The Influence of Product Quality and Word of Mouth on Wardah Cosmetics Purchase Decisions in Batam City

The aim of this study was to assess the impact of both product quality and creativity on the purchasing choices of Wardah Cosmetics customers in Batam City, both individually and in combination. This research employed a quantitative approach, utilizing a non-probability purposive sampling design or, alternatively, a probability sampling method. Data collection involved distributing questionnaires to 272 participants, and the analysis was conducted using SPSS version 25 software. The findings of the study are supported by the t-test statistic, where the calculated t-value for product quality (X1) was 10.649, exceeding the critical t-table value of 1.968. This indicates that product quality (X1) significantly and positively influences the purchasing decisions of Wardah Cosmetics customers in Batam City on a partial basis. Similarly, for the word of mouth variable (X2), the calculated t-value was 9.559, surpassing the critical t-table value of 1.968, indicating a significant and positive partial influence on the decision to purchase Wardah Cosmetics in Batam City. Furthermore, the study demonstrated a simultaneous and significant positive relationship between the product quality (X1) and word of mouth (X2) variables and the purchasing decision variable (Y), as evidenced by their combined F-value of 791.190, which is greater than the critical F-table value of 3.03. Additionally, the significance level for the product quality (X1) and word of mouth (X2) variables was 0.000, which is less than the significance threshold of 0.05. Consequently, this research supports the rejection of the null hypothesis (H0) and the acceptance of the alternative hypothesis (H1).


Introduction
In this era of industrial growt, all industry players are competing to sell goods or services to consumers.Increased competing to requires the industry to always pay attention to the needs and wants of consumers by developing product of superior quality compared to products offered by competitors.
According to Aaker (1994) in (Razak, 2019) product quality is the superiority with customer preferences or suitability with needs.There are various ways that the industry does is to carry out strict supervision when the product is processed.Consumers want to get quality raw product, althought there are some consumers who perceive high prices as quality product.
According to Kusumastuti (2013) in (Mawar et al, 2019) word of mouth is one of the promotional tools that people do to provide information about a product and have an impact on the assessment of a product.Word of mouth is marketing tool for building good relationships with consumers and making profits, where word of mouth comes from something that comes naturally and is not designed by the industry, by because it indirectly invites the closest person or other people to invite consumers curiosity about the quality of the product offered.The quality created does not have to be unique but can meet consumer desires.If thw quality of the product is not noticed by consumers do will result in a decrease in sales, where consumers do will not make repeat purchase.
According to Tjiptono (2008) in (Husen et al.,2018) the purchase decision is the preference of two or more options, where it is the consumer's decision to decide whether to buy the product or not.This stage is the consideration and completion stage in purchasing the product.The purchase decision can also be interpreted as the consumer's decision to actually buy the product.
Wardah is one of the well-known local cometics brands in Indonesia, a variety of wardah products intiated by PT PARAGON TECHNOLOGY AND INNOVATION which is known for its safe and quality.Wardah's various product include: moisturizer, lipstick, facial foam, parfume and so on.Wardah products are made from natural ingredients and referred to as a product that is labeled halal so as to provide convience for its users.Wardah in 2022 experienced a decline from the previous year, this was due to the many competitors with cosmetic product differentiation.What is the industry's effort for PT PARAGON TECHNOLOGY AND INNOVATION to continue to survive in the competition is to always produce products whit superior quality and product development compared to cosmetic products belonging to other industries.Industry events can be measured annually based on product developments in key markets and business turnover rates.Product quality can be adjusted by price, and the success of the marketing department influences customers decisions to purchase the product.Consumer buying decisions can increase the volume of sales to increase the level of profit made by the industry.If consumers can influence the consumer and make purchases and analyze consumers to develop product quality through word of mouth strategy, so that the industry competition become higher and more profitable for the industry itself.

Product Quality
According to Prajati (2013: 16) in (Arianto et al., 2022) product quality is a condition in which a product conforms to based on established standards.To [roduce good products, the industry needs to pay attention to whether the products manufactured meet the needs or wants of the customers, so that if the products meet the customers' needs, industry standards can be met.Which is used.If the industry continues to maintain the quality of its products, it will contribute to sales profits and customer loyalty.Products quality indicators in this research were taken according to Alfred (2013) in (Ristanti & Iriani, 2020) namely: 1. Performance: describes how a product works.2. Durability: describes how long a product will last.
3. Comformity: refers to the characteristics of the product that meets the standard.4. Priveleges: additional deatures that can increase the functional of a product.5. Aesthetics: the beauty of products that satisfy the five senses.6. Impression of uquality: this term refers to how consumers fell about a product.

Word Of Mouth
According to Pamuleh et al (2018) no one can build or create word of mouth, meaning that promotions carried out through word of mouth cannot be done by the industry but are carried out by consumers, where consumers will tell their experiences when using a product to their customers people around him.According to Lupoyoadi (2014) in (Husen et al., 2018) word of mouth suggestions about a product.Based on the explanation above, word of mouth as part of the promotional strategy does not come from the industry, but from consumers themselves, therefore to produce positive word of mouth promotions the industry needs to create ways that attract consumers so that consumers feel interested in sharing their experiences when buying these products.The word of mouth indicator in this research was taken according to Ronald (2016) in (Azlina & Noratika, 2022) as follows: 1. Do the talking: talking about positive things about a product that can ultimately change consumer understanding.2. Do the promoting: share information about a product with others by recommending the product.3. Do the selling: an activity that invites other people to use a product.

Buying Decision
The mai business can be drived when buying, purchasing decisions are closely related to the behavior of consumers when making a decision to buy a product.Consumers must decide which goods to use or consume to fullfill their desires (Yunus, 2018).When making an assessment, consumers will first check the product, brand, quantity to buy and other factors.Choices made usually based on consumer understanding of a product Alma (2016: 14) in Bakti et al (2021) defines purchasing decision occur after costumers evaluate a product.Through the general definition of purchase decisions, it can be concluded that purchase decisions are the final choices made by consumers regarding which products to purchase based on sonsumer evaluations.According to Kotler and Armstrong (2012: 176) in (Joesyiana, 2018) purchasing decision indicators are as follows: 1. Need recognition: need recognition is the initial stage in the buyer's decisions process, in which the consumer becomes aware of the need.2. Information search: at this stage consumers have a desire to find a lot of information about what to buy. 3. Evaluation of alternatives: based on the information received, consumers will evaluate alternatives from an existing set of choices.4. Purchase decision: is a decisions about what brand to buy, if there are no other factors that interfere.5. Post-purchase behavior: this stage where consumers feel satisfaction of dissatisfaction after making a purchase.

Hypothesis
Based on the proposed conceptual framework and theoretical framework, the hypotheses of this study are: H1: it is suspected that the quality of the product affect the purchasing decisions of Wardah cosmetics in Batam City.H2: is is suspected that word of mouth has an effect on Wardah's cosmetic the purchasing decisions in Batam City.H3: is is suspected that product quality and word of mouth simultaneously influence the purchasing decisions of Wardah cossmtics in Batam City.

Sampling
This research is characterized as a concise investigation due to its reliance on statistical data and estimations.The data analysis approach employed in this study is outlined below: 1. Descriptive analysis is a statistic used to analyzed data in order to draw general conclusions or general conclusions by describing or describing the data colloeted.2. Data quality control, including data accuracy and reliability testing.3. Classical hypothesis testing consists of normality test, multicollinearity test and heteroscedasticity tests.4. The test result consist of multiple linear regression analysis and analysis of the cooefficient of determination. 5.The hypothesis test consists of the T test and F test.
The objective of this research is to ascertain the impact of the independent variables, specifically product quality (X1) and word of mouth (X2), on the dependent variable, which is purchasing decisions (Y).This study follows a quantitative research approach, and it involves subscribing to a newsletter as a methodological step.
The research location is Batam City, especially female consumers who use Wardah cosmetics.Because the population in this research is not known with certainly, the researchers used Isaac and Michael's table to determine the sample size.Based on the following formula: Source: Sugiyono, Metologi Penelitian Bisnis, 2017

Sampling Technique
The researchers used a non-probability sampling method in this study.The non-popability technique is a way of taking samples whose population is not known and does not provide equal opportunities for each population unit to be sampled (Raihan, 2017: 94).The non-probability technique used is a purposive sampling technique because it involves taking samples while still paying attention to the researchers' considerations of the sample to be taken (Surahman et al., 2016:96).The characteristics used in this study are: 1. Female, aged 18-28 years.
3. Ever heard of wardah cosmetics from people around.4. Knowing about wardah cosmetics.

Validity Test Result
A data validity test is a method for assessing the validity and reliability of statement items used to assess the validity and reliability of statement items used to assrss the studied variables (Kurniawan & Puspitaningtyas, 2016: 97).The purpose of testing is to assess the validity of the questionnaire.A significant correlation coefficient test will reveal whether the questionnaire items are acceptable for use at the 0,05 level.The basis for decision making is as follows: 1.If the r count > r table is positive, the significant value is <0,05, then it is declared valid.
2. If the r count <r table value is positive, the significant value is > 0,05 then it is declared invalid.The validation test results are shown below:

Reability Test Results
The reliability test is designed to show how well the measurement results hold up when repeated measurements are made (Surahman et al., 2016: 114).All questionnaire items were subjected to a reliability test by calculating the Cronbach's Alpha correlation coefficient.If the alpha coefficient value is greater thans 0,70, the data is said to be reliable.Since there were more than 100 samples in this study, researchers used r table 30, which yielded a result of 0,349 with a significance level for a two-tailed test of 5%.The basis for decision making is: 1.If the r count > r table is positive, the significant value is <0,05, then it is declared valid.
2. If the r count <r table value is positive, the significant value is > 0,05 then it is declared invalid.The following shows the results of the reliability test:  1.3 shows that all statement items are considered reliable because the r count is greater than the r table and the r correlation is more than 0,70.

Classisc Assumption Test Normality Test
This test is used to determine whether the residuals are assumed to be normally distributed or not.Standard tests for normality are the Kolmogorov-Smirnov test, P-plot, and the bell curve for this experiment.Research shows that information that is collected is often shared.The residuals of the regression model are normally distributed.Typical test result are shown in the image below, useful for simplification.See the standard test result:  Data points are considered to exhibit a normal distribution when they are dispersed along a diagonal line in the same direction as the diagonal.Conversely, data is classified as having an abnormal distribution if the data points move counter to the diagonal line or fail to align with it.To verify the normal distribution of the data in this study, the Kolmogorov-Smirnov test was additionally employed.Kolmogorov-Smirnov test is as follows: 1.The Kolmogorov-Smirnov test value is above 0.05, suggesting that the data follows a normal distribution.2. A significance level below 0.05 in the Kolmogorov-Smirnov test indicates that the data does not conform to a normal distribution.

Source: SPSS Data Processing Results 25, 2022
The results of the Kolmogorov-Smirnov normality test, the significance value of the normality test is 0,228 > 0,05.Therefore, in this study it shows a normal distribution.

Multicollinearity Test
The independent variables that make up the equation need not be perfectly correlated.The multicollinearity test is used to determined whether there is a significnact correlation or relationship between independent variables.The reasons fot the adoption are as follows conclusion regarding the multicollinearity test: 1. Based on the tolerance value, multicollinearity does nt occur if the tolerance value is greater than 0,10.2. Based on the VIF value, multicollinearity does not occur if the VIF value is less than 10,00.Following are the results of processing multicollinearity test data: According to the data analysis results, it is evident that the tolerance value for both the product quality and word-of-mouth variables is 0.241, which exceeds the threshold of 0.10.Additionally, the VIF value for these variables is 4.141, which falls below the threshold of 10.00.Therefore, there are no indications of multicollinearity in this study.

Heteroscedasticity Test Results
If there is a problem with the heteroscedasticity test, it indicates that the variable has variance that is present or absent in the model.this condition can also be seen as evidence there is an imbalance between the model variant and the residual resulting from the observations used in the regression model.A heteroscedasticity test is used to determine whether a variable is derived from residuals of other residual variables.If the independent variable has a significant value > 0,05, it can be concluded that there are no signs of heteroscedasticity in this study.Below are the results of the heterokedasticity test: Based on the results of the Scatterplot Figure, it can be understood that there is data in the form of dotted graphs that point up and down and are around the number 0. The distribution of data points is not accompanied by a clear or persistent pattern.Therefore, the current research does not experience heteroscedasticity.

Influence Test Multiple Linear Regression Analysis Test Results
The technical structure and content of multiple linear regression analysis is similar to basic linear regression analysis.The result od the multiple linear regression analysis are as follows: Table 1.6 Multiple Linear Regression Test Results The regression equation derivered from this study is based on the data processing results: Y = 6,213 + 0,402 X1 + 0,531 X2

Model
The results of this processing lead to the following conclusions: 1.The product quality variable (X1) and word of mouth variable (X2) have a positive coefficient on the purchase decision variable (Y), indicating a positive relationship between the independent variable and the dependent variable.2. The two independent variables are product quality and the word of mouth value is 0 and the dependent vaiable is the purchase decision value of 6,213.3. The regressions coefficient of product quality variable is 0,402 against the word of mouth variable regression coefficient of 0,531.

Determination Coefficient Test (R2 Test)
Decision analysis can be used to estimate the percentage or total effect of independent variables on the dependent variable in regression model.Below is the result of the determination test: The test results in Table 1.7 indicate a correlation value of 0.925, signifying a strong relationship among the three variables in this study.Further examination of the test results reveals that the product quality variable, in combination with the word of mouth variable, contributes to an effect of 85.5%, while the product quality variable alone influences 14.5%.This is confirmed by the coefficient of determination, where the R-square value is reported as 0.85 and 0.855. .

Hypothesis testing T Test Results (Partial)
A useful test to determine whether the independent variable partially affects several variables significantly.Degree of freedom (df) = nk, where n is the number of samples = 272 and k is the number of research variables = 3, used to calculate the t table.Therefore df = 272 -3 = 269 and the significance level for the two-tailed test is 0,05.So α/2 = 0,05/2 = 0,025.Based on these calculations, this study's t table has a value of 1,968.The basis for making this test decision is: 1. H0 is rejected and H1 is accepted if tcount > ttable or a significant value less than 0,05.2. H0 is accepted and H1 is rejected if tcount < ttable or a significant value greater than 0,05.Based on the test results, it can be determined that the product quality variable (X1) exhibits a t-value of 10.649, which is greater than the critical t-table value of 1.968, and a significance level of 0.000, which is less than 0.05.Consequently, it can be concluded that this hypothesis demonstrates a positive and significant partial influence on purchasing decisions (Y) with respect to the product quality variable (X1).
Similarly, the t-value calculated for the word of mouth variable (X2) is 9.559, exceeding the critical t-table value of 1.968, and the significance level for the word of mouth variable is 0.000, which is also less than 0.05, as evident from the test results.Based on these findings, it can be inferred that this hypothesis partially exerts a positive and significant impact between word of mouth (X2) and purchasing decisions (Y)..

F Test Results (Simultaneous)
The F test is utilized to determine whether independent variables collectively influence the dependent variable.To initiate this analysis, we must first establish the F table.
The ANOVA table provides the outcome of the F test.Based on the computations, the F table value for this research is determined to be 3.03.The decision-making criteria for the F test are as follows:: 1. H0 is rejected and H1 is accepted if Fcount is greater than Ftable with a significance value lower than 0,05.2. H0 is accepted and H1 is rejected if Fcount is smaller than Ftable with a significance value greater than 0,0.Based on the result, it can be concluded that the hypothesis is that there is positive and significant influence between the product quality variables (X1) and word of mouth (X2) with the purchase outcome variable (Y) result.Processing of product quality data (X1) by word of mouth (X2).The value of word of mouth (X2) 791.190 > 3,03) f table) or the significance of the Product quality variable (X1) of word of mouth (X2) 0,000 < 0,05.

Conclusion
The findings from this research are that product quality has a partial effect on purchasing Wardah cosmetics in Batam.Word of mouth partially influences those who want to buy Wardah cosmetics in the city of Batam.Product quality and word of mouth have a large and large influence on purchasing decision for Wardah cosmetics in Batam.

Figure 1 . 4
Figure 1.4 Heteroscedasticity Test Results Source: SPSS Data Processing Results 25, 2022 Table Of Isaac and Michael

Table 1 .2
Validity Test Result Because r count is greater than r table, all clauses in the statement are declared valid using SPSS 25 as shown in Table 1.2.

Table 1 .5
Multicollinearity test results