Analysis of the Influence of the Marketing Mix on Consumer Purchasing Decisions Using the Structural Equation Modeling Method

This study aims to determine the effect of the 7P marketing mix, namely product, price, location, promotion, people, physical evidence, and process on Budiman Swalayan Jembes’s consumers that affect their decision-making when purchasing at Budiman Swalayan Jembes. This study uses the Structural Equation Modeling (SEM) method with the Smart PLS version 3.0 program. The data used in this study are data obtained from respondents totaling 155 people through filling out questionnaires. Data analysis conducted in this study is an evaluation of the measurement model (outer model), evaluation of the structural model (inner model) and hypothesis testing. The results of this study indicate that of the seven variables studied, three of them influence consumer purchasing decisions. The three variables that influence consumer purchasing decisions of Budiman Swalayan Jembes are product variables, price variables and location variables. As for the other four variables, namely the promotion variable, the people variable, the physical evidence variable and the process variable, the analysis shows that there is no influence on consumer purchasing decisions at Budiman Swalayan Jembes.


A. Introduction
The development of retail businesses or retail businesses such as supermarkets in Indonesia is quite rapid, this is marked by the increasing number of traditional retail businesses which are starting to reform themselves into modern retail businesses as well as the emergence of new modern retail businesses so that competition between these retail businesses is increasing and is very competitive.Retailers use various methods to win in business competition.A retailer must be able to see marketing opportunities that will affect its business.Marketing opportunities are opportunities for retailers to increase sales and profits in ways that generally satisfy consumers.Usually, retailers as companies in their efforts to increase sales carry out intensive promotions, add new products, open new outlets, increase prices or reduce costs, or other methods such as renovating outlets (Ma'ruf, 2005).
On the one hand, the presence of a retail business is very beneficial for consumers.In terms of shopping, consumers can enjoy comfort, safety, convenience, increasingly diverse product variations, increased product quality and of course lower product prices due to competition.(Widyastuti et al., 2020).In facing increasingly fierce competition, retailers are required to be able to anticipate changes that occur in the market and provide appropriate responses to meet the increasingly diverse needs and desires of consumers.
To attract consumer interest, retailers must be able to implement appropriate strategies to be able to compete with other retail businesses (Wichmann et al., 2022).Retailers must know what factors influence consumer purchasing decisions.The marketing mix is one of the marketing strategies that companies often use to influence consumer purchasing decisions (Abril & Rodriguez-Cánovas, 2016).A marketing mix is a set of marketing tools used by a company to reach its target market.This marketing mix is classified into seven groups, namely product, price, location, promotion, people, physical evidence and process (Kotler & Armstrong, 2016) Budiman Swalayan is a retail business that has been around for a long time in Bukittinggi City.Budiman Swalayan was originally founded in 1999 by H. Yasmar in the form of a Budiman daily goods shop located on Jalan Soekarno Hatta number 26 Pasar Bawah Bukittinggi City, now it has become Budiman Swalayan with an area of 240 M².Furthermore, in 2004 the Budiman daily goods shop was established which was located at Pasar Aur Kuning Bukittinggi, until 2010 the Budiman daily goods shop changed to Budiman Swalayan as well as relocating to its own land and building on Jalan Sutan Syahrir number 4 Tarok Dipo, Bukittinggi City.
In  From the process of Budiman Swalayan in establishing its branches since its founding in 1999, it can be seen that Budiman Swalayan has developed very rapidly among other supermarkets.This development is very significant, especially from 2013 to 2023.The average time span for the establishment of each branch is from one to two years and there are several branches established in the same year.
If you look at the table above, the supermarket that has been around the longest is Masyithah Swalayan, namely from 1986, but the number of branches currently only has one main store located on Jl.Prof. Hazairin, Bukit Cangang Kayu Ramang, District.Guguk Panjang, Bukittinggi City.Meanwhile, Niagara Swalayan, which was founded in 1996 until now, only has 4 branches located in Bukittinggi City and Payakumbuh City.When compared to these two supermarkets, Budiman Swalayan is growing very rapidly with a total of 16 branches throughout West Sumatra.From the table above, it can be seen that among the three supermarkets, the development that is quite competitive with Budiman Swalayan is Hoki Store, which has had 3 branches since its establishment in 2015.
The development of Budiman Swalayan which has increased significantly in the last few years cannot be separated from consumers who choose to shop at Budiman Swalayan so that Budiman Swalayan can develop rapidly.From the explanation above, it can be seen that Budiman Swalayan is developing very rapidly compared to other supermarkets in Bukittinggi City which have been around longer than Budiman Swalayan, but apart from its good development, Budiman Swalayan also has to face tough competition with other new supermarkets in Bukittinggi City, namely Hoki Store which can have 3 branches within 7 years.
The rapid development of retail businesses in Bukittinggi City has made competition between retail businesses increasingly tight.This makes Budiman Swalayan have to choose the right business strategy to excel in competition with other retail businesses in Bukittinggi City.According to the author's observations, one of the things that is different about Budiman Swalayan is the Islamic nuances applied in this retail business.This Islamic nuance can be felt directly by consumers through the polite and closed uniforms of Budiman Swalayan employees and the songs played in the shop are religious songs.
Apart from having the differences as explained above, to be able to maintain and increase the number of consumers, of course other appropriate business strategies are needed and Budiman Swalayan must be able to fulfill the factors that influence consumers' purchasing decisions.Understanding how and why consumers make purchasing decisions will help Budiman Swalayan make the right decisions to determine future business strategies.
Based on the above phenomenon, the author is interested in conducting research with the title "Analysis of the Influence of the Marketing Mix on Consumer Purchasing Decisions Using the Structural Equation Modeling Method (Study at Budiman Supermarket Jembes)" .

B. Research Methodology
This resear,cher use type study quantitative.According to (Sugiyono, 2015), quantitative research methods can be defined as research methods based on the philosophy of positivism, used to research certain populations or samples, collecting data using research instruments, quantitative/statistical data analysis with the aim of testing predetermined hypotheses.
The place of this research is Budiman Swalayan Jembes which is located on Jl.Sutan Syahrir No. 4, Tarok Dipo Village, Guguk Panjang District, Bukittinggi City.In this research, the population is all consumers or customers who have visited and made purchases at Budiman Swalayan Jembes, the number of which is not known with certainty.Considering the large population of consumers and the number is not known with certainty, in determining the sample the researchers were guided by Hair's opinion.According to the minimum sample (Hair et al., 2010) size for a population whose size is not known with certainty is 5-10 times the indicator variable.So the number of indicators is 31 times 5 (31 x 5 = 155).So the number of respondents in this research was 155 customers of Budiman Swalayan Jembes.
The data in this research was collected by distributing a questionnaire via Google Form which contained closed questions and statements to customers or consumers of Budiman Swalayan Jembes regarding the research conducted, namely regarding the analysis of the influence of the marketing mix on consumer purchasing decisions at Budiman Swalayan Jembes.Apart from using online questionnaires, researchers also distributed questionnaires offline .
This research uses a Likert scale for variable measurement.The Likert scale is used to measure the attitudes, opinions and perceptions of a person or group of people about social phenomena.In this research, social phenomena have been specifically determined by researchers which are referred to as research variables (Sugiyono, 2015).The variables in this research consist of independent variables and dependent variables.The independent variable consists of the 7P marketing mix ( product, price, place, promotion, people, physical evidence, and process ), while the dependent variable is the purchasing decision.
Data analysis in this research was carried out using the Partial Least Square (PLS) method.PLS is a component or variant-based Structural Equation Modeling (SEM) solution method.Partial Least Square (PLS) is a fairly strong analysis method because it is not based on many assumptions.The data used also does not have to have a multivariate normal distribution.Indicators with categorical, ordinal, interval and ratio scales can be used in the same model, and the sample used does not have to be large (Ghozali & Latan, 2012).In this research, the analytical tool used is the SmartPLS application for data processing.The data analysis carried out is as follows:

Convergent validity
Testing the validity of indicators with the SmartPLS program can be seen from the outer loading value for each construct indicator (Ghozali & Latan, 2015).The Rule of Thumb for assessing convergent validity is the outer value loading between 0.6-0.7 (Ghozali & Latan, 2015).In this research, the outer loading value used is above 0.6. .Apart from the outer loading value, convergent validity can be seen from the AVE ( Average) value Variance Extracted ).For good requirements, the AVE value used is ≥ 0.5 (Hair et al., 2010).

Discriminant validity
The discriminant validity test can be seen in the cross loading value, if the cross loading value of each variable statement item to the variable itself is greater than the correlation value of the statement item to other variables (Ghozali & Latan, 2012).

Composite reliability
Reliability tests are used to prove the accuracy, consistency and precision of instruments in measuring constructs (Ghozali & Latan, 2015).The reliability test can be seen from the composite reliability value.The composite reliability value is greater than 0.7 (Ghozali & Latan, 2015).

Structural Model Evaluation (Inner Model)
Testing of the structural model was carried out by looking at the R-square value (R2 ) .Changes in the R-square value can be used to assess the influence of certain independent latent variables on whether the dependent latent variable has a substantive influence (Ghozali & Latan, 2012).Apart from looking at the R-square value, the Partial Least Square (PLS) model was also evaluated by looking at the Q-square predictive relevance for the constructive model.Q-square measures how well the observed values are generated by the model and also its parameter estimates (Ghozali, 2016).

Hypothesis test
PLS does not assume normally distributed data, instead it relies on a nonparametric bootstrap procedure to test the significance of its coefficients (Hair et al., 2014).Hypothesis testing can be seen from the t-statistic value and probability value.To determine the t-table value, you need to calculate the degrees of freedom (df) value with the formula df = number of samples-1.To test the hypothesis using the t-statistical value, for an alpha of 5%, a sample size of 155, the df value is 154, so the t-table value used is 1.98.So the criteria for accepting or rejecting a hypothesis are Hi accepted and H0 rejected when the tstatistic ˃ 1.98.To reject or accept a hypothesis using probability, Hi is accepted if the p value ˂ 0.05 (Hussein, 2015).From the results data processing with SmartPLS shown in Table 2, the majority indicators for each variable in study This own greater outer loading value big of 0.6 and declared valid.Besides that there are 2 indicators which has the outer loading value is less of 0.6 , namely the KP4 indicator shows 0.593 and the LK4 indicator shows 0.465.Indicator variables that have outer loading value is more small of 0.6 have level low validity _ so that indicator the need eliminated from the model.Outer loading value after KP4 and LK4 are eliminated can seen in Table 3 following This: Based on Table 4, there was an increase in the outer loading value for the KP2, KP3, PD4, LK1, LK3, ORG3, BF1, BF2, BF4, PS1, and PS2 indicators.Apart from that, there was a decrease in the outer loading value for the KP1, PD2, PD3, HG4, PM1, PM2, and ORG4 indicators.However, the decrease that occurred in the outer loading indicator value did not cause the outer loading indicator value to be smaller than 0.6.The results of this outer loading test show that all indicators after elimination have an outer loading value above 0.6, which means all indicators have met the convergent validity criteria.

Measurement Model Analysis (Outer Model) Convergent validity
After carrying out the convergent validity test using the outer loading value, the Average Variance Extracted (AVE) value is then looked for.After processing the data using the SmartPLS application, the results of the AVE values can be seen in Table 5 below: Based on Table 5, all variables show an AVE value ≥ 0.5 with the highest AVE value being 0.78 for the location variable and the smallest AVE value of 0.50 for the product variable.The AVE value meets the requirements according to the minimum AVE value limit, namely 0.5.

Discriminant validity
Once done data processing with using the Smart PLS program cross loading results can be obtained seen in Table 6 following this: Based on Table 7, it shows that the composite reliability value for all variables is greater than 0.7.In accordance with the required composite reliability value limit, namely above 0.7, all variables in this study are declared reliable.

Structural Model Analysis (Inner Model)
The R-square value is used For evaluate influence variable independent to variable dependent .Once done data processing with use application SmartPLS results R -s quare value can be seen in Table 8 following This : Based on The R -s quare value in Table 8 above shows that variable product , price , location , promotion , people, evidence physical , and process capable influence variable decision purchase by 48% and the remainder 52 % is influenced by other variables outside those studied in study This .
Furthermore is testing predictive relevance or Q-square, goal testing This is measure how much Good mark observation generated by models and estimates the parameters.If the Q-square value > 0 then can said own mark good observation, meanwhile _ If Q-square value < 0 then can stated mark observation No Good .

Research Discussion of
The Influence of Sharia Compliance on Purchase Decisions Sharia Compliance can be defined as the ability of social networking sites selling their products to comply with and operate under the principles of Islamic economics.Assuring compliance with Sharia principles is crucial in conducting business by Sharia.The hypothesis test results in this study prove that Sharia Compliance does not significantly influence online purchases among millennial Muslim consumers.This research is in line with the study by Putri Diah Kartika, Rizal Agus (2021), which found that Sharia compliance does not have a significant partial influence on purchasing decisions.However, simultaneously, Sharia compliance does affect purchase decisions.Rohali et al. (2022), in a study examining the factors influencing companies' decision-making to transact with Islamic banks, found that Sharia compliance does not have a positive impact on transaction decisions in Islamic banks.
An interesting finding from this research is that Sharia compliance is a distinguishing factor in the behavior of Muslim consumers compared to other consumers, representing a unique behavior among Muslim consumers.However, it appears that Sharia compliance does not yet have an impact on the purchasing decisions of Generation Z Muslim consumers.

Trust significantly influences online purchase decisions
The hypothesis test results prove that trust significantly influences online purchase decisions.Trust can be interpreted as the belief in the ability and integrity of online shopping services on platforms like Shopee.Customer preferences require online businesses to be adaptive and maintain consumer trust to ensure continued transactions.In the final stages of the consumer's journey after completing a transaction on an online store, consumers will evaluate the performance of the app's usability and the services provided by the online store.
After purchasing and consuming a product, consumers will experience varying levels of satisfaction or dissatisfaction.If consumers feel confident in the service, they often express their opinion through star ratings, a popular rating scheme in online stores.This research aligns with a study conducted by Giampietri et al. (2018), where the trust variable directly influences purchase decisions, meaning that trust plays a significant role in influencing consumer behavior when making purchases.

Usability significantly influences online purchase decisions
The hypothesis test results in this study confirm that usability significantly influences online purchase decisions.In online trade transactions, usability is one of the factors considered by consumers.Consumers who find the online shopping system easy to use are more likely to make purchases and repeat them in the future.This research is consistent with previous studies, such as the one conducted by Wahyuni et al. (2017), which found that the usability variable has a positive and significant impact on consumer purchase decisions on Zalora.co.id.Initially, when someone or a potential consumer wants to shop online for the first time, they may encounter difficulties due to lack of familiarity with the process.Consumers who find online shopping systems difficult to navigate may abandon their intention to shop online and opt for offline stores that are more accessible.However, on the contrary, when someone experiences ease in online shopping, they are more likely to continue using online shopping apps.

Trend significantly influences online purchase decisions.
The hypothesis test results prove that trends significantly influence online purchases.This indicates that trends have a significant impact on online purchases.In online trade transactions, trends are one of the factors influencing consumer interest in purchasing a product.This is supported by Yolinda, Winatha, & Nurdin (2018) in their research, which found that trends have an influence on consumer clothing purchase decisions.However, Jayanti's study (2020) yielded different results, indicating that the trend variable, when considered individually, does not significantly affect purchase decisions.In this era of technological advancement, a trend has emerged in society in terms of consumer shopping behavior, with consumers preferring to shop online.Putro & Haryanto (2015) argue that online shopping behavior (also referred to as online buying behavior, internet shopping behavior, or shopping via the internet) involves the process of purchasing goods or services through the internet.
Sharia Compliance, Trust, Usability, and Trend together significantly influence online purchase decisions.
The hypothesis test results demonstrate that Sharia Compliance, trust, usability, and trend collectively have a significant impact on online purchase decisions.This is indicated by the obtained F-test result, with an F-value of 43.294 (F-value > F-table 2.47) and a significance level of 0.000 (sig < 0.05).This indicates that there is a positive and significant influence of the variables Sharia Compliance, trust, usability, and trend together on online purchase decisions.The variable trust (X2) related to online purchases has the highest influence on purchase decisions compared to the Sharia Compliance variable (X1), usability (X3), and the trend variable (X4).

D. Conclusion
Based on results data analysis and discussion _ done about influence mix 7P's of marketing viz product (product), price (price), location (place), promotion (promotion), people (people), evidence physical (physical evidence), and process towards decision purchase consumers at Budiman Swalayan Jembes , then can concluded of the 7 (seven) variables studied, 3 (three) of them own influence to decision purchase .3 (three) influencing variables decision purchase Budiman Swalayan consumers Jembes This is variable product, variable price and variable location.Whereas for 4 (four) variables other that is variable promotion, person variable, variable proof physical and process variables after done analysis show No exists influence to decision purchase consumers at Budiman Swalayan Jembes.
2013 and 2015 Budiman Swalayan established two branches in Padang City.Then in 2017 Budiman Swalayan established two more branches, namely in Padang City and in By Pass Pulai Anak Air, Bukittinggi City, which was previously Budiman Swalayan which was located on Jalan Batang Agam, Behind Balok, Bukittinggi City from 2007 to 2017.In 2018 Budiman Swalayan established two more branches located in Padang City and Payakumbuh City.In 2019 Budiman Swalayan opened another branch in Padang City, then in 2020 Budiman Swalayan opened three branches in three regions at once, namely in Padang City, Agam Regency and Payakumbuh City.In 2022 Budiman Swalayan will again open one of its branches in Bukittinggi City and three other branches in Padang City.To date, Budiman Swalayan has a total of 16 branches, 4 of which are located in Bukittinggi City.

Table 1 .
List of Years of Establishment of Supermarkets and Total Branches up to 2023

Table 3 .
Second Outer Loading Value Indicators that have the outer loading value is less of 0.6 is eliminated from the model.Outer loading value after PD1, LK2, BF3, PS3, and PS4 are eliminated can seen in Table 4 below This :

Table 5 .
AVE Value Results

Table 6 .
Cross Loading Results

Table 8 .
Results of R -s quare values

Table 9
Based onTable 9 Q -s quare value obtained big from 0, p This show that the model has nil values good observation ._ Test result hypothesis in study This can seen in Table 10 and for research model results This can depicted as in Figure 1 following this :

Table 10 .
Testing Hypothesis summary of the results of hypothesis testing can be seen in Table11 below: A