 |
Vol. 03, No. 12, December 2023
e - ISSN : 2807-8691 | p- ISSN : 2807-839X
Master of Management, Faculty Of Economics And Business, Universitas Pelita Harapan, Indonesia1,2 Email: [email protected]1, [email protected]2
ABSTRACT
Keywords
Social Media Marketing, Electronic
Word Of Mouth, Brand Awareness, Trust, Purchase Decision.
With the evolution of the times, there has been a rise in the
number of internet and social media users, leading to a shift from traditional marketing methods to digital marketing. The aim of
this research was to explore how social media marketing and
electronic word-of-mouth (e-WOM) affect the choices made in purchasing, with brand awareness and trust serving as mediating factors. Employing a quantitative approach, an online survey was conducted with 200 participants, and data processing was carried out using SmartPLS software. The findings revealed a notable and meaningful connection was observed, showing a positive influence among social media marketing, e-WOM, and purchasing decisions, wherein brand awareness and trust played intermediary roles., specifically in "X" ice cream outlets. Notably, brand awareness and trust were identified to have a partially mediating effect concerning the connection among social media marketing, e-WOM, and purchasing decisions, this study underscores the managerial significance of improving social media marketing, e-WOM, trust, and brand awareness to
positively impact purchase choices.
In the era of digitalization, human life cannot be separated from social media. Humans are social creatures that require interaction between individuals. Entering this digital era, interactions previously built directly are starting to switch to social media (Ansari et al. 2019). The development of this technology has dramatically facilitated human life for various activities, including business activities. The impact of this technology is the emergence of different creative and innovative ideas that can increase a business's competitiveness (Handi et al. 2018).
Based on the We Are Social survey, it is noted that Internet users in Indonesia will continue to increase from 2012 until 2023. In January 2023, the number of internet users in Indonesia will reach
212.9 million (Rizaty 2023; Clinten dan Kusuma 2023).
The increase in the number of users every year in Indonesia will undoubtedly change human lifestyles along with the times. Conventional marketing activities began to shift into digital marketing activities. Companies must adapt their marketing by switching to digital marketing activities (Upadana dan Pramudana 2020).
The emergence of the COVID-19 pandemic amid the digitalization era has various positive and negative impacts on the Indonesian people . With large-scale social restrictions, people cannot gather, gather, and carry out activities on a large scale. This impacts marketing methods that need to adapt to the times. Various companies innovate to develop creative ideas using social media for marketing needs (Saputra 2020).
IJSSR Page 3137
https://doi.org/10.46799/ijssr.v3i12.614
This work is licensed under a Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)
We Are Social research indicates approximately 167 million social media users in Indonesia. This represents 60.4% of Indonesia's population (Widi 2023). Every individual in Indonesia is estimated to spend 7 hours 42 minutes playing social media daily, the eleventh highest in the world (Bearman et al. 2023).
According to We Are Social, researching products and brands is one of the main reasons humans access the Internet. The survey shows that 43.4% of humans access the internet to find a product or brand, ranking seventh as the main reason humans access the internet (We Are Social, 2023). Based on this data, using social media to conduct marketing activities is the right choice.
Based on a survey conducted by Sea Insights, 54% of MSME entrepreneurs use social media as a marketing medium. The use of various social media as marketing media is a form of adaptation for business actors during the COVID-19 pandemic (Alika, 2020). Social media marketing is defined as a process carried out by a marketer to promote products and services provided through a social media network. The use of marketing through social media is not only carried out by a group of individual business actors but also carried out by industrial businesses (Ardiansyah dan Sarwoko 2020).
Previous research discovered that brand recognition and purchasing decisions positively and significantly correlate with social media marketing (Ardiansyah dan Sarwoko 2020; Upadana dan Pramudana 2020; Kodrat 2021). Brand awareness is a state in which consumers can recognize a brand that is part of a product category. In terms of increasing brand awareness, many business actors use social media to increase brand awareness of the products or services offered (Ardiansyah dan Sarwoko 2020).
According to research conducted by Pramudana and Upadana (2020), purchase intention is influenced by brand awareness. This is supported by Ardiyansyah and Sarwoko (2020), who revealed that a product or service with a higher brand awareness will increase the possibility of consumers purchasing. A Purchase decision is a process that an individual can carry out to determine a product or service through a series of evaluation processes (Upadana dan Pramudana 2020).
According to Budiatmo and Kurniasari (2018), purchase decision is a process carried out by a consumer in choosing products and services. In making decisions, there are six stages carried out by consumers, specifically: 1. Identification of the need; 2. Information seeking; 3. Alternative assessment;
4. Purchase choice; 5. Consumption; and 6. Post-purchase actions (Kodrat, 2021).
According to research by Angelyn and Kodrat (2021), brand knowledge impacts how social media marketing influences decisions about what to buy. When a company applies social media marketing, brand awareness increases, resulting in growth in purchase decisions. This research shows that brand awareness needs to be increased to the "top of mind" level by applying social media marketing to improve purchase decisions.
Another aspect that affects purchase intention is e-WOM. Based on previous research conducted by Moniharapon et al., (2022) and Handi et al., (2018), e-WOM positively and significantly influences purchase decisions. Electronic word of mouth (e-WOM) is defined as marketing activities using internet technology to spread word of mouth, which has the purpose of marketing media. In marketing through e-WOM, marketing is focused on viral marketing because viral marketing is contagious like a virus. In this marketing concept, consumers are encouraged to tell their experiences using a product or a service through word of mouth through online media (Sekaran dan Bougie 2016; Sulthana dan Vasantha 2019).
In research conducted by Park &; Seo, (2018), it is said that customers will look for information in the form of e-WOM through social media to reduce worry before making a purchase decision. Consumers will rely on trust derived from information obtained from e-WOM through social media regarding the experience of other consumers who have made product purchases first. Brand awareness is also another aspect that influences the relationship between eWOM and purchase decisions (Chatzipanagiotou et al. 2023; Kumar et al. 2023; S dan Chandra 2023). As a result of consumer trust in a brand, consumers will recognize one particular brand as brand awareness (Bastos dan Moore 2021; Huang 2022).
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2 Civelek also conducted other research &; Ertemel, (2018) In this research, it was found that
peer-to-peer interaction was built through e-WOM. Positive information obtained from e-WOM will increase brand awareness and will have a direct effect on trust. Purchase decisions will increase with the influence of positive information obtained from e-WOM.
In this modern era, many companies rely on social media marketing and e-WOM as marketing media to increase brand awareness, trust, and purchase decisions. This is also done by one of the "X" ice cream outlets that make social media marketing and e-WOM a marketing tool and strategy to increase purchase decisions. In Jakarta, there are many ice cream outlets, but "X" ice cream outlets mushroomed quickly and became one of the favorite and sought-after by consumers when they want to consume ice cream.
This study uses an explanatory research method to explain the relationship and influence between social media marketing (SMM) and electronic word of mouth (e-WOM) on purchase decisions at "X" ice cream outlets in the West Jakarta area. The object of the study was a customer of ice cream outlet "X" who actively used social media and knew the brand. The study was open to all genders, ages, and occupation types. The study population was all customers of "X" ice cream outlets that met these criteria.
The sample selection used the purposive sampling method, where respondents were selected based on specific criteria , namely individuals who use social media and know the "X" ice cream outlet. The number of samples is determined based on a range of numbers 5-10 with several indicators of 16, so a minimum sample number of 160 respondents is obtained.
Data collection is carried out through primary and secondary methods. Preliminary data was obtained through surveys by distributing questionnaires to respondents through social media, WhatsApp applications, and Instagram. Secondary data are obtained from scientific journals, books, survey data, and news relevant to the research topic (Bambale 2014; Suliyanto 2018).
The data analysis method uses SmartPLS software (Muhson 2022). According to research by Angelyn and Kodrat (2021), brand knowledge impacts how social media marketing influences decisions about what to buy.
Before actual testing, preliminary tests were conducted on 30 respondents to check the validity and reliability of the research instrument. The initial test results showed that all indicators and variables had AVE values above 0.50, outer loadings values above 0.70, and Cronbach's alpha and composite reliability values above 0.70. Therefore, all questions in the preliminary test will be used in the actual test.
Thus, this study uses the explanatory research method with the object of research on customers of ice cream outlet "X" in West Jakarta. Data was collected through a survey with a sample of 160 respondents, and data analysis was carried out with the help of SmartPLS software. The validity and reliability of the research instruments have been tested in preliminary tests to ensure the quality of the data obtained.
Based on the data obtained, of the 177 respondents who participated in this study, there were 93 female and 84 male respondents. In percentage, 53% of female and 47% of male respondents actively use social media and know there is an "X" ice cream outlet in the West Jakarta area.
Gender Profile of Research Respondents
Women; 53%;
53%
Man; 47%;
47%
Man
Women
Laki-laki Perempuan
Source: Results of Data Processing and Questionnaire
In this study, respondents will be grouped into age ranges: under 18 years old, 19-40 years old, 41-59 years old, and over 60 years old. Based on the data obtained, of the 177 respondents who participated in this study, there were 27 (15%) respondents aged under 18 years, 102 (58%) respondents aged 19-40 years, 34 (19%) respondents aged 41-59 years, and
14 (8%) respondents over 60 years.
Profile of Age Range of Research Respondents
60 Years and
over; 8%; 8%
!8 years and
above; 15%;
15%
41-59 years
old; 19%; 19%
19-40 years
old; 58%; 58%
18 years and above 19-40 years 41-59 years 18 years and above
18 tahun kebawah
19-40 tahun
41-59 tahun
60 tahun keatas
Source: Results of Data Processing and Questionnaire
The respondents involved in this study were spread across West Jakarta, North Jakarta, South Jakarta, Central Jakarta, East Jakarta, and Tangerang. Based on data obtained from 177 respondents who participated in this study, there were 88 (50%) respondents domiciled in West Jakarta, 26 (15%) respondents domiciled in North Jakarta, 21 (12%) respondents domiciled in
South Jakarta, 12 (7%) domiciled in Central Jakarta, 6 (3%) domiciled in East Jakarta, and 24 (14%) domiciled in Tangerang.
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2
Profile of City of Residence of Research Respondents
Tangerang; 14%
East jakarta 3%
Central Jakarta; 7%
West Jakarta; 50%
South Jakarta; 12%
Nort Jakarta; 15%
West Jakarta Nort Jakarta South Jakarta
Central Jakarta East Jakarta Tanggerang
Jakarta Barat Jakarta Utara Jakarta Selatan
Jakarta Pusat Jakarta Timur Tangerang
Source: Results of Data Processing and Questionnaire
Descriptive statistics is a statistical analysis that will describe the characteristics of each research variable in general. Descriptive statistics can be seen in the minimum, maximum, and mean values.
In this study, the Likert Scale was used to make it easier for respondents to provide answers.
The Likert scale used is as follows.
Strongly disagree
Disagree
Neutral
Agree
Totally Agree
From the data obtained using the Likert Scale, categorization can be carried out using the interval scale as follows.
Interval scale = (Highest score - Lowest score) / Number of scales Interval scale = (5-1) / 5
Interval scale = 0.8
Based on the calculation above, respondents' answers can be categorized according to Table
2.
1,00 - 1,80 Strongly disagree
1,81 - 2,60 Disagree
2,61 - 3,40 Neutral
3,41 - 4,20 Agree
4,21 - 5,00 Totally agree
In Table 3. The results of data processing in the form of minimum, maximum and mean values for social media marketing variables will be displayed as follows.
Indicators | Statement | Min | Max | Mean | Category | |||
SMM1 | When accessing "Instagram" | the social | 1 | 5 | 2,90 | Neutral |
media of ice cream outlet "X", I interact with other individuals.
SMM2 When accessing the
"Instragram" social media of ice cream outlet "X", I was interested in sharing it with other individuals.
SMM3 When accessing social
media "Instagram ice cream outlet "X", I feel the content shared can be entertaining.
1 5 3,32 Neutral
1 5 3,43 Agree
Source: Results of Data Processing and Questionnaire
Based on Table 3, in the social media marketing variable, respondents tend to give neutral answers for SMM1 and SMM2 indicators, and give affirmative answers for SMM3 indicators. Based on these three indicators, a total average value of 3.22 was categorized in the neutral category with a minimum value of 1 which is strongly disagree and a maximum value of 5 which is strongly agree.
Based on the total average value obtained, it can be said that there is still a need to increase the use of social media as a marketing tool used by "X" ice cream outlets, especially for "Instagram" social media. Based on Table 4.3, respondents have been quite entertained by the content shared by the Instagram account of ice cream outlet "X" but not all respondents feel interested in interacting on the ice cream shop's social media account and sharing content on the Instagram account of ice cream outlet "X".
In Table 4. The results of data processing will be displayed in the form of minimum, maximum and mean values for electronic word of mouth variables, which are as follows.
Indicators | Statement | Min | Max | Mean | Category |
EWOM1 | I am interested in | 1 | 5 | 3,38 | Neutral |
posting positive | |||||
comments about the "X" | |||||
ice cream parlor through social media. | |||||
EWOM2 | I am interested in | 1 | 5 | 3,54 | Agree |
recommending to buy | |||||
ice cream from ice | |||||
cream outlet "X" through social media. | |||||
EWOM3 | I am interested in | 1 | 5 | 3,54 | Agree |
recommending to buy |
ice cream from ice cream outlet "X"
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2
through social media to my colleagues
Source: Results of Data Processing and Questionnaire
Based on Table 4, on the electronic word of mouth variable, the average respondent gave a neutral answer to the EWOM1 indicator and agreed to the EWOM2 and EWOM3 indicators. Based on these three indicators, a total average value of 3.49 was obtained which was categorized in the agree category, Having a scale ranging from 1, indicating strong disagreement, to 5, denoting strong agreement..
Based on the total average value obtained, it can be said that ice cream outlet "X" has been quite good at building electronic word of mouth from consumers through social media owned by ice cream outlet "X". Based on Table 4, it can be seen that respondents have felt interested in recommending ice cream outlet "X" through social media and also recommend it to colleagues but not all respondents have been interested in giving positive posts related to ice cream outlet "X" through social media accounts.
In Table 5, the results of data processing in the form of minimum, maximum and mean values for brand awareness variables will be displayed as follows.
Indicators | Statement | Min | Max | Mean | Category |
BA1 | When craving ice cream, | 1 | 5 | 3,84 | Agree |
the brand "X" ice cream | |||||
BA2 | crossed my mind. I used the Instagram | 1 | 5 | 3,49 | Agree |
app to get more | |||||
information about "X" | |||||
ice cream parlors. | |||||
BA3 | I can recognize the "X" ice cream logo. | 1 | 5 | 4,50 | Totally agree |
BA4 | I can remember the "X" | 1 | 5 | 4,46 | Totally agree |
ice cream logo. | |||||
Total Average 4,07 | Agree | ||||
Source: Results of Data Processing and Questionnaire
Based on Table 5, on the average brand awareness variable, respondents answered in agreement for indicators BA1 and BA2, and answered firmly in agreement for indicators BA3 and BA4. Based on these four indicators, a total average value of 4.07 was obtained, categorized in the agreed category, with a minimum value of 1, strongly disagree, and a maximum value of 5, strongly agree.
Based on the total average value, it can be said that ice cream outlet "X" already has a pretty good brand awareness because consumers can remember and recognize the logo and have a sense of wanting to consume ice cream "X" when they want ice cream. Based on Table 5, respondents can be said to have significantly recognized and remembered the ice cream “X” logo. Based on the data in Table 5, respondents are interested in consuming ice cream from ice cream outlets "X" and accessing Instagram social media accounts to get more information from ice cream outlets "X".
In Table 6, the data processing results in the form of minimum, maximum and mean
values for brand awareness variables are displayed as follows.
Indicators | Statement | Min | Max | Mean | Category |
T1 | Ice cream stand "X" | 2 | 5 | 4,08 | Agree |
celebrated. | |||||
T2 | I think the "X" ice cream | 1 | 5 | 3,88 | Agree |
stand doesn't hide any | |||||
important information I need to know. | |||||
T3 | Ice cream stall "X" kept | 2 | 5 | 3,84 | Agree |
its promise. | |||||
Total Average 3,93 | Agree | ||||
Source: Results of Data Processing and Questionnaire
Based on Table 6, all respondents tend to give affirmative answers in indicators T1, T2, and T3 in the trust variable. Based on the three indicators, a total average value of 3.93 was obtained, categorized in the category of agree with a minimum value of 1, strongly disagree, and a maximum value of 5, strongly agree.
Based on Table 6, it can be said that ice cream outlet "X" has gained a sense of trust from consumers or has fostered a sense of trust in consumers. Based on Table 6, consumers have instilled confidence in the "X" ice cream outlet by not hiding important information that consumers need to know and can keep their promises.
In Table 7. The data processing results in minimum, maximum, and mean values for purchase decision variables are displayed as follows.
Indicators | Statement | Min | Max | Mean | Category | |
WW1 | In choosing | a product, | 1 | 5 | 4,34 | Totally agree |
information product | about the becomes |
important to me.
WW2 In choosing a product, I will choose the best alternative in my opinion.
1 5 4,40 Totally agree
WW3 | I use social media to information about product. | get a | 1 | 5 | 4,30 | Totally agree |
Total Average | 4.35 | Totally agree |
Source: Results of Data Processing and Questionnaire
Based on Table 7, the purchase decision variable tends to give agreeable answers in PD1, PD2, and PD3 indicators. Based on these three indicators, a total average value of 4.35 was obtained, categorized in the category of strongly agree with a minimum value of 1, strongly disagree, and a maximum value of 5, strongly agree.
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2 Based on Table 7. Consumers strongly agree that product information and the best product choices can be considered in choosing products. Based on Table 7, social media is also
essential in providing consumers with information about a product.
In this study, inferential analysis will be carried out, which will be used to test the outer and inner models of this study. The inferential analysis in this study will use SmartPLS version 3 devices. This research will use questionnaire-filling data in Google Forms as primary data. Questionnaires used in actual research are questionnaires that have passed preliminary tests that are valid and reliable.
In this study, the outer model analysis used is validity testing and reliability testing, which includes validity indicators, convergent validity, construct reliability (Cronbach's alpha and composite reliability), and Forner-lacker criteria—the following results from calculating the outer model of the research path analysis.
Latent Variable | Indicator | Loading (>0.70) | AVE(>0.5) |
Brand Awareness | BA1 | 0.853 | 0.670 |
BA2 | 0.769 | ||
BA3 | 0.808 | ||
BA4 | 0.841 | ||
Electronic word of mouth | EWOM1 | 0.897 | 0.860 |
EWOM2 | 0.953 | ||
EWOM3 | 0.930 | ||
Purchase Decision | WW1 | 0.903 | 0.789 |
WW2 | 0.886 | ||
WW3 | 0.875 | ||
Social media marketing | SMM1 | 0.919 | 0.854 |
SMM2 | 0.945 | ||
SMM3 | 0.908 | ||
Trust | T1 | 0.861 | 0.812 |
T2 | 0.934 | ||
T3 | 0.907 |
Source: Data Processing Results
Latent Variable | Cronbach's Alpha | Composite Reliability |
Brand Awareness | 0.836 | 0.890 |
Electronic word of mouth | 0.918 | 0.948 |
Purchase Decision | 0.866 | 0.918 |
Social media marketing | 0.915 | 0.946 |
Trust | 0.885 | 0.928 |
Source: Data Processing Results
Latent variables | BA | EWOM | PD | SMM | T |
Brand Awareness | 0.818 | ||||
Electronic word of mouth | 0.373 | 0.927 |
Purchase Decision | 0.583 | 0.440 | 0.888 | ||||
Social media marketing | 0.363 | 0.061 | 0.406 | 0.924 | |||
Trust | -0.065 | 0.246 | 0.178 | -0.080 | 0.901 | ||
Source: Data Processing Results
Based on Table 8, all indicators of each latent variable have values above 0.70 and have AVE values above 0.50. Based on the results obtained from Table 8. So, all indicators of each indicator have met the validity criteria and can be said to be valid. According to Hair et al. (2010), validity indicators can be measured based on outer loading values and average values (AVE), valid if they have outer loading values above 0.70 and AVE values above 0.50. As long as the indicator's AVE value is more significant than 0.50 and its outer loading value is less than 0.70, it can still be considered valid.
Table 9 shows that all indicators have composite reliability values above 0.70 and Cronbach's Alpha values above 0.70. Considering the outcomes shown in Table 9., the construct reliability computation results (Cronbach's Alpha and Composite reliability values) can be said to have met the reliability testing criteria, or it can be said that variable constructs can be declared reliable. According to Hair et al., (2014), a variable construct can be reliable if it meets the criteria, namely having a composite reliability value above 0.70 and a Cronbach's alpha value above 0.70.
Table 10 demonstrates no multicollinearity issue between latent variables and that each variable in the study has a discriminant validation value that satisfies the Fornell-Lacker criteria. Hair et al. (2010) state that to satisfy the Fornell-Lacker criteria for discriminant validity testing, the square root of each construct's AVE value must be greater than the correlation value with other constructs.
In this study, the inner model testing carried out was collinearity testing, coefficient of determination (R2), and effect size (f2). The following results were obtained based on testing using SmartPLS software version 3.
Latent Variables | Brand Awareness | Purchase Decision | Trust |
Brand Awareness | 1.374 | ||
Electronic word of mouth | 1.004 | 1.277 | 1.000 |
Social media marketing | 1.004 | 1.162 | |
Trust | 1.099 |
Source: Data Processing Results
Latent Variables | R Square | R Square Adjusted |
Brand Awareness | 0.256 | 0.247 |
Purchase Decision | 0.476 | 0.464 |
Trust | 0.061 | 0.055 |
Source: Data Processing Results
Latent Variables | Brand Awareness | Purchase Decision | Trust |
Brand Awareness | 0.241 | ||
Electronic word of mouth | 0.166 | 0.077 | 0.064 |
Social media marketing | 0.156 | 0.106 |
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2
Trust 0.050
Source: Data Processing Results
Table 11. indicates that all latent variables in the study have VIF values less than 5.0. Therefore, based on the findings, it is possible to conclude that none of the study's variables exhibit multicollinearity symptoms and can proceed with additional analysis.. According to Hair et al. (2014), variables that do not have symptoms of multicollinearity have a VIF value smaller than 5.0.
Table 12 shows that the R-value for the brand awareness variable is 0.256. While other factors not included in the research model influence the remaining 74.4% of brand awareness, it can be concluded that social media marketing and electronic word-of-mouth moderately impact brand awareness, accounting for 25.6% of the total. The decision regarding the variable purchase has an R2 value of 0.476. The research model indicates that social media marketing, electronic word-of-mouth, brand awareness, and trust moderately influence purchase decisions, accounting for 47.6%. The remaining 52.4% of the decisions are influenced by factors not included in the model.
Table 12 indicates the value of R2. Electronic word-of-mouth has a 6.1% influence on the trust variable (0.061), indicating a low influence. The remaining 93.9% of the trust variable is determined by factors not included in the research model. The coefficient of determination (R2) is measured, per Hair et al. (2010), in order to gauge how accurate an estimate is. Large estimation accuracy values are found in variables with R values of 2.75, moderate estimation accuracy values in variables with R values of 0.50, and small estimation accuracy values in variables with R values of 2.25.
Based on Table 13, the value of f2 for variable construct models in this study can be known. For the variable construct model, brand awareness can affect purchase decisions by
0.241 and can be said to have a moderate estimation value. The electronic word-of-mouth construct model can affect brand awareness by 0.166 and can be said to have a moderate estimation value. The electronic word-of-mouth construct model can affect the purchase decision by 0.077 and can be said to have a small estimation value. For the constructed model, the electronic word-of-mouth variable can affect trust by 0.064 and can be said to have a small estimation value.
Based on Table 13, the value of f2 For the social media marketing construct model for brand awareness is 0.156, which is relatively low. The social media marketing construct model can affect the purchase decision by 0.106, which can be said to have a small estimation value, and for the trust variable construct model affects the purchase decision variable by 0.050 and can be said to have a small estimation value. According to Hair et al., (2010), the value of f2 0.02 has a small effect size value, F value2 0.15 has a medium effect size value and an F value2 0.35 has a large effect size value.
Based on the results of data obtained in the outer model and inner model analysis, the analysis can be continued in hypothesis testing using the bootstrapping test. The following results were obtained based on the hypothesis test that was carried out using SmartPLS Version 3 software.
Hypothesis Path Coefficient Original Sample
T Statistics P Values
Information
H1 Social media marketing -> Brand Awarenesss
0.341 4.504 0.000 Supported
H2 | Social media marketing -> | 0.254 | 5.451 | 0.000 | Supported | ||
Purchase Decision | |||||||
H3 | Brand Awareness -> | 0.416 | 7.154 | 0.000 | Supported | ||
H5 | Purchase Decision Electronic word of mouth - | 0.353 | 5.322 | 0.000 | Supported | ||
> Brand Awareness | |||||||
H6 | Electronic word of mouth - | 0.228 | 3.440 | 0.001 | Supported | ||
> Purchase Decision | |||||||
H8 | Electronic word of mouth - > Trust | 0.246 | 3.173 | 0.002 | Supported | ||
H9 | Trust -> Purchase | 0.169 | 3.161 | 0.002 | Supported | ||
Decision |
Source: Data Processing Results
Hypothesis | Path Coefficient | Original Sample | T Statistics | P Values | Information |
H4 | Social media marketing -> | 0.142 | 3.817 | 0.000 | Supported |
Brand Awareness -> | |||||
H7 | Purchase Decision Electronic word of mouth | 0.147 | 4.060 | 0.000 | Supported |
-> Brand Awareness -> | |||||
Purchase Decision | |||||
H10 | Electronic word of mouth | 0.042 | 2.027 | 0.043 | Supported |
-> Trust -> Purchase Decision |
Source: Data Processing Results
Based on Hair et al. (2010), structural model coefficient analysis is carried out to analyze a hypothesis by analyzing the relationship between which variables have a significant relationship. A relationship is declared to have a significant effect if it has a p-value smaller than 0.05, if the relationship between variables has a p-value greater than 0.05 then it can be said that the relationship has no significant effect.
Based on Table 14. and Table 15. It can be seen that all hypotheses in this study have a p-value smaller than 0.05, so all hypotheses in this study can be supported, both in direct influence and also in indirect influence.
It is evident from Table 14's hypothesis test results that hypothesis 1 has a t-statistic value of 4.504, which is higher than the t-table value of 1.645. Additionally, Table 14 shows that hypothesis 1 has a p-value of 0.000, which is less than 0.05, suggesting that the hypothesis can be supported.
According to Table 14. Furthermore, hypothesis 1's positive path coefficient value of 0.341 indicates that social media marketing significantly and favorably affects brand awareness. Therefore, it can be concluded that the "X" ice cream outlet's brand awareness increases with the quality of its social media marketing.
The results obtained from this study can strengthen the results obtained in previous studies conducted by Kodrat and Angelyn (2021), Ardiyansyah and Sarwoko (2020), and Pramudana and Upadana (2020) which stated that social media marketing has a positive influence on brand awareness. Based on the results obtained, it can be concluded that hypothesis 1 can be supported.
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2 Kodrat and Angelyn (2021) conducted research at Haroo Table, which is an online-based bakery located in the city of Surabaya, the respondents involved in this study were 88 respondents and used purposive sampling and Slovin methods. Data analysis was carried out using SEM-PLS and primary data was obtained using questionnaires and Likert scales. In contrast, secondary data was obtained from
company data, competitor data, and related literature.
Ardiyansyah and Sarwoko (2020) conducted research on sportswear products under the SeaGods brand produced in the city of Bali. The respondents involved in this study were 100, using the purposive sampling method, and data analysis was carried out using ordinary least square regression. Primary data were obtained using questionnaires and Likert scales, while secondary data were obtained from company data and related literature.
Pramudana and Upadana (2020), researched Starbucks Coffee company located in Denpasar. The number of respondents involved in this study was 110 respondents using the purposive sampling method. The analysis method used in this study was multivariate. Data collection was carried out using the distribution of questionnaires through Google Forms conducted in Denpasar City using the Likert scale.
Analyzing the outcomes of the hypothesis test provided in Table 14, it becomes evident that hypothesis 2 exhibits a t-statistic value of 7.154, surpassing the t-table value of 1.645. Additionally, Table 14 reveals that hypothesis 3 yields a p-value of 0.000, indicating it is less than 0.05. Therefore, hypothesis 2 can be supported.
It is based on Table 14. It can also be seen that hypothesis 1 has a favorable path coefficient value of 0.254, so effective social media marketing positively and significantly influences purchasing decisions. Hence, it can be asserted that the more proficient the social media marketing efforts of the "X" ice cream shop, the more purchase decisions of the "X" ice cream outlet will increase.
The results obtained from this study can strengthen the results obtained in previous studies that have been conducted by Kodrat and Angelyn (2021), Ardiyansyah and Sarwoko (2020), and Pramudana and Upadana (2020), which stated that social media marketing has a positive influence on purchase decisions. Based on the results obtained, it can be concluded that hypothesis 2 can be supported.
Kodrat and Angelyn (2021) conducted research at Haroo Table, which is an online-based bakery located in the city of Surabaya. The respondents involved in this study were 88, and purposive sampling and Slovin methods were used. Data analysis was carried out using SEM-PLS, and primary data was obtained using questionnaires and Likert scales. In contrast, secondary data was obtained from company data, competitor data, and related literature.
Ardiyansyah and Sarwoko (2020) conducted research on sportswear products under the SeaGods brand produced in the city of Bali. The respondents involved in this study were 100, using the purposive sampling method, and data analysis was carried out using ordinary least square regression. Data collection was carried out using questionnaires and Likert scales.
Pramudana and Upadana (2020), researched Starbucks Coffee company located in Denpasar. The number of respondents involved in this study was 110 respondents using the purposive sampling method. The analysis method used in this study is multivariate. Data collection was carried out using the distribution of questionnaires through Google Forms conducted in Denpasar City using the Likert scale.
Based on the results of the hypothesis test contained in Table 14., it can be seen that hypothesis 3 has a t-statistic value of 5.451 which is greater than the t-table value of 1.645. In Table 14. It can also
be seen that hypothesis 3 has a p-value of 0.000 which is smaller than 0.05, so hypothesis 3 can be supported.
It is based on Table 14. It can also be seen that hypothesis 3 has a favorable path coefficient value of 0.416, so brand awareness has a positive and significant effect on purchase decisions. Thus, it can be said that the better the brand awareness owned by the ice cream shop "X", the purchase decision of the ice cream outlet "X" will also increase.
The results obtained from this study can strengthen the results obtained in previous studies that have been conducted by Kodrat and Angelyn (2021), Pramudana and Upadana (2020), and Budiatmo and Kurniasari (2018), which stated that brand awareness has a positive influence on purchase decisions. Based on the results obtained, it can be concluded that hypothesis 3 can be supported.
Kodrat and Angelyn (2021) conducted research at Haroo Table, which is an online-based bakery located in the city of Surabaya. The respondents involved in this study were 88, and purposive sampling and Slovin methods were used. Data analysis was carried out using SEM-PLS, and primary data was obtained using questionnaires and Likert scales. In contrast, secondary data was obtained from company data, competitor data, and related literature.
Pramudana and Upadana (2020), conducted research on Starbucks Coffee located in the city of Denpasar. The number of respondents involved in this study was 110, using the purposive sampling method. The analysis method used in this study is multivariate. Data collection was carried out using the distribution of questionnaires through Google Forms conducted in Denpasar City using the Likert scale.
Budiatmo and Kurniasari (2018) researched J.CO Donuts & Coffee in Semarang. The respondents involved in this study were 100, using purposive and accidental sampling methods. Data collection is carried out by distributing questionnaires.
Based on the results of the hypothesis test contained in Table 15., it can be seen that hypothesis 4 has a t-statistic value of 3.817 which is greater than the t-table value of 1.645. In Table 15. It can also be seen that hypothesis 4 has a p-value of 0.000, which is smaller than 0.05, so sis 4 can be supported.
Based on Table 15. It can also be seen that hypothesis 4 has a path coefficient value of 0.142 which is positive, social media marketing has a positive and significant effect on purchase decisions mediated by brand awareness. Thus, it can be said that the better the social media marketing and brand awareness of the "X" ice cream shop, the purchase decision of the "X" ice cream outlet will be.
The results obtained from this study can strengthen the results obtained in previous studies conducted by Kodrat and Angelyn (2021) and Pramudana and Upadana (2020), which stated that brand awareness positively influences purchase decisions mediated by brand awareness. Based on the results obtained, it can be concluded that hypothesis 4 can be supported.
Kodrat and Angelyn (2021) conducted research at Haroo Table, which is an online-based bakery located in the city of Surabaya. The respondents involved in this study were 88, and purposive sampling and Slovin methods were used. Data analysis was carried out using SEM-PLS, and primary data was obtained using questionnaires and Likert scales. In contrast, secondary data was obtained from company data, competitor data, and related literature.
Pramudana and Upadana (2020), conducted research on Starbucks Coffee located in the city of Denpasar. The number of respondents involved in this study was 110, using the purposive sampling method. The analysis method used in this study is multivariate. Data collection was carried out using the distribution of questionnaires through Google Forms conducted in Denpasar City using the Likert scale.
Based on the results of the hypothesis test contained in Table 14., it can be seen that hypothesis 5 has a t-statistic value of 5.322 which is greater than the t-table value of 1.645. In Table 14. It can also
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2 be seen that hypothesis 5 has a p-value of 0.000, more diminutive than 0.05, so hypothesis 5 can be supported.
It is based on Table 14. It can also be seen that hypothesis 5 has a favorable path coefficient value of 0.353, so e-WOM has a positive and significant effect on brand awareness. Thus, it can be said that the better the e-WOM owned by the "X" ice cream shop, the better the brand awareness of the "X" ice cream outlet.
The results obtained from this study can strengthen the results obtained in previous studies conducted by Choi et al. (2020), Handi et al. (2018), and Seo and Park (2018), which stated that e-WOM has a positive influence on brand awareness. Based on the results obtained, it can be concluded that hypothesis 5 can be supported.
Choi et al., (2020) researched the social media accounts of an airline. The respondents involved in this study were 450 respondents, and the amount of data processed in this study was 430 data obtained from questionnaires. Sampling is carried out using a non-probability sampling method, namely random sampling. Data analysis was carried out using SPSS 21.0 and AMOS 20.0 software (Ghozali 2016). Handi et al., (2018) conducted a study on Go-Food application users in Jakarta, The respondents involved in this study were 175 respondents. The sampling method uses non-probability sampling, namely the convenience sampling technique. Data analysis is carried out using SEM and processed using
AMOS software. Data collection was carried out using questionnaire dissemination.
Seo and Park (2018) researched the social media accounts of an airline. The respondents involved in this study were 430 passengers who knew the airline's social media.
Based on the results of the hypothesis test contained in Table 14., it can be seen that hypothesis 6 has a t-statistic value of 3.440, which is greater than the t-table value of 1.645. In Table 14. It can also be seen that hypothesis 6 has a p-value of 0.001 which is smaller than 0.05, so hypothesis 6 can be supported.
It is based on Table 14. It can also be seen that hypothesis 6 has a favorable path coefficient value of 0.228, so e-WOM has a positive and significant effect on purchase decisions. Thus, it can be said that the better the e-WOM owned by the "X" ice cream shop, the purchase decision of the "X" ice cream outlet will increase.
The results obtained from this study can strengthen the results obtained in previous studies conducted by Moniharapon et al. (2022) and Handi et al. (2018), which stated that e-WOM has a positive influence on purchase decisions. Based on the results obtained, it can be concluded that hypothesis 6 can be supported.
Moniharapon et al. (2022) researched Manado Bag Store. The respondents involved in this study were 100 respondents obtained from questionnaires. Sampling is carried out using the Accidental sampling method. Data analysis was carried out using SPSS 25.0 software. Data was collected by distributing questionnaires and using Likert scale data measurements.
Handi et al., (2018) conducted a study on Go-Food application users in Jakarta. The respondents involved in this study were 175 respondents. The sampling method uses non-probability sampling, namely the convenience sampling technique. Data analysis is carried out using SEM and processed using AMOS software. Data collection was carried out using questionnaire dissemination.
Based on the results of the hypothesis test contained in Table 15., it can be seen that hypothesis 7 has a t-statistic value of 4.060 which is greater than the t-table value of 1.645. In Table 15. It can also be seen that hypothesis 7 has a p-value of 0.000 smaller than 0.05, so hypothesis 7 can be supported.
Based on Table 15. It can also be seen that hypothesis 5 has a favorable path coefficient value of 0.147, so e-WOM has a positive and significant effect on purchase decisions mediated by brand
awareness. Thus, it can be said that the better the e-WOM and brand awareness of the "X" ice cream shop, the purchase decision of the "X" ice cream outlet will be.
The results obtained from this study can strengthen the results obtained in previous studies conducted by Kodrat and Angelyn (2021) and Choi et al. (2020), which stated that brand awareness has a positive influence on purchase decisions and that e-WOM has a positive effect on brand awareness. Based on the results obtained, it can be concluded that hypothesis 7 can be supported.
Kodrat and Angelyn (2021) conducted research at Haroo Table, which is an online-based bakery located in the city of Surabaya. The respondents involved in this study were 88, and purposive sampling and Slovin methods were used. Data analysis was carried out using SEM-PLS, and primary data was obtained using questionnaires and Likert scales. In contrast, secondary data was obtained from company data, competitor data, and related literature.
Choi et al. (2020) researched the social media accounts of an airline. The respondents involved in this study were 450 respondents, and the amount of data processed in this study was 430 data obtained from questionnaires. Sampling is carried out using a non-probability sampling method, namely random sampling. Data analysis was carried out using SPSS 21.0 and AMOS 20.0 software.
Based on the results of the hypothesis test contained in Table 14., it can be seen that hypothesis 8 has a t-statistic value of 3.173, which is greater than the t-table value of 1.645. In Table 14. It can also be seen that hypothesis 8 has a p-value of 0.002, more diminutive than 0.05, so hypothesis 8 can be supported.
It is based on Table 14. It can also be seen that hypothesis 8 has a favorable path coefficient value of 0.246, so e-WOM has a positive and significant effect on trust. Thus, it can be said that the better the e-WOM owned by the "X" ice cream shop, the more trust in the "X" ice cream outlet will also increase.
The results obtained from this study can strengthen the results obtained in previous studies that have been conducted by Choi et al. (2020), Civelek and Etermel (2018), and Seo and Park (2018), which stated that e-WOM has a positive influence on trust. Based on the results obtained, it can be concluded that hypothesis 8 can be supported.
Choi et al., (2020) researched an airline's social media accounts. The respondents involved in this study were 450 respondents, and the amount of data processed in this study was 430 data obtained from questionnaires. Sampling is carried out using a non-probability sampling method, namely random sampling. Data analysis was carried out using SPSS 21.0 and AMOS 20.0 software.
Civelek and Etermel., (2018) conducted research conducting customer research on B2C websites. The respondents involved in this study were 400 respondents and the amount of data processed in this study was 305 data obtained from questionnaires. Data analysis is done using SEM, and data processing using SPSS and AMOS software.
Seo and Park (2018) researched the social media accounts of an airline. The respondents involved in this study were 430 passengers who knew the airline's social media.
Based on the results of the hypothesis test contained in Table 14., it can be seen that hypothesis 9 has a t-statistic value of 3.161, which is greater than the t-table value of 1.645. In Table 14. It can also be seen that hypothesis 9 has a p-value of 0.002 smaller than 0.05, so hypothesis 9 can be supported.
It is based on Table 14. It can also be seen that hypothesis 9 has a favorable path coefficient value of 0.169, so trust has a positive and significant effect on purchase decisions. Thus, it can be said that the better the e-WOM owned by the "X" ice cream shop, the purchase decision of the "X" ice cream outlet will increase.
The results obtained from this study can strengthen the results obtained in previous studies conducted by Handi et al. (2018), Civelek and Etermel (2018), and Seo and Park (2018), which stated that trust has a positive influence on purchase decisions. Based on the results obtained, it can be concluded that hypothesis 9 can be supported.
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2 Handi et al., (2018) conducted a study on Go-Food application users in Jakarta. The respondents involved in this study were 175 respondents. The sampling method uses non-probability sampling, namely the convenience sampling technique. Data analysis is carried out using SEM and processed using
AMOS software. Data collection was carried out using questionnaire dissemination.
Civelek and Etermel., (2018) conducted research conducting customer research on B2C websites. The respondents involved in this study were 400 respondents, and the amount of data processed in this study was 305 data obtained from questionnaires. Data analysis is carried out using SEM, and data processing using SPSS and AMOS software.
Seo and Park (2018) researched the social media accounts of an airline. The respondents involved in this study were 430 passengers who knew the airline's social media.
Based on the results of the hypothesis test contained in Table 15. it can be seen that hypothesis 10 has a t-statistic value of 2.027 which is greater than the t-table value of 1.645. In Table 15. It can also be seen that hypothesis 10 has a p-value of 0.043, smaller than 0.05, so hypothesis 10 can be supported.
Based on Table 15. It can also be seen that hypothesis 10 has a favorable path coefficient value of 0.042, so e-WOM has a positive and significant effect on purchase decisions mediated by trusts. Thus, it can be said that the better the e-WOM and trust of the "X" ice cream shop, the better the purchase decision of the "X" ice cream outlet will be.
The results obtained from this study can strengthen the results obtained in previous studies conducted by Choi et al. (2020), Handi et al. (2018), and Civelek and Etermel (2018), which stated that e-WOM has a positive influence on purchase decisions and that e-WOM has a positive effect on trust. Based on the results obtained, it can be concluded that hypothesis 10 can be supported.
Choi et al., (2020) researched the social media accounts of an airline. The respondents involved in this study were 450 respondents, and the amount of data processed in this study was 430 data obtained from questionnaires. Sampling is carried out using a non-probability sampling method, namely random sampling. Data analysis was carried out using SPSS 21.0 and AMOS 20.0 software.
Handi et al., (2018) conducted a study on Go-Food application users in Jakarta. The respondents involved in this study were 175 respondents. The sampling method uses non-probability sampling, namely the convenience sampling technique. Data analysis is carried out using SEM and processed using AMOS software. Data collection was carried out using questionnaire dissemination.
Civelek and Etermel., (2018) conducted research conducting customer research on B2C websites. The respondents involved in this study were 400 respondents and the amount of data processed in this study was 305 data obtained from questionnaires. Data analysis is done using SEM, and data processing using SPSS and AMOS software.
Based on the results of testing that has been carried out in the hypothesis test, it is found that all hypotheses can be supported. Brand awareness and trust partially mediate the influence of social media marketing and e-WOM on the relationship with purchase decision variables. Based on Table 14, path coefficient data can be calculated to determine the magnitude of the influence of mediation variables, which can be seen in Table 16.
Variable | Path Coefficient | Direct Effect | Indirect Effect | |
Brand Awareness | Social media marketing Brand Awareness | -> -> | 0,065 | 0,142 |
Purchase Decision
Brand | Electronic word of mouth -> | 0,052 | 0,147 |
Awareness | Brand Awareness -> | ||
Purchase Decision | |||
Trust | Electronic word of mouth -> Trust -> Purchase Decision | 0,052 | 0,042 |
Source: Data Processing Results
Based on Table 16, one can conclude that brand awareness has a partially mediated nature and can have a more significant influence on purchase decisions than the direct influence of social media marketing and e-WOM. The path coefficient for the brand awareness variable has a more excellent value in the indirect effect.
Table 16 also indicates that the trust variable's path coefficient is smaller in the indirect effect, suggesting that while trust is partially mediated, it cannot substantially influence purchase decisions more than e-WOM's direct influence.
This research aims to assess the effects of social media marketing and electronic word of mouth (e-WOM) on purchasing decisions, considering whether they involve mediating factors such as brand awareness and trust. Employing the PLS-SEM approach, ten hypotheses were tested and analyzed. The findings indicated a positive and significant impact of social media marketing on increasing brand awareness at "X" ice cream outlets. Additionally, social media marketing demonstrated a positive and significant influence on purchasing decisions, underscoring the potential for enhancing consumer purchasing decisions by improving the quality of social media marketing strategies. Moreover, brand awareness was found to positively and significantly affect purchasing decisions, emphasizing the significance of boosting brand awareness for increased consumer appeal. The combined effect of social media marketing and brand awareness was also observed to positively and significantly influence purchasing decisions.
Furthermore, e-WOM was confirmed to elevate brand awareness and contribute positively and significantly to purchasing decisions. Notably, e-WOM was identified to influence trust, which, in turn, positively and significantly impacted purchasing decisions. Additionally, e-WOM was found to influence purchasing decisions through trust mediation. In summary, this study underscores the critical role of social media marketing, brand awareness, e-WOM, and trust in shaping and enhancing purchasing decisions at the ice cream outlet "X."
Alika R. 2020. Survei: 54% UMKM Pakai Media Sosial untuk Pacu Penjualan saat Pandemi. Katadata. co. id. 2.
Ansari S, Ansari G, Ghori MU, Kazi AG. 2019. Impact of brand awareness and social media content marketing on consumer purchase decision. Journal of Public Value and Administrative Insight. 2(2):5–10.doi:10.31580/jpvai.v2i2.896.
Ardiansyah F, Sarwoko E. 2020. How social media marketing influences consumers purchase decision: A mediation analysis of brand awareness. JEMA: Jurnal Ilmiah Bidang Akuntansi Dan Manajemen. 17(2):156–168.
Bambale AJ. 2014. Research methodological techniques as a model for quantitative studies in Social Sciences. British Journal of Economics, Management & Trade. 4(6):862–879.
Bastos W, Moore SG. 2021. Making word-of-mouth impactful: Why consumers react more to WOM about experiential than material purchases. J Bus Res. 130:110–
123.doi:10.1016/J.JBUSRES.2021.03.022.
Regina Raina Rania Citrasumidi1, Lamhot Henry Pasaribu2 Bearman M, Nieminen JH, Ajjawi R. 2023. Designing assessment in a digital world: an organising
framework. Assess Eval High Educ. 48(3):291–304.doi:10.1080/02602938.2022.2069674.
Chatzipanagiotou K, Azer J, Ranaweera C. 2023. E-WOM in the B2B context: Conceptual domain, forms, and implications for research. J Bus Res. 164.doi:10.1016/J.JBUSRES.2023.113957.
Civelek ME, Ertemel AV. 2018. Trust Building Model of Customers on B2c Websites: A Research on Generation Y Customers/B2c Web Sitelerinde Müşterilerin Güven Oluşturma Modeli: Y Jenerasyonu Müşterileri Üzerine Bir Araştırma. Journal of History Culture and Art Research. 7(1):332–340.doi:10.7596/taksad.v7i1.1381.
Clinten B, Kusuma WK. 2023. Pengguna Internet di Indonesia Tembus 212, 9 Juta di Awal 2023. Diakses dari https://tekno. kompas. com/read/2023/02/13/19300087/pengguna-internet-di-indonesia-tembus-212-9-juta-di-awal-2023.
Ghozali I. 2016. Aplikasi analisis multivariete dengan program IBM SPSS 23.
Hair JF, Black WC, Babin BJ, Anderson RE, Tatham RL. 2010. Multivariate data analysis New Jersy: Pearson Education.
Handi H, Hendratono T, Purwanto E, Ihalauw JJOI. 2018. The effect of e-WOM and perceived value on the purchase decision of foods by using the go-food application as mediated by trust. Quality Innovation Prosperity. 22(2):112–127.doi:10.12776/qip.v22i2.1062.
Huang YC. 2022. How marketing strategy, perceived value and brand image influence WOM outcomes— The sharing economy perspective. Journal of Retailing and Consumer Services.
68.doi:10.1016/J.JRETCONSER.2022.103071.
Kodrat DS. 2021. The Effect of Social Media Marketing on Purchase Decision with Brand Awareness as Mediation on Haroo Table.
Kumar S, Prakash G, Gupta B, Cappiello G. 2023. How e-WOM influences consumers’ purchase intention towards private label brands on e-commerce platforms: Investigation through IAM (Information Adoption Model) and ELM (Elaboration Likelihood Model) Models. Technol Forecast Soc Change.
187.doi:10.1016/J.TECHFORE.2022.122199.
Kurniasari M, Budiatmo A. 2018. Pengaruh Social Media Marketing, Brand Awareness Terhadap Keputusan Pembelian Dengan Minat Beli Sebagai Variabel Intervening Pada J. Co Donuts & Coffee Semarang. Jurnal Ilmu Administrasi Bisnis. 7(3):152–159.doi:10.14710/jiab.2018.20968.
Muhson A. 2022. Analisis Statistik dengan SmartPLS: Path Analysis, Confirmatory Factor Analysis, & Structural Equation Modeling.
Rizaty MA. 2023. Pengguna Internet di Indonesia Sentuh 212 Juta pada 2023. Dataindonesia. id, Mar. 2. S B, Chandra B. 2023. The influence of intrinsic and extrinsic motivational factors on e-WOM behaviour: The role of psychological impact during the time of COVID-19 crisis. Heliyon.
9(2).doi:10.1016/J.HELIYON.2023.E13270.
Saputra GW, & AIGAKS. 2020. Pengaruh digital marketing, word of mouth, dan kualitas pelayanan terhadap keputusan pembelian.
Sekaran U, Bougie R. 2016. Research methods for business: A skill building approach. john wiley & sons.
Seo E-J, Park J-W. 2018. A study on the influence of the information characteristics of airline social media on e-wom, brand equity and trust. The Open Transportation Journal. 12(1).doi:10.2174/1874447801812010289.
Seo EJ, Park J-W, Choi YJ. 2020. The effect of social media usage characteristics on e-WOM, trust, and brand equity: Focusing on users of airline social media. Sustainability. 12(4):1691.doi:10.3390/su12041691.
Suliyanto P, & MP. 2018. Metode Penelitian Bisnis untuk Skripsi .
Sulthana AN, Vasantha S. 2019. Influence of electronic word of mouth eWOM on purchase intention.
International Journal of Scientific and Technology Research. 8(10):1–5.
Tuyu V, Moniharapon S, Poluan JG. 2022. The Effect Of Online Advertising And E-Wom On Online Product Purchase Decisions (Study On Consumer Bag Store Manado). Jurnal EMBA: Jurnal Riset Ekonomi, Manajemen, Bisnis dan Akuntansi. 10(3):526–537.doi:10.35794/emba.v10i3.42743.
Upadana MWK, Pramudana KAS. 2020. Brand Awareness Memediasi Pengaruh Social Media Marketing Terhadap Keputusan Pembelian. E-Jurnal Manajemen. 9(5):1921–1941.
Widi S. 2023. Pengguna Media Sosial di Indonesia Sebanyak 167 Juta pada 2023. Retrieved from DataIndonesia. id: https://dataindonesia. id/digital/detail/pengguna-media-sosial-di-indonesia-sebanyak-167-juta-pada-2023.
Regina Raina Rania Citrasumidi, Lamhot Henry Pasaribu (2023)
International Journal of Social Service and Research (IJSSR)
