CC BY
18
DOI 10.18551/rjoas.2020-07.07
THE ROLE OF IMAGE BRANDS MEDIATES E-WOM EFFECTS ON BUYING INTEREST
IN TRITON LEARNING INSTITUTIONS
Asdiana Luh Devi Priyanthi, Yasa Ni Nyoman Kerti*
Master of Management Study Program, Faculty of Economics and Business, University of Udayana, Indonesia *E-mail: [email protected]
ABSTRACT
This study aims to examine the role of brand image mediating the effect of e-WOM on purchase intentions at the Triton Tutoring Institute. The number of respondents in this study was 128 respondents, all respondents were high school students in Denpasar City who had not yet followed the Triton Study Guidance. The sampling technique in this study was purposive sampling. Data from the questionnaires were then analyzed using the Partial Least Square (PLS) analysis technique with the help of the Smart PLS program. Based on the results of the analysis in this study, several conclusions are obtained, namely (1) e-WOM has a positive and significant effect on purchase intention, (2) e-WOM has a positive and significant effect on brand image, (3) brand image has a positive and significant effect on purchase intention and (4) the role of the brand image significantly mediates the effect of e-WOM on purchase intentions. This shows the higher brand image, the higher the influence of e-WOM for students' interest in following Triton's tutoring in Denpasar.
KEY WORDS
E-WOM, brand image, purchase intention.
Tutoring education services are currently in great demand by students because of the application of value standards and the level of competition in improving learning achievement. Many educational institutions have sprung up that provide course services, increasing competition among similar educational institutions in terms of offering the best products to meet the needs of consumers. One of the tutors in Bali is Triton, which has been established since 2010. The number of students who take tutoring in Triton in the past three years has decreased. This proves the purchase intention of students to take tutoring in tutoring institutions decreases. Purchase intention is an important component in consumer buying behavior because it influences consumers to make different choices in the market based on their preferences (Kazmi et al., 2016). If the purchase intention is high, then there is a high chance for the customer to buy the product or service, but the purchase intention depends on various factors such as customer readiness and interest in the product or service, social pressure from others and a general understanding of what the customer feels (Vahdati et al., 2016).
In the technological era, consumers have more interaction with each other and convey information about products to decide whether to buy or not, depending on the arguments of other customers who currently use these types of products. Consumers tend to be interested in writing positive and negative experiences from the experience of buying or using the product on an online site, so when customers are satisfied with a product or service, e-WOM occurs which can help others to choose better products or services as well (Kazmi et .al., 2016). e-WOM is interesting to be highlighted by marketing management because consumers of social media users post, repost, and provide reviews of products or services that they have consumed (Yunus et al., 2016). The likes or dislikes of customers are very dependent on customer online reviews. Purchase intention is influenced by feedback reviews from other consumers (Abubakar et al., 2016). Electronic word of mouth influence brand image positive and significantly, each electronic word of mouth and brand image have a positive and significant effect on brand attitude and there significant role of brand image in
strengthening the influence of electronic word of mouth on brand attitude (Pratiwi and Yasa, 2019).
Research conducted in Cyprus found that e-WOM results had a positive and significant effect on purchase intentions (Abubakar et al., 2016). Likewise, research conducted on students in Rawalpindi City and Islamabad, e-WOM has a positive and significant effect on purchase intention (Tariq et al., 2017). Research conducted in the City of Samarahan, Malaysia, e-WOM has a positive and not significant relationship to consumer purchase intentions (Shahrinaz et al., 2016). Research conducted by Chin et al. 2018 e-WOM has a positive and not significant relationship to consumer purchase intentions. This means customers don't refer to online reviews when they have the intention to use Grab.
There is an e-WOM research gap on purchase intentions. Looking at the existing research gap, it is necessary to add a mediating variable that is suitable to influence e-WOM towards purchase intention is the brand image. This is because e-WOM is getting more positive, so being able to strengthen brand image is also supported by empirical studies on research conducted by Jalilvand et al. 2012, showed that e-WOM had a positive and significant impact on brand image and brand image significantly mediated the effect of e-WOM on purchase intentions. Research conducted by Suyoga et al. 2018 which shows a positive and significant brand image effect on purchase intentions and can mediate the influence of e-WOM on purchase intentions. Based on the story above, this study aims to explain the role of brand image mediating the effect of e-WOM on purchase intentions.
LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT
E-WOM is a way of communication that provides information to consumers about products, how to use products and services perceived by consumers through internet-based technology (Torlak et al., 2014). Consumers have the opportunity to make reviews for potential consumers by sharing their opinions through blogs, chat rooms, and social networking websites (Panjakajornsak, 2014). Users on social media like Facebook can easily reach out and get information about products or services. When customers are satisfied with a product or service, it encourages e-WOM to help others choose better products or services (Kazmi et al., 2016). WOMMA (2012) states that 90% of small businesses consider Word of Mouth very important as one of the marketing tools, and 59% say it is a driving force in their purchasing decisions based on e-Word of Mouth.
Research conducted in Cyprus found that e-WOM results had a positive and significant effect on purchase intentions (Abubakar et al., 2016). Research conducted on smartphone brands in the North Sirpus Country found that e-WOM had a significant positive and significant effect on purchase intentions (Alrwashdeh et al., 2019). Research with the same results also conducted in Haripur Pakistan found that e-WOM had a significant positive and significant effect on purchase intention (Kazmi et al., 2016).
H1: E-WOM has a positive and significant effect on purchase intentions.
E-WOM is a social communication carried out on a website, where consumers send and receive information related to each other products online. Consumers write and read pleasant and unpleasant experiences written on internet networks (Jalilvand et al., 2012). Various useful information and opinions from others that are positively related to a brand, so e-WOM can encourage brand image (Samuel et al., 2014).
Research carried out in the automotive industry in Iran found that E-WOM communication had a positive and significant impact on the brand image (Jalilvand et al., 2012). Other research also supports that e-WOM has a positive and significant impact on the brand image (Tolak et al., 2014). Research conducted on the food industry also shows results that e-WOM has a positive and significant impact on the brand image (Charo et al., 2015). Research conducted on students at Islamabad and Rawalpindi universities found that e-WOM had a positive and significant relationship to brand image (Tariq et al., 2017).
H2: E-WOM has a positive and significant effect on brand image.
Brand image can contribute to the success of a company when customers are willing to buy products or services at a higher price. Brand value is about the value and satisfaction
that makes customers interested in using certain brands (Kazmi et al., 2016). Individual purchasing decisions on a particular brand occur after an in-depth evaluation so that individuals who have positive feelings towards a particular brand will increase purchase intentions (Moslehpour et al., 2017). WOMMA (2012) reports that nine out of ten people mention brand names 60 times a week in their daily conversation.
Research conducted on students in Samarahan City, Malaysia found that brand image had a positive and significant effect on smartphone purchase intentions (Shahrinaz et al., 2016). The same results are also shown in research conducted in Malaysia about Grab online motorcycle taxi, that brand image has a positive and significant effect on purchase intention (Chin et al., 2018). Based on empirical studies, the following hypotheses can be arranged:
H3: Brand image has a positive and significant effect on purchase intentions.
Purchase intention is a necessary component of consumer buying behavior because it influences consumers to make choices among various alternatives on the market with information about specific purchases of products or services. The purchase intention of a product depends on consumer confidence in the brand (Kazmi et al., 2016). Consumers consider the reviews of other consumers and get information about the product in the buying process (Torlak et al., 2014).
Previous studies conducted by Jalilvand et al. 2012, show that brand image significantly mediates the effect of e-WOM on purchase intentions. Research conducted by Suyoga et al. 2018 which shows the brand image mediates the influence of e-WOM on purchase intentions. Based on empirical studies, the following hypotheses can be arranged:
H4: Brand image plays a significant role in mediating E-WOM's purchase intentions.
CONCEPTUAL FRAMEWORK
Consumer behavior is influenced by several factors, according to the Theory Of Planned Behavior, one of which influences consumer behavior is subjective norms. Subjective norms are influenced by a person's beliefs obtained from the views of others associated with them (normative beliefs). In this study, testing the e-WOM variable as one of the variables affecting purchase intention. By doing e-WOM, consumers can get reviews from consumers who have previously bought a product before deciding to buy. Brand image is a thing that can be remembered and can make an impression in the minds of consumers very easily. Companies that have a good brand image will make consumers easily attracted to these products and increase consumer purchase intentions (Wang et al., 2008).
Based on the thinking and conceptual framework above, the research model can be described as follows:
This research was conducted by collecting data through a survey of high school students in the city of Denpasar. Data collection was carried out at once in one stage (one short study) or a cross-section through a questionnaire. The scope of this research is high school students in Denpasar City who have never and are not currently studying in Triton's tutoring to find out the influence of e-WOM and brand image that leads to purchase intentions.
The population in this study were all high school students attending school in Denpasar. In this study, a research sample is a group of research subjects drawn from the population. The criteria in selecting samples are high school students in grades X to XII who are studying in Denpasar City and have never attended tutoring in Triton's tutoring. The sampling technique in this study uses Nonprobability Sampling. The nonprobability sampling technique approach in this study is by using purposive sampling. This study uses 10 indicators so that using the estimation above the sample size is obtained by 100 samples.
The type of data used in this study is the type of primary data with the results of filling out the questionnaire by respondents using Google forms. The questionnaire was filled in by respondents who met the specified sample criteria. This study uses a questionnaire as an instrument in collecting data from respondents. The questionnaire contains several items consisting of questions about the identity of the respondent and items of statements about the research variables that must be answered by the respondent.
This research uses Partial Least Square (PLS), is a structural modeling technique that is very suitable for complex research models that involve many constructs and indicators (Hair, Ringle, and Sarstedt, 2011) as a general method for estimating the path of a model that describes latent constructs with multiple indicators. PLS is intended for causal-predictive analysis in situations of high complexity and low theoretical support. Wold mentioned PLS with soft modeling. PLS is a powerful analytical method because it can be applied at all data scales, does not require a lot of assumptions, and can also be used for relatively small sample sizes.
RESULTS OF STUDY
This study uses a two-step approach to measure the model before it is used for hypothesis testing, aimed at verifying the validity and reliability of a research model. First by analyzing convergent validity followed by analyzing discriminant validity.
An outer model test is carried out to ensure that the research indicators are feasible to be used as their role in measuring the research variables, then to see whether a model is valid to be the basis of the research three criteria must be met (Hair Jr., Hult, Ringle, & Sarstedt, 2016), namely: (1) all loading indicators must be above 0.65 (2) composite reliability (CR) must be above 0.8, and (3) average variance extracted (AVE) for each construct must exceed 0.5.
Table 1 - Model Size Results
Construct Indicator Outer Loading Composite Reliability Average Variance Extracted (AVE)
X1 0.810
X2 0.852
E-WOM X3 0.863 0.894 0.737
X4 0.886
M1 0.875
Brand Image M2 0.806 0.915 0.728
M3 0.892
Y1 0.901
Purchase Intention Y2 0.764 0.879 0.709
Y3 0.856
Source: Data processed, 2020.
Based on Table 1 shows that all outer loading indicators have values above 0.65 with a range between 0.764 to 0.901 which means they are in the recommendation limit, then the
composite reliability (CR) value is in the range between 0.879 to 0.915 all of which are above 0.8 meaning all The construct formed has good consistency as a research model, the third is the average variance extracted value (AVE) where everything is above 0.5, with a range from 0.709 to 0.737 so it can be concluded that the research model in this study has good validity.
To evaluate discriminant validity, a research model is recommended to ensure that the average variance extracted (VAVE) root value of a latent variable must be greater.
Table 2 - Correlation between Latent Variables
Construct e-WOM Brand Image Purchase Intention
e-WOM 1.000
Brand Image 0.790 1.000
Purchase Intention 0.772 0.751 1.000
Source: Data processed, 2020.
Table 3 - Root value of AVE
Construct Average Variance Extracted (AVE) Akar AVE
e-WOM 0.728 0.853
Brand Image 0.737 0.858
Purchase Intention 0.709 0.842
Source: Data processed, 2020.
Discriminant validity is considered good if the smallest root value of AVE (VAVE) in Table 3 is greater than the correlation value between the highest latent variables in Table 3. The research model proposed in this study can be considered well, where the smallest VAVE value is 0.842 greater than the correlation value between the highest latent variables that is equal to 0.790.
Structural models focus on hypothesized relationships or paths between latent variables. A structural model can be used to draw conclusions about the significance of the relationship between constructs and provide an explanation of the predictive power of the proposed model for research. The results of testing the inner model can be seen in Figure 1.
Figure 2 - Structural Model
The structural model is evaluated using R-square for the dependent construct and t-test as well as the significance of the coefficient of structural path parameters.
In this study, a bootstrap will be conducted which will produce two structural model measurements, namely: the value of t (t-test) and R2 which will be interpreted the same as multiple regression analysis in general. The predictive power of a research model can be
seen by looking at the R2 value generated by the bootstrap process, in Table 4. R2 values will be presented for each exogenous variable contained in the model.
Table 4 - Coefficient of Determination
Construct R2
Brand Image 0.624
Purchase Intention 0.649
Note: only endogenous (dependent) variables have the value R2
Source: Data processed, 2020.
Based on Table 4 above, it can be explained that the highest R2 value is found in the purchase intention variable of 0.644, which can be interpreted that as much as 64.4% of the purchase intention variable can be explained by the construct contained in the model, namely e-WOM and company image, while the lowest value is found in brand image variable with 0.621 which means 62.1% brand image variables can be explained constructs that affect these variables namely E-WOM. From the examination of the value of R2, it can be concluded that the predictive ability of this research model is moderate, seen from all variables that have the same R2 value or above 0.33 (Hair et al., 2011).
Changes in the value of R2 can be used to see the effect of exogenous latent variables on endogenous latent variables having substantive effects or not. This is measured using the effect size test (f2). The effect size test in this study can be seen in Table 5 below:
Table 5 - Effect Size
Construct e-WOM Brand Image Purchase Intention
e-WOM 1.675 0.245
Brand Image 0.150
Purchase Intention -
Source: Data processed, 2020.
In Table 5, the f2 value for the e-WOM variable to the brand image is 1.657 which is classified as strong because it is above 0.35. The f2 value for the e-WOM variable towards purchase intention is 0.245, which is considered moderate because it is above 0.15. The f2 value for the brand image variable for e-WOM is 0.150, which is classified as moderate, where the value is at 0.15.
The significance of the estimated parameters provides very useful information about the relationship between the research variables. The basis used in testing hypotheses is the value contained in the output path coefficients presented in Table 6 below.
Table 6 - Path Coefficient
Correlation between Variables Path Coefficient t-statistic P Values Information
e-WOM -> Purchase Intention 0.477 4.611 0.000 Accept
e-WOM -> Brand Image 0.790 12.941 0.000 Accept
Brand Image -> Purchase Intention 0.374 3.602 0.000 Accept
Source: Data processed, 2020.
Hypothesis testing is done by using t-statistics and looking at p-values. If p-value <0.05, the hypothesis is accepted. Based on table 6 above, it can be explained that e-WOM towards purchase intention has a t-statistic value of 4,611 with a p-value of 0,000 <0.05, then the hypothesis is accepted. This means that the better e-WOM, the higher the interest of students in following Triton's tutoring. e-WOM on the brand image has a t-statistic value of 12,941 with a p-value of 0,000 <0.05, the hypothesis is accepted. This means that the better e-WOM, the higher the student's perception of the Triton tutoring brand image. The brand image of purchase intention has a t-statistic value of 3.602 with a p-value of 0.000 <0.05, then the hypothesis is accepted. This means that the better the Triton's tutoring brand image, the higher the interest of students following the Triton's tutoring.
Test the role of brand image mediation on the influence of e-WOM on purchase intentions by examining indirect effects which are Smart PLS outputs as presented in Table 5.13. Based on Table 7, it can be explained that the t-statistic value is greater than the t table value (3.372> 1.96), then the brand image significantly mediates e-WOM towards purchase intention.
Table 7 - Indirect Effects
Correlation between Variables Path Coefficients t- Statistic P Values Information
e-WOM (X) - Brand Image (M) -Purchase Intention (Y) 0.295 3.372 0,001 Accept
Source: Data processed, 2020.
DISCUSSION OF RESULTS
E-WOM toward purchase intentions. Hypothesis testing is done by using t-statistics and looking at p-values. If p-value <0.05, the hypothesis is accepted. Based on table 6 above, it can be explained that e-WOM towards purchase intention has a t-statistic value of 4,611 with a p-value of 0,000 <0.05, then the hypothesis is accepted. This means that the better e-WOM, the higher the intention of students to follow the Triton Tutoring.
The results of this study were supported by Nuseir (2019), in his research conducted in the United Arab Emirates, it was found that E-WOM had a significant and positive effect on consumer purchase intentions in the United Arab Emirates. Research conducted on electronic consumers in Pakistan showed that e-WOM significantly and positively influenced purchase intentions. Consumers know the product and decide to buy or not depends on the online reviews of other customers who have used the product (Kazmi et al., 2016). e-WOM will positively influence consumer purchase intentions on a product.
E-WOM toward a brand image. e-WOM on a brand image has a t-statistic value of 12,941 with a p-value of 0,000 <0.05, the hypothesis is accepted. This means that the better e-WOM, the higher the consumer's perception of the Triton Learning Guidance brand image.
The results of this study are in line with research conducted on MIUI Indonesia members, e-WOM significantly, and positively influences brand image (Atika et al., 2016). Research conducted on the food industry also shows results that e-WOM has a positive and significant impact on the brand image (Charo et al., 2015). This means that the more often consumers do e-WOM positively then it affects the brand image in the minds of consumers.
Brand Image toward Purchase Intention. The brand image of purchase intention has a t-statistic value of 3.602 with a p-value of 0.000 <0.05, then the hypothesis is accepted. This means that the better brand image the Triton Tutoring has, the higher the intention of students to follow the Triton Tutoring. The results of this study are supported by research conducted on Facebook advertisements, the brand image of companies that advertise on Facebook has a significant and positive effect on consumer purchase intentions (Dehghani et al., 2015). Research on the Traveloka.com company in Banda Aceh found that the greater the brand image owned by traveloka.com, the stronger the purchase intention of the traveloka.com website in Banda Aceh (Rahmi et al., 2017).
The role of brand image mediates the effect of e-WOM on purchase intentions. Hypothesis testing for the indirect effect correlation of brand image mediation variables on e-wOm on purchase intentions produces a t-value of 3.372 with a p-value of 0.001 <0.05, thus it can be said that the brand image significantly influences indirectly the effect of e- WOM against purchase intentions. This means that the higher the brand image, the higher the influence of e-WOM to increase students' interest in following the Triton Tutoring, then H4 is accepted. In this study, brand image partially mediates the effect of e-WOM on purchase intentions.
The results in this study are in line with the results from previous studies conducted by Jalilvand et al. 2012, which shows that brand image significantly mediates the effect of e-WOM on purchase intentions. This means that e-WOM has a large effect on brand image and indirectly leads to intention to buy, especially in the automotive industry in Iran. Research conducted on the AirAsia airline company, based on calculations from the influence of e-WOM through a brand image on purchase intentions obtained VAF values as large as 40.6%, this value is between 20% to 80%, so the brand image can be categorized as partial mediation (Daughter et al, 2019).
CONCLUSION
The results of this study have shown that e-WOM has a positive and significant effect on purchase intentions in this case consumers are students of Denpasar City High School. E-WOM has a positive effect on brand image. Brand image has a positive effect on purchase intentions. Brand image mediates positively and moderately the influence of e-WOM on purchase intentions. These findings support the theory that the higher the brand image on student perceptions, the higher the desire of students to spread positive e-WOM on online media that will increase purchase intentions. The results of this study are expected to be empirical evidence for future research and can enrich the development of marketing management science related to e-WOM, brand image, and purchase intention. This research can be used as input to companies which in this study are Triton's Study Guidance to pay attention to the factors that can improve e-WOM, brand image, and purchase intention have been proven to be able to improve these factors. Companies need to always improve brand image so that candidates for the creation of positive e-WOM will increase student interest in attending tutoring at Triton.
REFERENCES
1. Abubakar, A. M., llkan, M. & Sahin, P. (2016). "eWOM, eReferral and gender in the virtual community", Marketing Intelligence & Planning, Vol. 34 No. 5, pp. 692-710.
2. Alrwashdeh, Muneer., Okechukwu, Lawrence Emeagwali., & Hasan, Yousef Aljuhmani. (2019). "The effect of electronic word of mouth communication on purchase intention and brand image: An applicant smartphone brands in North Cyprus", Management Science Letters, Vol. 9, pp. 505-518.
3. Atika., Andriani, Kusumawati., & Mohammad, Iqbal. (2016). "The effect of electronic word of mouth, message source credibility, information quality on brand image and purchase intention", Jurnal Ekonomi dan Keuangan, Vol. 20 No. 1, pp. 94-108.
4. Charo, N., Sharma, P., Shaikh, S., Haseeb, A. and Sufya, M.Z. (2015), "Determining the impact of eWOM on brand image and purchase intention through adoption of online opinions", International Journal of Humanities and Management Science, Vol. 3 No. 1.
5. Chin, ThooAoi, Nur Ain Nasyazwanie Salimi, Zuraidah Sulaiman. (2018). "Determinants of Brand Image and their Impacts on Purchase Intention of Grab". Journal of Arts & Social Sciences, Vol. 2 No. 1, pp. 26-36.
6. Dehghani, Milad., Mustafa Tumer. (2015). "A research on effectiveness of Facebook advertising on enhancing purchase intention of consumers". Computers in Human Behavior, Vol.49, pp. 597 -600.
7. Hair, J. F. Ringle, C.M. and Sarstedt, M. (2011). PLS-SEM: indeed a silver bullet. Journal of Marketing Theory and Practice, Vol. 19 No. 2, pp.139-151.
8. Jalilvand, Mohammad Reza dan Samiei, Neda. (2012). "The effect of electronic word of mouth on brand image and purchase intention (An empirical study in the automobile industry in Iran)", Jurnal Marketing Intellegence & Planning.
9. Kazmi, A. & Mehmood, Q. S. (2016). "The effect of electronic word of mouth communication and brand image on purchase intention: A case of consumer electronics in Haripur, Pakistan". Management Science Letters,Vol. 6, pp. 499-508.
10. Moslehpour, M., Wong, W.-K., Lin, Y. H., & Le Huyen Nguyen, T. (2017). "Top purchase intention priorities of Vietnamese lowcost carrier passengers: expectations and satisfaction". Eurasian Business Review, Vol. 8 No. 4, pp. 371-389.
11. Nuseir, Mohammed T. (2019). "The impact of electronic word of mouth (E-WOM) on the online purchase intention of of consumers in the islamic countries - acase of (UAE)", Marketing Intelligence & Planning.
12. Panjakajornsak, W. L. (2014). "The impact of electronic word-of-mouth factors on consumers' buying decision-making processes in the low cost carriers: A conceptual framework". International Journal of Trade, Economics and Finance, Vol. 5 No. 2, pp. 357.
13. Putri, Ni Putu Ayu Kartika Kurnia Putri., Tjokorda Gde Raka Sukawati. (2019) "Peran citra merek memediasi pengaruh e-WOM terhadapniat beli (studi kasus pada maskapai penerbangan airasia di kota denpasar)". E-Jurnal Manajemen, Vol. 8 No.7, pp. 4268 -4297.
14. Pratiwi, W.A., and Yasa, N., N., K. (2019), The Role of Brand Image in Mediating the Effect of Electronic Word of Mouth on Brand Attitude of Shopee Fashion E-Commerce in Denpasar. European Journal of Business and Management. Vol.11, No.18
15. Rahmi, Delsi Tuttia., Ahmad, Nizam. (2017). "Pengaruh citra merek terhadap niat beli pembelian tiket pesawat secara online pada konsumen traveloka.com dengan kepercayaan sebagai variabel mediasi". Jurnal Ilmiah Mahasiswa Ekonomi Manajemen, Vol. 2 No. 3, pp. 70-85.
16. Samuel, H. and Lianto, A. S. (2014). "Analisis Ewom, Brand Image, Brand Trust dan Minat Beli Produk di Surabaya". Jurnal Manajemen Pemasaran, Vol. 1 No. 2, pp. 47-54.
17. Shahrinaz, Irwan., Jati, Kasuma., Yusman, Yacob., Dayang, Hummida Abang Abdul Rahman., & Ahmad, Faisal Mahdi. (2016). "Relationship and impact of e-WOM and brand image toward purchase intention of smartphone?". Journal of Scientific Research and Development, Vol. 3 No. 5, pp. 117-124.
18. Suyoga, I. B. G. A, & Santika I. W. (2018). Peran Brand Image dalam Memediasi Pengaruh Electronic Word of Mouth terhadap Niat Beli. E-Jurnal Manajemen Unud, Vol. 7 No. 6, pp. 3230-3257.
19. Tariq, M., Abbas, T., Abrar, M., & Iqbal, A. (2017). "EWOM and brand awareness impact on consumer purchase intention: mediating role of brand image". Pakistan Administrative Review, Vol. 1 No. 1, pp. 84-102.
20. Torlak, Omer., Behcet Yalin Ozkara., Muhammet Ali Tiltay., Hakan Cengiz., dan Mehmet Fatih Dugler. (2014). "The Effect of Electronic Word of Mouth on Brand Image and Purchase Intention: An Application Concerning Cell Phone Brands for Youth Consumers in Turkey". Journal of Marketing Development and Competitiveness, Vol. 8 No. 2, pp. 6168.
21. Vahdati, H, Mousavi Nejad, S. H. (2016). "Brand Personality Toward Customer Purchase Intention: The Intermediate Role Of Electronic Word-Of-Mouth And Brand Equity". Asian Academy of Management Journal, Vol. 21 No. 2, pp. 1-26.
22. Wang, X, & Yang, Z., (2008), "Does country-of-origin matter in the relationship between brand personality and purchase intention in emerging economies?", International Journal of Marketing, Vol. 25 No. 4, pp. 458-474.
23. WOMMA (2012), "Word of mouth marketing association", available at: www.womma.org/
24. Yunus, Nur Helyana, Mohd Shoki Md Ariff, Norhuzaimah Mohd Som, Norhayati Zakuan, Zuraidah Sulaiman. (2016). "The Mediating Effect of Brand Image between Electronic Word of Mouth and Purchase Intention in Social Media". American Scientific Publisher, Vol. 22, pp. 3176-3180.
