THE EFFECT OF E-WOM, PERCEIVED QUALITY, ON PURCHASE INTENTION AT SHOPEE ONLINE STORE IN MANADO

: This study aims to find out how much influence e-WOM has on consumer purchase intentions that occur in the community in Manado City during the COVID-19 pandemic using one of the platforms that continues to grow, namely Shopee with the research title "The Effect of e-WOM, Perceived Quality, on Purchase Intention at Shopee Online Store in Manado." Electronic word of mouth is a positive or negative statement made by potential customers, actual customers, and former customers about a product or company via the internet. This study uses quantitative methods. To achieve this goal, the researchers distributed questionnaires and the sample used was 100 respondents. The results of this study indicate that e-WOM (X 1 ) partially has a significant effect on consumers' purchase intentions, while perceived quality (X 2 ) partially have no significant effect on consumers' purchase intentions, and e-WOM and perceived quality simultaneously have a positive effect on purchase intentions consumer. satu platform yang terus yaitu Shopee judul penelitian “Pengaruh e-WOM, Perceived Quality, terhadap Purchase Intention di Online Shopee Manado.” Electronic word of mouth adalah pernyataan positif negatif dibuat oleh calon pelanggan, internet. Penelitian


INTRODUCTION
The presence of COVID-19 has changed new habits in society. There have been many shifts and changes in consumer behavior that have occurred after the emergence of the pandemic. Starting from small habits in everyday life to habits that change lifestyle. Changes in consumer behavior today, of course, occur because of the encouragement of changing conditions that force them to adapt to these conditions. This is because consumers also need time to readjust to their old habits and behaviors. Consumers of course are now accustomed to shopping or doing activities with minimal physical contact and carrying out health protocols anywhere because they still don't feel safe since the presence of the COVID-19 pandemic. When there is a large-scale The statistical results of the multiple regression test indicated that eWOM quality, eWOM credibility, and eWOM quantity respectively; are significantly and positively impact purchase intention. In terms of the mediating role of corporate image the results indicated that corporate image mediate the relationship between the independent and dependent variables. Hanslim, Jaya, and Prasetyawati (2020) determined the effect of perceived quality on buying interest through the Culinaria Modena event. The results of the study stated that there is a strong influence between perceived quality on buying interest through events. This states that the event is a strategic marketing communication to build buying interest.

Figure 1. Conceptual Framework
Source: Literature Review

Research Hypothesis
H1: There is a significant effect of Electronic Word of Mouth and Perceived Quality on Purchase Intention H2: There is a significant effect of Electronic Word Of Mouth on Purchase Intention H3: There is significant effect of Perceived Quality on Purchase Intention

Research Approach
This research will use quantitative approach in order to collect the data. Quantitative research is the measurement of quantitative data and objective statistics through scientific calculations derived from a sample of people or residents who are asked to answer a number of questions about the survey to determine the frequency and percentage of their responses.

Population, Sample, and Sampling Technique
The target population and sample are explained below; include the sampling technique and sampling size, to describe the respondents' or informants' criteria before doing the research

Data Collection Method
Data collection method is a technique or ways that the researcher used to collect the data. The data consist of two types between primary data and secondary data. Primary data are gained from semi in-depth interview and secondary data are taken from several books, journals, and previous research.

Validity and Reliability
Validity is about the accuracy of the measure. According to Sugiyono (2018), Validity is a measure that shows the degree of accuracy between the data that occurs on the object with data collected by researchers. The questionnaire can be said to be valid if the statement in the questionnaire is able to express something that can be measured. According to Ghozali (2013), Reliability is a toll for measuring a questionnaire which is an indicator of a variable. A questionnaire can be said to be reliable if the respondent's answers to questions are consistent or stable from time to time. Therefore if measurements are made twice or more for the same symptoms the results remain consistent and can be trusted, then the measurement is considered to have a good level of reliability.

Multiple Linear Regression Model
According to Anderson et al (2011), Multiple regression analysis is the study of how a dependent variable y is related to two or more independent variables. The starting point of multiple regression analysis is the conceptual model and the hypotheses derived from that model that the researcher has developed in an earlier stage of the research process. The formula of multiple regression models in this research is as follows:

Result Validity Test
The purpose of validity test is to know whether the instrument is valid or not. Theoretically, if the correlation coefficient between the value of one indicator and the total value of all indicators is equal to and more than 0.3 (r ≥ 0.3) and below the significance level of 5%.  (2021).

Variable
The output of SPSS shows that the value of Cronbach's Alpha of 13 items used in this research is 0.927. The data is considered as reliable since the value of Cronbach's Alpha is above the minimum value which is 0.6.

Figure 2. Normality Test Source: SPSS Output (2021)
Figure shows that the dots are spreading near to the diagonal line and follow the direction of the diagonal line. Therefore, the normality test is accepted which means the data of this research is normally distributed.

Figure 3. Heteroscedasticity Test
Source: SPSS Output (2021) Recall that the Multiple Linear Regression model is used to determine the effect of several independent variables on a dependent variable. The computation was done using SPSS 22 software. From the result in the  Table, with level of significance is 0.05 (α = 0.05) and the level of confidence is 95% then Ftable is 3.09. The result is Fcount (68.132) > Ftable (3.09). Therefore, since Fcount is greater than Ftable, Hypothesis is accepted which means the independent variables simultaneously influence the dependent variable significantly.  (2021) The value of t-count of X1 is 5.404 since the value of tcount = 5.404 > ttable 1. = 1.984 meaning that hypothesis is accepted. The result of this declares that X1 (Electronic Word of Mouth) partially does have an effect on Purchase Intention (Y). 2. The value of t-count of X2 is 1.367 since the value of tcount = 1.367 > ttable = 1.669 meaning that hypothesis is rejected. The result of this declares that X2 (Perceived Quality) partially does have an effect on Purchase Intention (Y).

Electronic Word of Mouth on Purchase Intention
In this study, there are 4 indicators of e-WOM, namely platform assistance, online consumer reviews, loyalty, concern for other customers. In this case, the object is customers who shop through the Shopee platform. On the platform assistance indicator, most of them answered agree because they felt that the Shopee platform could make shopping online easier. So that it makes customers feel comfortable in shopping at Shopee. Furthermore, the online consumer review indicator, some respondents answered agree because they felt that they often read interesting reviews from other consumers on social media such as Instagram, Facebook, Twitter, and others. So that it makes customers feel interested in the products that Shopee offers. After that, the loyalty indicator, some respondents agreed because they felt that they would make the Shopee platform their subscription platform for online shopping. So they will make Shopee the most preferred online shopping platform. And the last indicator is concern for other customers, some respondents answered agree because they feel they will recommend Shopee to other consumers. So they will recommend Shopee to consumers who have never used Shopee as an online shopping platform. This means that e-WOM has a positive and significant effect on purchase intention at shopee. The results of this study support previous research conducted by Bataineh (2015) that eWOM significantly and positively impact purchase intention.

Perceived Quality on Purchase Intention
The results of multiple regression analysis show that perceived quality is partially insignificant or has no effect on purchase intention. The problem that may occur in terms of perceived quality is the quality of Shopee itself. Based on indicators of perceived quality, namely serviceability, reliability, product characteristics and conformance to specification, some respondents may still be dissatisfied with the quality of services provided by the Shopee platform, and the features available at Shopee are also still poorly understood by respondents such as chat features, Shopee live, and Shopee pay so that it can be concluded that there is no