THE EFFECTS OF PRODUCT QUALITY, SERVICE QUALITY AND ATTRIBUTES RELATED TO SERVICE AS BRAND SWITCHING ANTICIPATION AND THEIR IMPLICATION ON CUSTOMERS’ SATISFACTION (Empirical Study on the Customers of Clean Water in Nizam Zachman Ocean Area

Utilization of clean sources of SWRO products, tank water, and groundwater which are processed by themselves into clean water becomes an indication of brand switching by a company in Nizam Zachman Ocean Fisheries Port area. Research literature conducted by researchers based on re-purchase decisions led researchers to develop five empirical hypotheses, namely service quality, attribute related to services, product quality, brand switching and customer satisfaction. Data on observed variables were obtained through an interview with a questionnaire to 108 at clean water customer of Indonesian Fisheries Public Company (Perum Perikanan Indonesia) within Ocean Fisheries Port of Nizam Zachman Jakarta and tested by using AMOS statistic software 21. The test result showed that Brand Switching is statistically proven to be influenced by Product Quality, Service Quality and Attributes Related To Service and Customers’ Satisfaction statistically proven to be influenced just by Service Quality while product quality and Brand Switching are not statistically proven influenced on Customer Satisfaction.


INTRODUCTION
Referring to the Data of the Clean Water Supply during the 2014-2015 period, in Nizam Zachman Fisheries Port area Jakarta, managed by Indonesian Fisheries Public Company, when there was damage in one of the SWRO (Sea Water Reverse Osmosis) units had an impact on the reducing of clean water production. Tanks water from outside only slightly increased (May-August, October-December 2014 continued to February-April 2015). Especially in July, the demand decreased. This was due to Idul Fitri holiday in that month. When SWRO clean water production decreased, the demand for Tank Water only slightly increased. The companies in Nizam Zachman Fisheries Port area utilized the three sources of clean water, whether clean water of SWRO products, tank water or groundwater that were processed by themselves into clean water. This adjusted to the needs and the weather conditions. The unstable SWRO Clean Water production from time to time creates brand switching of clean water use, i.e switching to use Clean Water Tank Truck and the using of Ground Water from their drill well that are processed into clean water.
If you look at data from 2014-2015, the number of SWRO sales is increasing. However, if further examined, the target set is decreasing. In addition, although the number of SWRO customers still remains, the amount/volume of SWRO usage by customers is decreasing. These findings can be early indications of brand switching.
The movement of product/brand is influenced by many factors. Certain behavioral factors, times and competitive scenarios are complex phenomena that can affect the movement of products/brands to other products/brands (Srinivasan, 1996in Ishadi, 2012. Among the reasons is the quality of the product is not good, the level of service quality is not comparable with expectations, with the cost is too high and after-sales service is not in line with expectations. In addition to these findings, this study is interesting to do considering the discovery of differences in the results of previous researches on the factors that explain brand switching. The testing of the service quality and brand switching effects performed by Afzal, Sarwat et. al (2013), Jabeen, Hamid & Rehman (2015) asserted that service quality gives significant positive impact on brand switching. However, it is not the case with Kumaradeepan (2012) which shows that brand switching is significantly negatively affected by the service quality.
Other findings on product quality and brand switching have not been conclusive. In a study conducted by Anisa & Harti (2015), G. Biedenbach (2015), Loprang (2015), Khasanah & Kuswati (2013) and Wijaya, Fauzi, Sunarti (2014) showed that product quality had a positive effect on brand switching. However, in the study conducted by Yusup, Zainudin & Yoyok Soestyo (2014) showed that product quality was proven to have a negative effect on brand switching.
The quality of service impact on the customers' satisfaction also shows results that have not been conclusive. As in the study of Hizza & Cheng (2014), Sabir, et. al (2014) stated that the service quality gave a positive effect on customers' satisfaction. However, in Thuong study (2016) it showed a negative effect on the service quality to customers' satisfaction.
The study on customers' satisfaction described by product quality has been studied by previous researchers but the results are still different. In the study conducted by Hayati & Sekartaji (2015) also showed the effect of product quality to customers' satisfaction. The study conducted by Hidayat (2009), Sembiring, Suharyono & Kusumawati (2013, Wijayanto (2015) on the same variables showed a significant effect on product quality for customers' satisfaction. However, Alfin, et. al (2013) in his research shows the existence of product quality that has a positive impact on customers' satisfaction.
Referring to these findings then in this study the existing problems need to be formulated, namely: "explaining the brand switching behavior of clean water customers".

Customer Satisfaction
Customers' satisfaction is a reflection of the fulfillment of the needs, expectations and customers' wishes that give impact on the continuity of loyalty or repeat purchase (Band, 1991). The performance of agents that are interpreted as the quality of the agent becomes an important factor in explaining customers' satisfaction is (Mowen, 1995). The comparison of accepted performance and expectations is the function of the satisfaction level (Kotler, 1997). Therefore, the background of customers' expectations can be judged differently by its customers even in the companies with the same type of business.

Brand Switching
The distinguishing nature of names or symbols (such as logo, stamp, packaging) aimed to identify the goods or services of a seller with another group of sellers is the meaning of brand, according to Aaker (1991). Thus, the goods and services produced by competitors can be differentiated based on their brand.
Identity and differentiation on the competitor's product embedded in the form of symbol, name, term, design, color, motion, sign or combination of other product attributes is the definition of brand according to Tjiptono (2002). Certain reasons for consumers movement can also be interpreted as brand switching behavior which is vulnerable to switch to other brands (Dharmamesta, 1999).
Brand switching is consumers' behavior that is reflected through the shift from the brand product commonly consumed with other brand products.

Service Quality
Suitable requirements or demands, free of damages or defects, the suitability for the use of continuous improvement, customers' needs fulfillment, everything performed for happiness is quality definition often defined by the customers. Not only the aspect of the results emphasized, but also includes the process, environment, and human being is the quality definition in the perspective of TQM (Total Quality Management) is a broad quality. Products, services, people, processes, and environments that meet or exceed expectations related to quality dynamically is the definition seen in the formulation of Goeth and Davis cited Tjiptono (2012).

Attributes Related to Service
One of the factors that affect consumers' satisfaction is Attributes Related to Service. Attributes related to service includes (Ardhanari, 2008): 1. Guarantee or warranty describes the guarantees provided by producers of products or services in repairing or replacing products that are not in accordance with the order from the consumers. 2. Delivery refers to the speed and accuracy of shipments from the producers of products or services. 3. Complaint handling is an activity or effort made on complaints from handled complained consumers that are conducted by the manufacturers. 4. The problem-solving (resolution of the problem) related to the efforts of the producers trying to solve the problem complained by consumers when consumers complain. 5. Product benefits are the consumption of products produced by the manufacturer which benefits acceptable can be accepted by the consumers. 6. Product reliability and consistency is the reliability that the failure of the use of something in a certain period of time will not happen.

Product Quality
Quality can be defined as the characteristics of a product (goods or services) in total to support its ability to meet the needs. Anything that satisfies the consumers or meets the requirements or needs is also the meaning of quality. A set of related attributes (intangibles) associated in an identifiable form, within which are already covered the price, color, packaging, prestige of retailers, factory prestige, and the service of the factory as well as the retailers that may be accepted by the buyer as something that can satisfy the desire is the definition of product stated by Stanton (1985).

The Effect of Service Quality on Brand Switching
Getting and keeping customers are not easy tasks for any company. At the beginning of getting customers, every company can do it by using its various marketing strategies. So, getting customers in the early launching of products is easier to do. But to retain customers, there are not many companies manage to do this. Many companies' customers move to other products for various reasons. Brand switching is determined by many factors. The decided shift from one brand to another is a complex phenomenon whereby certain behaviors, competition scenarios, and time are the affecting factors (Srinivasan, 1996in Ishadi, 2012. Among the reasons are the quality of the product is not good, the quality of service is not as expected, the price is too high and after-sales service is not in line with expectations. Afzal et. al (2013) and Jabeen, Hamid & Rehman (2015) conducted a study that studies the effect of service quality on brand switching indicating that service quality is proven to have a significant positive effect on brand switching. However, this is not the case with Kumaradeepan (2012) study which stated that service quality significantly affects brand switching negatively.
The above description leads to the development of the following hypothesis: H1: Service quality affects brand switching negatively.

The Effect of Attributes Related to Service on Brand Switching
Attributes related to service are attributes of job satisfaction. Attributes related to service becomes part of service quality, meaning that attributes related to service provide support to the service quality. Attributes related to service can be given as a support for consumers to buy products or be given on post sales to increase loyalty.
According to Dutka (1994), attributes related to service may include (a) guarantee or warranty, is guarantee provided by a company on the product that can be returned if the performance of the product is not satisfactory; (b) complaint handling, is the handling of complaints made by consumers against the company; (c) delivery, is the speed and accuracy of the process of delivering the products and services the company provides to its customers; (d) resolution of problem, is the company's ability to solve the problems faced by consumers seriously. Unfulfilled aspects may lead consumers to switch to other brands.
The above description becomes a reference for developing the hypothesis as follows: H2: Attributes related to service negatively affect brand switching.

The Effect of Product Quality on Brand Switching
What is interesting and needs to be studied further by marketers who do not want to lose their customers is brand switching. Keeping their customers, in the long run, is the hope of every marketer even for good if possible. By knowing some of the factors behind the decision of brand switching, marketers can take advantage of brand switching behavior as a strategy to fit into a reasonable and attract competitors' customers. On the other hand, it becomes the behavior that companies need to avoid in maintaining their existence so that able to minimize the possibility of their customers moving.
Sukhla study (2004) found that product quality is the universal reason encountered when consumers perform the brand switching behavior. So, the important reason for brand switching is product quality. Similarly, studies conducted by G. Biedenbach (2015), Loprang (2015), Khasanah & Kuswati (2013), Anisa & Harti (2015) and Wijaya, Fauzi & Sunarti (2014) also show that product quality is statistically proven as factors that can affect the consumer's brand switching decision. However, the study conducted by Yusup & Soestyo (2014) on product quality and brand switching variables indicates that product quality has a negative effect on brand switching.
Referring to the above description, the hypothesis developed is as follows: H3: Product quality affects Brand Switching negatively.

The Effect of Service Quality on Customers' Satisfaction
The relationship between service quality and satisfaction theoretically as well as empirically is positive in the literature especially marketing which has been documented by Sonderlund (1988). When the given service is able to meet or exceed expectations or customers' expectations, then the customers are satisfied. It is a type of theoretical relationship (Andreaseen andLindestad, 1998 andParasuraman et. al., 1988).
Therefore the gap between customers' perceptions of service quality and customers' expectations on the services offered is a form of the paradigm that often arises in the definition of customers' satisfaction. Thus the quality of service is inherent with customers' satisfaction, where positively increase in service quality is used as a reflection of the increased of customers' satisfaction (Parasuraman et. al, 1998). Anderson et. al's research model (1994 expresses satisfaction as the function of the service quality explicitly.
The evidence that service quality and customers' satisfaction are positive are the empirical relationship appearing in various studies with different sample backgrounds. The proof of service quality affecting satisfaction is found in Sivadas study (2000( ) & Selnes (1993. Mukherjee and Nath study (2005), there are five dimensions of service quality that becomes an important factor for customers' satisfaction. The dimension of responsiveness service quality is the most important factor for bank customers in India, followed by reliability, empathy, tangible and assurance. This finding is in contrast to the results of Parasuraman, Zeithalm, and Berry (1991) studies which yield finding that reliability is the most important dimension of service quality (Mukherjee and Nath, 2005).
Andaleeb and Conway study (2006), states that by making a modification of service quality dimension it indicates the existence of positive effect on the three dimensions of service quality namely, responsiveness, reliability, and price to customers' satisfaction. The responsiveness dimension is the most influential factor in consumers' satisfaction.
Olorunniwo et. al study (2006), found tangible, responsiveness, knowledge and recovery are the factors that affect consumers' satisfaction. The factors of responsiveness and knowledge are the factors that have a higher effect on customers' satisfaction than tangible and recovery factors. While the tangible factor is the less important factor compared to the other three factors.
Customers' satisfaction that is only slightly affected by the service quality was found in Hellier (2002) study. Powpaka (1996 in Hellier (2002) also found that to improve customers' satisfaction comprehensively, high standard of service quality is important but not sufficient.
The above description becomes a reference for the development of the following hypotheses: H4: Service Quality has a positive effect on customers' satisfaction.

The Effect of Product Quality on Customers' Satisfaction
How the level of product quality offered determines customers' satisfaction but unfortunately, Naser et. al (1999) noted a lack of attention to the relationship between product quality and customers' satisfaction in the context of a service company. Product attributes affecting customers' satisfaction are shown in Naser's research (1999). While the performance of the product as perceived by customers which affect customers' satisfaction is shown in Selnes study (1993). While product quality is measured from customers' perceptions of the damage level of product affecting the level of satisfaction is found in Andreassen and Lindestad studies (1998).
Similarly, the results of other studies, such as Hayati & Sekartaji (2015) also indicate the effect of product quality on customers' satisfaction. The study conducted by Hidayat (2009), Sembiring, Suharyono & Kusumawati (2013), Wijayanto (2015) on the same variable also proves that customers' satisfaction can be explained by the existence of the qualified product.
The development of the fifth hypothesis is as follows: H5: Product quality affects Customers' Satisfaction positively.

The Effect of Brand Switching on Customers' Satisfaction
Customers' satisfaction survey conducted by Koskela (2002), Griffin (1995) found that 85% of them stated that they still wanted to move to another provider or supplier. This means that customer loyalty cannot be guaranteed only with customers' satisfaction (Singh, 2006 andStorback andLentinen, 2001). 75% of customers in certain industries who moved to other service providers said that they are satisfied or even very satisfied with the previous service provider. Substitution of the service provider by the customers is caused by the price or offer of new opportunities by the competitor, or the desire of variation. Factors such as switching costs, brand equity, market regulation, and loyalty programs are mapped in Fornell study (1992) in Lee et. al (2001) which determine the relationship between customers' satisfaction and loyalty. Hauser et. al (1994) in Lee et. al (2001) also stated that customers become less sensitive to satisfaction as switching cost increases. Switching cost plays an important role by making it of high value to move to another provider (Fornell, 1992in Lee, 2001. Anderson and Sullivan (1993) study in Lee (2001) suggests that high switching costs exist in the aircraft industry while in the bank's supermarket industry have low switching costs. The market structure also determines the effect of switching costs on the relationship of satisfaction and loyalty. The small effect of switching cost occurs on monopolistic markets. The absence of alternatives leads to unsatisfied customers will not move. Therefore, some switching cost providers are important.
However, in a study conducted by Putri, Arifin & Wilopo (2016) on brand switching and customers' satisfaction variables showed a significant effect of both variables with the positive effect direction. In contrast to the studies that have been done by Saleh et. al (2015) in which brand switching proved significantly had a negative impact on customers' satisfaction.
The above explanation leads to the development of the sixth hypothesis as follows: H6: Brand switching affects customers' satisfaction negatively.

Population
The population in this research was the clean water customers of Indonesian Fisheries General Company (Perum Perindo) within the Ocean Fishing Port of Nizam Zachman Jakarta, amounting to 239 customers.

Sample
Hair, et. al in Ferdinand (2005) found that the size of the sample corresponding to SEM was between 100 -200 samples. With reference to the opinion of Hair then the number of samples used in this study was 108 respondents.

Sampling Technique
Sampling technique used was purposive sampling that is sampling determination technique with a certain consideration. Customer samples were prioritized for customers with large water volume utilization. Because they were the customers who play a big role in the small or large amount of clean water production.

Service Quality Variables
Service quality variables are shaped by five dimensions which include questions related to SERVQUAL approach adopted from Aydin and Ozer (2004), Parasuraman (1994) which includes Tangible, Responsiveness, Assurance, Reliability, and Empathy.

Attributes Related to Service Variables
Attributes related to service variables are measured through six indicators adopted from Ardhanari (2008) covering warranty or guarantee, delivery, complaint handling, resolution of the problem, product benefit, and product reliability and consistency.

Product Quality Variables
The attributes of quality perception were formed by eight indicators adopted from Selnes (1993), Zhang (2001 including Appearance, Privilege, Trust, Conformity, Endurance, Ease in improvement, Aesthetics/beauty, and useful Quality.

Brand Switching Variables
Brand switching variables are measured using four indicators adopted from Zainudin & Soestyo (2014) studies that include Dissatisfaction experienced postconsumption, Desire for variation, Desire to speed termination of use, Decision on own desire, or by forced or accidentally discontinue the use of the product used and then proceed with purchasing and using other brand products.

Customer Satisfaction Variables
Customers satisfaction is measured using several indicators that are adopted from Petri & Graumann study (2014) which include Generally satisfactory, Can be recommended, Worth consuming for the present time and upcoming, Advantage and benefits are higher.

Method of Collecting Data
The questionnaire was chosen as the method of data collection in this study. The question type in the questionnaire is closed question by using Agree-Disagree Scale which results in strongly disagree answer -the answer strongly agree on the range of values 1 to 7.

Analysis Technique
To test the model and the relationship developed in this study required analysis technique. The analysis technique used in this research is Structural Equation Modeling (SEM) which is operated by AMOS program.

Evaluation of Data Normality
Estimation with Maximum Likelihood requires that the observed variable must satisfy the multivariate normality assumption. Normality analysis was performed by observing CR value for multivariate with range 2.58 at 1% significance level (Ghozali, 2004). Normality test results show that the value of CR for multivariate is 1.201 which is below 2.58, so it can be said there is no evidence that the distribution of observed variable data is not normal.

Evaluation of Outliers
The Mahalanobis Distance Test was used to see whether there are multivariate outliers in multivariate or not. To calculate Mahalanobis Distance based on the Chi-Square value on the degree of free 22 (number of indicators) at p level <0.001 is 2 (26; 0.001) = 54,052 (based on distribution table 2). Based on the results of data processing, it can be seen that the maximum Mahalanobis distance is 44.584 which is smaller than 2 table (54,052) which means that there are no multivariate outliers. So it can be concluded that there is no outliers problem in research data.

Evaluation of Multicollinearity and Singularity
From the results of data processing, it can be seen that value of the determinant of sample covariance matrix is 4626071,679 which is far from zero. So it can be concluded that there is no multicollinearity in the research data used.

Evaluation of Residual Value
From the results of the statistical analysis conducted in this study, the standardized residual value of covariance more than 2.58 is not found. So it can be concluded that residual requirements are met.

Feasibility Test of Research Model
The results of data processing for full model SEM analysis are described below.

Figure 1 Research Model Test
Source: Primary data processed, 2017 The first stage in testing the research model is to conduct a feasibility analysis of research model. The results of feasibility testing on the research model developed in this study is presented in Table 1. This research model shows that Chi-Square count value is 313,750, while Chi-Square critical/table value with df = 162 is 192,700. Since the Chi-Square count value in this study is smaller than the critical / table value, it means that this research model is not different from the estimated population/model which is considered good (accepted).

Hypothesis Testing
The fourth testing of hypothesis proposed in this research was done by analyzing the value of Critical Ratio (CR) and the probability of a causality relationship.

Hypothesis 1
The estimation parameter for testing the effect of service quality on brand switching shows CR value of -2.204 with probability is equal to 0,028. Therefore probability value <0,05 hence it can be concluded that service quality variable is proven significantly to have a negative effect on brand switching.

Hypothesis 2
The estimation parameter for testing the effect of attributes related to service on brand switching shows CR value of -2.039 with probability equal to 0,041. Because probability value <0,05 it can be concluded that attributes related to service variable are proven significantly to have a negative effect on brand switching.

Hypothesis 3
The estimation parameter for testing the effect of product quality on brand switching shows CR value of -3.382 with probability is 0.000. Therefore probability value <0,05 hence it can be concluded that variable of product quality is proven significantly to have a negative effect on brand switching.

Hypothesis 4
The estimation parameter for testing the effect of service quality on customers' satisfaction shows CR value of 2.070 with a probability of 0,038.
Because probability value <0,05 hence it can be concluded that service quality variable is proven significantly to have a positive effect on customers' satisfaction.

Hypothesis 5
The estimation parameter for testing the effect of product quality on customers' satisfaction shows CR value of 1,423 with probability equal to 0,155. Because probability value >0,05 hence it can be concluded that variable of product quality is proven to have an insignificant effect on customers' satisfaction.

Hypothesis 6
The estimation parameter for testing the effect of brand switching on customers' satisfaction shows CR value of 1.612 with a probability of 0.107. Since probability value >0,05 hence it can be concluded that variable of brand switching is proven to have an insignificant effect on customers' satisfaction.

Hypothesis Conclusions
The hypothesis proposed in this research are six hypotheses. The conclusions of the six hypotheses are as follows: 1. The test results of the service quality on brand switching can be concluded that service quality affects brand switching negatively. 2. The test results of the effect of an attribute related to services on brand switching can be concluded that attribute related to services does not affect brand switching. 3. The test results of the effect of product quality on brand switching can be concluded that product quality affects brand switching negatively. 4. The test results influence the quality of service to customer satisfaction can be concluded the quality of service has a positive effect on customer satisfaction. 5. The test results of the effect of product quality on customers' satisfaction can be concluded that product quality does not have a positive effect on customers' satisfaction. 6. The test results of the effect of brand switching on customers' satisfaction can be concluded that brand switching affects customers' satisfaction negatively.