The Impact of Word of Mouth ( WOM ) on the Bank Selection Decision of the Youth : A Case of Bahrain

For the youth, word of mouth (WOM) is an influential marketing tool in determining their bank selection decisions. The extent to which WOM influences the youth in selecting their banks depends on the amount of trust they might have in the source of WOM and how experienced said source is with the bank that they are recommending. This study found that WOM sources are of two types: personal and impersonal. Among the personal sources identified in this study, the study found WOM from parents who have accounts in the recommended banks is the most influential. Among the impersonal sources, the study found that online reviews posted in specialized media have the most impact on young people’s bank selection decisions.


Introduction
Word of mouth (WOM) means exchanging opinions (positive or negative) about a firm's products.Such opinions may have a significant impact on consumers' behavior (Goyette, Richard, Bergeron, & Maricotte, 2008;East, Hammond, & Lomax).Literature indicates that the effectiveness of WOM in changing consumers' attitudes and influencing their buying decisions is more than advertising or even direct sales (Katz & Lazarsfeld, 1955;Day, 1971;Morin, 1983).Duhan, Johnson, Wilcox and Harrell (1997) found that WOM is the most powerful means of promotion for attracting and retaining customers.There is a strong positive association between WOM and satisfaction (Anderson, 1998).Satisfied customers will pass positive WOM to others while dissatisfied customers will pass negative WOM (Tucker, 2011;Vermeulen & Seegers, 2009;Jalilvand, Samiei, Dini, & Manzari, 2012;Dennis, Merrilees, Jayawardhena, & Wright, 2009).Therefore, given that satisfaction is one of the major sources of WOM, it is important that companies consider customer satisfaction to be a core focus of their businesses.In addition to satisfaction, a feeling that one might get after using a product, other sources of WOM include communications from trusted people (Bergeron, Ricard, & Perrien, 2003), service quality (Parasuraman, Zeithmal, & Berry, 1988), or the level of relationship that the buyer has with the seller (Boles, Barksdale, & Johnson, 1977).
The objective of this study is to consider personal and impersonal sources and to investigate the effectiveness of WOM in influencing the youth in Bahrain in their bank selection decisions.

Importance of WOM as a Marketing Tool
The importance of WOM in shaping consumers' attitudes and buying decisions led many researchers to examine its effectiveness in stimulating demand within various industries (for example : Walker, 1995;Suderlund & Rosengren, 2007;East, Hammond, & Lomax, 2008;Silverman, 2001;Aghdaie, Piraman, & Fathi, 2011;Lim & Beatty, 2005;Mir, 2011;Chen, 2011).Lytras et al. (2009) found that positive WOM will most likely result in having a stronger relationship between consumers' emotional trust and their intention to shop online.WOM complements advertising by disseminating the information that is embedded in the advertisement (Goldenberg, Libai, & Muller, 2001;Day, 1977).Past studies have shown that WOM has significant impact on consumers' attitude and buying decisions (Soderland & Rosengren, 2007;Bone, 2005;Charlette, Garland, & Marr, 1995) especially in situations where products perform above or below customers' expectations (Bone, 1995).WOM also plays an important role in changing negative or neutral attitudes into positive attitudes (Day, 1971).Lee and Lee (2009) reported that WOM has an influence on the consumers' buying judgments.This was confirmed by Ladhari (2007) who found that there is a positive relationship between satisfaction and positive WOM.Similarly, Chaniotakis and Lymperopoulos (2009) found such a relationship in the service markets.On the other hand, other studies reported that dissatisfaction results in negative WOM (Richins, 1983).Matos and Rossi (2008) explained the situation from a different perspective.They reported that dissatisfaction leads to disloyalty which results in spreading negative information about a product, which is to say, negative WOM. Firms spend a lot of money and time to avoid negative WOM because of its ability to damage a firm's reputation or the reputation of its products (Hart, Heskett, & Sasser, 1990) and its effect is stronger than positive WOM (Weinberger & Dillon, 1980;Chevalier & Mazlin, 2006;Scott & Tybout, 1981;Charlette, Garland, & Marr, 1995).Fails & Francis (1996) investigated the ability of demographic variables to predict consumers' WOM.These studies reported that variables such as sex, educational level and age were predictors of positive WOM, while variables such as income and age were predictors of negative WOM. Kempf and Palan (2006) reported that the demographic factor of sex plays an important role in forming WOM, stating "positive WOM is most influential on brand evaluations when the communicator's sex and the WOM recipient's sex are opposite".

WOM and Banking Services
One of the service industries that received a good amount of attention from researchers is banking (Shirsaver, Gilaninia, & Almani, 2012;Najmi, 2009).Dillon (2013) investigated the financial services sector in India and found that the majority of respondents considered WOM an effective means of communication for attracting new customers and retaining existing ones.A similar result was reported by Gremler and Brown (1996); Reichheld and Sesser (1990);Ennew, Banerjee, and Li (2000).Zhou (2004) investigated the banking services in China and reported that, due to some social and cultural factors, many Chinese consumers do not have confidence in the quality of banking services provided by Chinese banks.Such negative perception is therefore likely to lead to negative WOM. Lack of confidence in the banking services might also be the result of an increase in perceived risk, which can reduce customers' willingness to use banking services (Aurier & Siadou-Martin, 2007).Shirsaver, Gilaninia and Almani (2012) conducted a study on bank customers in Iran and reported that WOM is an important competitive advantage through which Iranian banks can increase their acquisition of customers and retain existing ones.This study found that the major determinant factors of positive WOM are corporate image, relationship marketing, perceived value, perceived risk, satisfaction, and loyalty.According to Jarvenpaa & Todd (1997), positive WOM helps consumers have more trust in Internet banking.Yavas, Benkenstein, and Stuhldreier (2004) studied the relationship between service quality and the behavior of private bank customers in Germany and reported that tangible elements of service quality is positively associated with positive WOM.Lymperopoulos and Chaniotakis (2008) studied the impact of WOM through the human factor of service quality in the banking sector in Greece.This study found that personnel efficiency and price satisfaction are antecedents of customer satisfaction, which is a prerequisite for positive WOM.The relationship between satisfaction and WOM was also investigated by File, Judd, and Prince (1992) who interviewed 325 medium-sized business owners in the USA and reported that bank marketers put in a lot of effort to increase customer satisfaction because they believe that satisfied customers will disseminate positive WOM, which in turn serves as a powerful factor in customers' decisions to purchase financial services.This indicates that WOM plays a significant role in the growth of the banking business.Such a finding is supported by Metcalfe (2008) who reported that Signature Bank in New York achieved growth through positive WOM.

The Power of WOM
Based on the literature, it becomes evident that there are two major sources of WOM.First, personal sources, which include friends, family, acquaintances, and experts (Goyette, Ricard, Bergeron, & Marticotte, 2010;Brown & Reingen, 1987;Duhan, Johnson, Wilcox, & Harrell, 1997).Second, impersonal sources, which include the news, articles, views and reviews found in printed media, broadcast media, specialized publications, or online discussion forums (Goyette, Ricard, Bergeron, & Marticotte, 2010;Senecal, Kalczynski, Nantel, 2005;Senecal & Nantel, 2004).The literature also indicates that the effectiveness of WOM from these two sources very much depends on two factors: trust and experience.In other words, WOM's effectiveness depends on the amount of trust that the WOM receiver has in the WOM sender.The more the receiver trusts the sender, the more effective WOM becomes.The other factor that determines the power of WOM is how experienced the WOM's source is with the product recommended.Personal sources who used the product and impersonal sources that are more specialized in the product would have more impact on the attitudes and buying decisions of the WOM receiver.
As for trust factor, Murray (1991) found that service buyers have more confidence in personal sources of information and that personal WOM has a greater impact on service buyers than on goods buyers.Mangold, Miller, & Brockway (1999) stated that interpersonal communications have a significant influence on consumers' attitudes and buying decisions.In some studies, such influence is called normative social influence (Asch, 1953;Stafford, 1966;Venkatesan, 1966;Pincus & Waters, 1977).This finding is also supported by Ennew, Banerjee, & Li (2000) who stated that the influence of WOM would be stronger when it originates from social contacts due to their perceived reliability.One of the influential social contacts are people with whom the WOM receiver has strong relationships such as close family and friends, referred to as "in-groups" by Lam and Mizerski (2005).Therefore, how close the communicator is with the receiver plays a significant role in how effective the WOM is likely to be. Brown and Reingen (1987) reported that WOM from strong ties (close friends and relatives) has more impact than WOM from weak ties (acquaintances and distant relatives).Such a finding is also supported by East, Hammond, and Lomax. (2008) who said that solicited advice from close friends and relatives has more impact on the receivers' attitudes and buying decisions (East, 2003;Bansar & Voyer, 2000;Gremler, 1994).WOM from close friends and relatives has more impact due to the perceived trust and credibility (Sweeney, Soutar, & Mazzarol, 2008).The importance of trust and credibility in achieving positive buying behavior has been investigated in a number of studies (for example, Moorman, Gerald, & Rohit, 1993;Morgan & Hunt, 1994;Garbarino & Johanson, 1999;Singh & Sirdeshmukh, 2000;Chaudhuri & Holbrook, 2001).In this respect, Dillon (2013) found that WOM becomes more powerful if WOM sender is known and trusted by the receiver.A similar result was reported by Reichheld and Sasser (1990) who stated on page 65 of their study that General Electric found that "recommendations from friends and acquaintances carry twice the impact of paid advertising when consumers make purchasing decisions".Similarly, Ennew, Banerjee, and Li, (2000) found that personal influences is seven times more effective than print advertising in magazines or newspapers.A similar result was also reported by Katz and Lazarfeld (1955).In addition to family and friends, the opinions of experts may also have an impact on the effectiveness of WOM.Experts' advice is seen to be more persuasive if the expert does not have any ties to the recommended company or product (Keaveney, 1995;East, Hammond, & Lomax, 2008;Herr, Kardes, & Kim, 1991).This is assured by Silverman (2001) who mentioned that for WOM to be convincing, the message being transmitted and the sender of the message must be perceived as independent from the company whose product is being promoted (Goyyette, Ricard, Bergeron, & Marticotte, 2010).Hence, the advice of people who we trust has a positive impact on our attitudes and buying behavior (Beldad, Jong, & Steehouder, 2010;Lu, Zhao, & Wang, 2010).
In addition to trust, WOM sender's experience may also play a significant role in the WOM's effectiveness (Chiu, Huang, & Yen, 2010).In a review of services literature, Gabbott and Hogg (2004) reported that reliance on WOM sources increase when experience-related information is offered.This is also supported by Murray (1991) who stated that, in making service decisions, consumers like to listen to the opinions and experiences of others.Opinion leadership research assures the impact of source expertise on interpersonal influence (Bansal & Voyer, 2000;Gilly, Graham, Finley, & Yale, 1998;Fitzgerald, 1995;Wangenheim & Bayon, 2004).To show the importance of experience in WOM effectiveness, some researchers investigated the issue from a satisfaction point-of-view and suggested that customer satisfaction results from purchase experience (for example : Oliver, 1993;Meuter, Ostrom, Roundtree, & Bitner, 2000;Maxham, 2001;Yu, Wu, Chiao, & Tai, 2005;Chen & Chen, 2009).With respect to experience-related satisfaction, other researchers reported that customer satisfaction is an overall cumulative evaluation of a company after consumers have experienced their products (Anderson, Fornell, & Lehman, 1994;Goyette, Ricard, Bergeron, & Marticotte, 2010).The influence of experience can also be viewed through an impersonal source of WOM, that is, reviews.Reviews are usually posted online by people who have experienced the product (Chen, 2011).A reviewer's comments about a product might have more influence if it is posted on reputable sites (Ping, Luping, & Luluo, 2013).Nowadays, many people like to share their thoughts and experiences about the products they use (Bone, 1992;Jalilvand, Samiei, Dini, & Manzari, 2012).With online facilities, such reviews are on the rise and quickly becoming one of the sources that customers refer to before buying their products (Goldenberg, Libai, & Muller, 2001;Gretzel & Yoo, 2008).

The Pilot Study
Having reviewed the literature on WOM, we found that WOM has two major sources, personal and impersonal (Goyette, Ricard, Bergeron, & Marticotte, 2010).In addition, we found that the elements of personal and impersonal sources and their effectiveness depend on the nature of the product being recommended as well as on who the WOM recipient is.Therefore, to identify the appropriate WOM elements for the youths' bank selection decisions, we conducted a pilot study using four focus groups of eight young persons each, 50% of them male and the other 50% female.After spending twenty minutes with each focus group, the pilot study's results were as follows: 1) In their bank selection decision, the youth might be influenced by three personal sources (relatives, friends, and experts) and two impersonal sources (online reviews and news or articles).
2) The effectiveness of each WOM source depends on the amount of trust the youth have in that source.
3) For the effectiveness of personal sources, the pilot study findings indicate that among relatives, the youth mostly trust the recommendations of their parents followed by close relatives (for example, brother, sister, uncle, aunt) and then distant relatives.With respect to friends, WOM from close friends is trusted more than that of acquaintances.For experts, the recommendations of those who are independent from the bank being recommended are trusted more than those who are associated with the bank.4) For impersonal sources, the degree of trust that the youth might have in the source depends on the nature of the media on which the recommendations are posted.Online reviews and news published on specialized media are more trusted than those published on non-specific ones.5) In addition to trust, the effectiveness of WOM sources on the youth's bank selection decisions also depends on the experience of WOM source with the recommended bank.The pilot study found that the recommendations of those who dealt with or have accounts in the bank are more effective than those who have no association with the bank.

Study Conceptual Model
Based on the literature review and the pilot study explained in section 3 we developed the study model, shown in Figure 1.The model indicates that in selecting their banks, the youth are influenced by WOM from personal and impersonal sources.Personal sources are those with whom the WOM recipients interact directly such as relatives, friends and experts.Impersonal sources are those that WOM recipients refer to such as online reviews and news or articles in various media.The model also suggests that the effectiveness of both personal and impersonal WOM depends on the degree of trust that recipients have in the WOM senders and the experience that the sender might has with the bank being recommended.

Research Hypotheses
Based on the review of relevant literature on WOM, the findings of the pilot study, and the conceptual model shown in Figure 1, we propose the following three hypotheses: H1: Both Personal and impersonal WOM have a significant influence on the youth's bank selection decision.WOM from personal sources is expected to have more impact than WOM from impersonal sources.H2: WOM from sources that the youth trust more will have a greater impact on their bank selection decision.H3: WOM from sources who have experience with the bank being recommended will have a greater impact on the youth's bank selection decision.

Study Sample and Data Collection
The study recruited 300 young customers in Bahrain, aged 20-25 years old, 60% of them were female while the other 40% were male.The required data were collected through personal interviews using a specially designed questionnaire.

The Questionnaire
The questionnaire includes 23 questions.Two of the questions are demographical related to age and gender.The remaining 21 questions intended to determine the effectiveness of WOM through trust (11 questions) and experience (10 questions).

Variables of Interest and How They Are Measured
In this study, we proposed that the effectiveness of personal and impersonal WOM on the youth's bank selection decision depends on the amount of trust WOM receiver has in WOM source and the experience that WOM source has with the recommended bank.In this study, we measured trust and experience as follows:

 Trust
We identified the degree of trust in the recommendations of personal and impersonal sources by asking the following direct question: Parents, close relatives, distant relatives, close friends, acquaintances, independent experts, associated experts, online reviews in a general site, online reviews in a specialized site, news/articles in a general media, news/articles in a specialized media?" Each of these 11 sources was followed by a five-point scale: {very effective, effective, somewhat effective, not effective, not effective at all}.

 Experience
We gauged the impact of the WOM on the impact of the WOM sender's experience on the youth's bank selection decision through the following five personal sources: Parents, close relatives, distant relatives, close friends, and acquaintances).For each of the five sources the respondents were given two scenarios.In one scenario, the WOM sender had an account in the bank that he/she was recommending.In the other scenario, the WOM sender did not have an account.For instance, for parents, the impact of experience was tested through asking the following question:

"Suppose you intend to open an account in a commercial bank, to what extent may the following WOM scenarios influence your bank selection decision?
A parent who has an account in bank X told you positive things about bank X A parent who has no account in bank X told you positive things about bank X Each scenario was followed by a five-point scale {very effective, effective, somewhat effective, not effective, not effective at all} from which respondents were asked to choose one.The difference between the responses obtained in the two scenarios indicated the importance of experience in the youth's bank selection criteria.
We applied the same process to the other four personal sources.

Research Findings
Research analysis started by calculating Cronbasch Alpha (α) in order to test the reliability of the internal consistency of the scales used.Results of Cronbasch Alpha revealed that the scales used were reliable with an alpha value of 0.663 for trust and 0.717 for experience.
The first hypothesis was tested by averaging the mean scores of the seven elements of WOM personal sources (parents, close relatives, distant relatives, close friends, acquaintances, independent experts, and associated experts).Results of the analysis revealed that the mean of personal WOM is 3.4.Such a significant result indicated that the impact of personal WOM on the youth's bank selection decisions is effective.To find the impact of the impersonal sources of WOM on bank selection decisions, the average mean of the four elements used in this study (online review posted in a specialized media, online review posted in a general media, new/articles posted in a specialized media, and news/articles posted in a general media) was calculated and resulted in having an average mean of 2.7.This result indicates that impersonal WOM is somewhat effective for influencing the youth to choose their banks.Comparatively, it is clear that personal WOM (mean = 3.4) has more impact on the youth's bank selection decisions than impersonal WOM (mean = 2.7).Furthermore, results of t-test show that the difference between the mean score of personal and impersonal sources of WOM is significant (t = 28.573,df = 201, Sig = 0.000).The results discussed above fully support hypothesis 1.
Before testing the second hypothesis, data on personal and impersonal WOM was analyzed in order to determine the ranking of such elements based on the trust that the youth have in each element.To achieve this, we analyzed the responses of the youth to the following question: "which source do you trust more in choosing your bank?" The results are shown in Table 1.The findings detailed in Table 1 are similar to the findings of the pilot study discussed earlier, except for experts.
With respect to the personal sources of WOM, Table 1 shows that when it comes to selecting a bank, young customers trust their parents the most, followed by their close relatives and then their distant relatives.Among friends, WOM from close friends is trusted more than that from acquaintances.For experts, the recommendations and advice of those experts who are independent from the bank are trusted less than that of experts who are associated with the bank being recommended.For the impersonal sources of WOM, Table 1 indicates that the young customers trust online reviews posted in specialized media more than those posted in general media.A similar result was found for the impact of news and articles; those that were posted in specialized media seem to be more trusted than those posted in general media.
After ranking the elements based on trust, we can now test the second hypothesis.T-test was used to examine the role of trust on the WOM's impact.The findings revealed that for all the WOM sources that we tested, a WOM's effectiveness was positively associated with trust.In other words, WOM from sources that the youth trusted more had more impact on their bank selection decision, as shown in Table 2.
As detailed in Table 2, those WOM sources that scored higher in terms of trust factor also scored higher in terms of WOM impact factor.For example, for selecting their banks, recommendations and advice of parents have the most significant influence on youth's decision (mean = 4.31) because they trust their parents the most, followed by their close relatives (mean = 3.91) and then distant relatives (mean = 2.94).T-test shows that the difference between the WOM impact of the three relative sources is significant (t=*************, Dig ***************).
For friends, results in Table 2 shows that close friends who are trusted more than acquaintances scored a higher WOM impact (mean = 3.69 and 2.36 respectively) and the difference between the WOM impact of the two friend sources is significant (t = 11.897,Sig = 0.000).The scenario was different for experts.Results indicate that the WOM from experts associated with the bank being recommended is marginally more effective (mean = 3.23) than that from experts independent from the bank being recommended (mean = 3.19).This is also confirmed through the result of the t-test which shows that the difference between the two is insignificant (t = 0.326, Sig = 0.745).With regard to online reviews, online review on specialized media that scored higher trust than online review posted on general media also scored higher for WOM impact (mean = 3.10 and 2.56 respectively) and the difference between the two found significant (t= 4.492, Sig = 0.000).Finally, a similar scenario was found for news and articles.As shown in Table 2, news posted on specialized media which scored higher trust than news posted on general media, also scored higher for WOM impact (mean = 2.96 and 2.52 respectively) and the difference between the impact of the two sources found significant (t= 4.640, Sig = 0.000).The above results give full support to accept hypothesis 2.
For testing the third hypothesis, we considered five scenarios related to five personal sources of WOM (Parents, close relatives, distant relatives, close friends, and acquaintances).Each scenario contained two cases, one with experience and one without experience.The difference between the impacts of each scenario indicates the role of experience in WOM effectiveness.The youth's responses to the two cases within each of the five scenarios were analyzed and compared using t-test statistics.The summary of the findings is shown in Table 3.
As it is clear from Table 3, experience plays a significant role in the impact of WOM from different personal sources.For example, the youth stated that the recommendations of a parent who has experience with bank X would be more effective (mean = 3.38) than the recommendations of a parent who has no experience with bank X (mean = 2.42).T-test shows that the difference between the impact of the two scenarios is significant (t = 12.642, Sig = 0.000).A similar argument is true for the other four sources shown in Table 3.People who have more experience with a bank that they recommend are in a better position to judge the positive and negative traits of the bank.Besides, the advice of experienced people is more trusted by WOM recipients.
The above findings and discussions provide enough support to accept hypothesis 3.
Table 3.The role of experience in the impact of WOM on the youth's bank selection decisions

Conclusion
This study has proven that WOM has a positive influence on the youth's decision-making process when choosing where to open a bank account.Youth receive their WOM from two main sources, personal and impersonal.As for personal sources, relatives, friends and experts leave the strongest mark on a person's decision.Relatives with the most influence are parents, then close relatives, followed by distant relatives.As for friends, close friends have a stronger influence compared to acquaintances.Finally, associated experts are marginally more influential compared to experts independent from certain banks.The experience of relatives and friends with a particular bank highly influences the WOM that they provide.
Youth are also influenced through the WOM of impersonal sources, such as online reviews and news.Online reviews that come from specialized media sources tend to better influence WOM compared to online reviews on general websites.News and articles follow the same pattern as before, whereas specialized sources for news and articles are considered to leave a stronger impact on a person's decision-making compared to news and articles coming from the general media.
With respect to both (personal and impersonal) sources of WOM, this study shows that sources with a higher trust level have a greater impact on recipients.
In conclusion, this study proves that WOM plays an important role when it comes to the youth's decision-making process about where to open a bank account.

Figure 1 .
Figure 1.The conceptual model of the role of trust and experience in the effectiveness of WOM

"
In selecting your bank, to what extent do you trust the recommendations or advice of the following sources: A parent who has an account in bank X told you positive things about bank X Scenario 2: A parent who has no account in bank X told you close relative (i.e.brother, sister, cousin, etc.) who has an account in bank X told you positive things about bank X Scenario 2: A close relative (i.e.brother, sister, cousin, etc.) who has no account in bank X told you positive things about bank A close friend who has an account in bank X told you positive things about bank X Scenario 2: A close friend who has no account in bank X told you positive things about bank X An acquaintance who has an account in bank X told you positive things about bank X An acquaintance who has no account in bank X told you positive things about bank X 3.07 Note.* Very effective = 5, Effective = 4, Somewhat effective = 3, Not effective = 2, Not effective at all = 1.

Table 1 .
The trust the youth have in personal and impersonal sources of WOM for selecting their banks

Table 2 .
Role of trust in WOM effectiveness on the youth's bank selection decision