness = 0.828 KL, Reliability = 0.722 KL, Empathy = 0.687 KL, and Tangible = 1.007 KL. From this equation, it can be explained that the values of the five dimensions are positive and significant. In conclusion, the higher the value of the five dimensions, the higher the value of the quality of service provided by McDonald's according to customers.
The next analysis is the analysis of direct effect, indirect effect, and total effect. This analysis aims to determine the strong influence between constructs both directly, indirectly, and totally. Direct effect is the coefficient of all lines with arrows having one end. Indirect effect or indirect influence is an influence that arises through intervening variables. Total effect is the influence of all relationships. For the calculation of direct effect of this study can be seen in table 10. For this study also has no indirect effect or indirect effect, so the value of direct influence or direct effect is equal to the value of total influence or total effect. In table 11 for total effect, it can be explained that if one point of value is increased on these five dimensions, it will increase the value of service quality. For assurance one point value will increase the value of service quality by 0.981 points, for responsiveness by 0.896 points, for reliability by 0.760 points, for empathy by 0.632 points , and for tangible by 0.682 points. Of the five dimensions, the most efficient dimension to increase the value of service quality is assurance.
After doing all these tests and calculations, researchers can make modifications to the model that has been developed. From the modification of the model, if it is obtained that the estimated value has a prediction level that is not as expected, it means that there is a high residual value. If this happens, then you can see the value of standardized residual covariance matrice. If there is a value of >2.58 in the standardized residual covariance matrice, then there is a problem in the formation of the model. For this study, there was no standardized residual covariance matrice value of >2.58 which means that the model is accepted.

Hypothesis Testing
The results proposed for the hypothesis in the study can be seen in table 6 with the following conclusions. There is an influence between service quality and customer satisfaction, the first hypothesis (H1) which reads "There is an influence on the quality of McDonald's service in the Jabodetabek area to McDonald's customer satisfaction in the Jabodetabek area". From the test results obtained have a value of 0.720 and a p value of less than 0.005, the customer satisfaction variable has a positive and significant effect on customer satisfaction.
According to Eko Cahyadi and Hari Sandi Atmaja in a similar study, service quality will affect customer satisfaction with the meaning that service quality itself is the main key in outperforming other companies.
Then for the second hypothesis (H2) which reads "There is an influence on the quality of McDonald's service in the Jabodetabek area to McDonald's customer loyalty in the Jabodetabek area". From the test results obtained have a value of 0.814 and a p value of less than 0.005, the customer satisfaction variable has a positive and significant effect on customer satisfaction.
In addition to service quality, there is also a small effect of service quality that can be assessed and analyzed by other factors but can affect customer loyalty strongly. There are three basic performance scores for a restaurant, among others, service quality, food quality, and price quality (Gupta, McLaughin, and Gomez;2007) Finally, for the third hypothesis (H3) which reads "There is an influence on McDonald's customer satisfaction in the Jabodetabek area to McDonald's customer loyalty in the Jabodetabek area". From the test results obtained have a value of 0.935 and a p value of less than 0.005, the customer satisfaction variable has a positive and significant effect on customer satisfaction.
According to Eko Cahyadi and Hari Sandi Atmaja in a similar study, customer satisfaction will depend on the company's performance in meeting customer expectations and expectations. If the company's performance is not in accordance with customer expectations, customers will not be satisfied but if the company's performance meets customer expectations, customers will be satisfied (Kotler, 2005). Customer satisfaction will have an impact on purchasing behavior in customers until the future.

CONCLUSION
From the results of the research discussed, it was shown and obtained that the five dimensions that make up the quality of service tested proved to be able to affect service quality positively and significantly. The dimensions that most influence the quality of service are Empathy, Responsiveness, Tangible, Assurance, and Reliability. The results also showed a positive and significant influence on the quality of McDonald's service in the Jabodetabek area to McDonald's customer satisfaction in the Jabodetabek area. With calculations on the quality of McDonald's service in the Jabodetabek area to McDonald's customer satisfaction in the Jabodetabek area by 72%, the quality of McDonald's service in the Jabodetabek area to McDonald's customer loyalty in the Jabodetabek area by 81.4%, and McDonald's customer satisfaction in the Jabodetabek area to McDonald's customer loyalty in the Jabodetabek area by 93.5%.