The Impact of Word of Mouth on Customer Perceived Value for the Malaysian Restaurant Industry

Abstract

Purpose - The purpose of this research is to determine the impact of word of mouth on customer perceived value for restaurants in Malaysia. The objectives of this research include determining how word of mouth (WoM) factors - frequency of word of mouth messages, reputation of word of mouth messenger, richness of word of mouth message, dispersion of word of mouth conversations and manner of word of mouth delivery impact customer perceived value in Malaysian restaurant industry. Research Design, Data, and Methodology - The research follows causal / explanatory research method based on quantitative data. A sample of 150 restaurant customers in Kuala Lumpur, Malaysia was selected using convenience sampling technique. Likert scale questionnaire is used to collect data and data is analysed using regression analysis through SPSS 22. Results - The statistical analysis revealed that independent variable 'manner of delivery' significantly and positively impacts customer perceived value for restaurants in Malaysia. Conclusions - To build strong positive customer perception, Malaysian restaurants can enhance word of mouth campaigns' 'manner of delivery' by making them passionate, exciting and with high emotional appeal.

1. Introduction

1.1. Background

Word of Mouth (WoM) is popular research since long, particularly when it comes to relational analysis (Walsh et al., 2009). Many times customers pass the information about their purchase experience to someone else through oral or any other communication means. WoM is act of telling such experience to at least one of the acquaintances (Richins, 1983). It is information diffusion about the experience of purchase (Higie et al., 1987). Acts of such information diffusion might be are higher when it comes to unsatisfactory experiences (Singh, 1990). The information spreading can be in small as well as in big groups (Brown & Reingen, 1987) and so WoM can be a group phenomenon of exchanging the comments (Bone, 1992). WoM is an important promotional medium for businesses. According to Day, 1971 WoM has more effectiveness than traditional advertising techniques. WoM is one of the preferred factors that customers generally rely on while making purchase decisions (Hawkins et al. 2004). Cost effectiveness makes WoM highly effective advertising tool (Godes et al., 2005) and so the powerful factor in marketing (Taghizadeh et al., 2013). Another aspect underpins the importance of WoM in marketing is consumer trusts social circle more than advertisements by companies. (Ng et al, 2011). Consumers experience the product or services and they make opinions about it. Such opinions are shared and informed to others through personal communications or other tools (Brown et al., 2015). Advertising might be important in attracting customers for first time and then WoM plays its role. It is important to note, in most of the cases, WoM is difficult to manage unlike regular advertising (Chevalier & Msyzlin 2006).

In the pioneer study conducted by Ernest Ditcher, it was argued that consumers reject advertising messages because of perceiving it as sales tools until the products are humanized would it appeal to consumers. It has been investigated that based on WoM customers will evaluate benefits and sacrifices (Gwinner, Gremler & Bitner, 1998; Rust, Zeithaml, & Lemmon, 2000). Customers carry out assessment of value based on WoM (Walsh et al., 2009, Sundaram, Mitra, & Webster, 1998). Word of Mouth can significantly affect customers’ perception (Anderson et al., 1994; Athanassopoulos et al., 2001). There have been many investigations later to further explore relationship between word of mouth and customer perceived value. More recent studies focus on offline as well as online word of mouth and its impact on perceived value. It was determined that while social and functional drivers were most important for online word of mouth, emotional driver is the most important for offline word of mouth (Lovett, Peres, & Shachar, 2013). Positive word of mouth can happen organically but it can also be managed or facilitated by the business (Kotler & Keller, 2012).

It can also be observed that irrespective of how different the cultures may be, the impact of word of mouth on customer perceived value cannot be under estimated (Chu & Choi, 2011). In the Malaysian context, customer satisfaction and trust significantly affects loyalty through word of mouth (WOM) communications resulting in more repeat purchases (Kassim & Abdullah, 2010). WoM has positive and significant influence on brand image, brand loyalty, brand leadership and brand preference (Hanayshya, 2016). This shows that WoM is quite significant factor in Malaysian market. Thus, there is a clear link between WOM communication and customer perceived value. WoM either through face-toface or electronic-mediated communication provides a means of suggestion or recommendation or lack thereof of a product or service based on the participants thus creating feelings of attachment or affiliation for a particular product or service stronger than that felt for the competitor, hence a consumer perceives more value for the product and makes a choice to purchase the product. Focus of this research is to determine the impact of word of mouth on customer perceived value in the Malaysian restaurant industry.

1.2. Significance

The Malaysian food services sub-sectors include fullservice restaurants, cafés/bars, fast food, street kiosks/stalls, self-service cafeterias, and 100% home delivery and takeaway. Data suggest that the Malaysian food services industry by market value continues to grow. As per Euromonitor 2014, between 2008 and 2012 the Malaysian food services subsectors saw a 4.6 percent compound annual growth rate by market value and is forecasted to experience a compound annual growth rate increase to 5.3 percent in market value from 2013 to 2017. The “Full service restaurant subsectors is expected to grow by 5.4 percent in market value while the highest growth rate is expected to come from “100% home delivery/takeaway” subsector. The Malaysian food services subsector had a total of 30,721 outlets with “street stalls/kiosks” having the most number of outlets (11,201 outlets) followed by “Full-service restaurants with 10,231 outlets. Full-service restaurants was the third by transactions in 2012 after “street stalls/kiosks” and “Fast food” (Agriculture and Agri-Food Canada 2014, p. 2). Restaurant operators will be able to identify the factors that significantly contribute to customer perception of value and use the recommendations of this research for improvement.

These days, Malaysian urban consumers are increasingly dining out in restaurants as people are busy in their routine activities with hectic schedule. Some of the key reasons are food price, convenience, individual characteristics, seeking pleasure and social attraction (Cardas Research, 2015). But on the other hand competition is quite tough with high buyer power, low supplier power and high threat of new entrants. Data suggests increased industry competition hence necessitating restaurants to invest in developing personal relationships with customers through word of mouth. However, the exact factors of word of mouth that would deliver the most customers value needs to be identified which is the focus of this research and makes this research significant. Moreover, this research will add to the marketing literature in terms of identifying the important word of mouth factors that significantly influences perception of restaurants’ perceived value in the Malaysian context. Moreover, advertisers can leverage on the findings of this research to improve word of mouth implementations among restaurants operating in the Malaysian food service industry.

1.3. Problem Statement

Many organizations have traditionally resorted to mass media advertising using television, newspapers, radio, magazines, yellow pages, outdoor spaces, internet, fliers, billboards and posters (Armstrong & Kotler, 2013). Due to limitations of some of the above media types such as high cost of television advertisements, fleeting nature of radio presentations and increasing clutter of the internet organizations are increasing favoring personal communications with customers (Kotler & Keller, 2012). Moreover, this concern is especially difficult for small businesses as small and medium scale businesses (SMEs) do not have big marketing budgets usually available to large corporations and international food chains. The presence of a large number of food services organizations in Malaysia has resulted in stiff competition. Food service businesses are seeking the most effective means of promoting its services in order to attract and retain customers. Therefore, it is desired that small Malaysian food services organizations are able to access various and effective means of promoting its offering just like the large businesses in order to increase customer perceived value and continued patronage.

1.4. Research Questions

To what extent does ‘frequency of word of mouth messages’ influence customer perceived value in the Malaysia restaurant industry? To what extent does ‘reputation of word of mouth messenger’ influence customer perceived value in Malaysia restaurant industry? To what extent does ‘richness of word of mouth message’ influence customer perceived value in Malaysia restaurant industry? To what extent does ‘dispersion of word of mouth conversations’ influence customer perceived value in Malaysia restaurant industry? To what extent does ‘manner of word of mouth delivery’ influence customer perceived value in Malaysia restaurant industry?

2. Literature Review

2.1. Discussion on Previous Frameworks

‘Mere Exposure Theory’ suggests that the more frequent an individual is exposed to a stimulus; more the individual tends to like the presented stimulus (Toomey & Francis, 2013). However, Mere Exposure Theory presumes that recipients of the stimulus will always take action resulting from the stimulus which means assuming that listeners to messages passed through word of mouth will always develop a likeness thus increasing value perception of the restaurant. This may not always be the case especially in situations when customers determine that companysponsored messages were disguised as word of mouth (Magnini, 2011). Sweeney et al. 2008 framework shows that personal, interpersonal, message characteristics and situational characteristics can influence word of mouth messages to the receiver by reducing the receiver’s perceived buying risk, improve firm perceptions and increase the likelihood of buying. However, this model ignores factors such as the influence of frequency of word of mouth message and extent of dispersion of the word of mouth message on the actions of the receiver (Kucukemiroglu & Kara, 2015; Wang, Sun, & Peng, 2013). Kucukemiroglu & Kara(2015) framework shows that individual’s innovativeness, social capital and trust will influence opinion leadership and opinion seeking that results in individuals engaging in WoM activities online. In other words, if an individual is not able to navigate online (innovativeness), have the people to get and provide information (social capital) or trust the circle of friends, WoM would not occur or facilitated. Other factors ignored by this model included message characteristics such as strength of delivery and vividness or personal characteristics such as credibility of the sender (Sweeney, Soutar & Mazzarol, 2008). Tam, 2010 model shows that perceived value and customer satisfaction are critical factors affecting post-purchase behaviors of customers. This model indicates that in order for organizations to ensure repeat purchases, it must identify customer perception of the value of its services.

Chen & Dubinsky (2003) model shows that perceived customer value in an ecommerce setting is directly affected by valence of experience, perceived risk, product price and perceived product quality. However, this model ignored the effect word of mouth can have on E-tailer reputation, perceived risks, perceived customer value and purchase intention. Hence, this study suggests that word of mouth (online or face-to-face) influences customer perceived value and eventual purchase decisions (offline or online). An exploratory study that investigated the concept of word of mouth found that word of mouth (especially with richness of content and strength of implied or explicit advocacy) can significantly influence consumer attitudes and behaviors (Mazzarol, Sweeney, & Soutar, 2007). This suggests that customer perceived value can be increased when word of mouth is positive and appealing. Some drivers of word of mouth include customer satisfaction and favorable business activities (Longart, 2010). Also, in a different scenario, firm’s customer perceived value, loyalty and trust can be eroded if customers find out that an organization’s sponsored marketing activities is disguised as genuine word of mouth (Magnini, 2011)

2.2. Discussion on Concept of Word of Mouth

WoM refers to the face-to-face or electronic-mediated communication process that includes a suggestion or recommendation or lack thereof of a product or service based on the participant or consumer’s experiences (Kotler & Keller, 2012). WOM is influenced by key factors such as frequency of message, reputation of messenger, richness of message, dispersion of conversations and manner of message delivery. The frequency of WoM messages can influence customer perception of a business (Wang, Sun, & Peng, 2013). However, other studies suggest that in terms of reach, standard metrics for measuring word of mouth especially using social media can be overestimated. A study by Groeger & Buttle(2014) highlights that frequency of reach can be misleading and suggests that proper measurement of the effects of word of mouth is critical in order not to confuse individuals with high frequency of message reach with individuals whose perception of the organization’s value has increased. Wu & Wang(2011) suggest when the reputation of the word of mouth message source is highly credible, listeners expressed better attitude towards the brand praised compared to when the message source is less reputable. Notwithstanding, other studies have suggested that credibility of a word of mouth message source is subjective and cannot always truly influence customer perception of business value especially when organizations can disguise sponsored messages as genuine word of mouth (Magnini, 2011). The reputation of word of mouth message sources/sender can influence the impact of the message on the receiver (Lim & Chung, 2014).

Richness of word of mouth message can influence customer perceived value of an organization (Yoon & Lee, 2007). A study examining the factors of word of mouth effectiveness from a receiver’s perspective found that receivers are likely to be influenced by the word of mouth message depending on the nature of the sender-receiver relationship, richness and strength of message among other factors (Sweeney, Soutar, & Mazzarol, 2008). Notwithstanding, other studies suggest that receiver’s perception of a business’ country of origin can affect the perceived value of the organization irrespective of richness of word of mouth messages. The extent of dispersion of word of mouth conversations is important to the way the message influences customer perception of business value. Studies have shown that individuals are increasingly taking to social media and blogging sites to inform others of service experiences thus increasing the dispersion of electronic word of mouth communications (Kucukemiroglu & Kara, 2015). Also, the study noted that firms realized that consumers seek to associate with firms that are socially acceptable hence seek firms whose messages expressed through electronic WOM is well known or dispersed (Magnini, 2011). Notwithstanding, other studies have suggested that although word of mouth messages may be widely dispersed, what actually influences customer perceived value is the most recent word of mouth message that reaches the prospective customer before the service encounter. This means that although a positive word of mouth message can be widely dispersed some customers may not attach high value, purchase or associate with the organization if the most recent information or messages reaching the customer is negative.

The manner of delivery of word of mouth messages can indicate the strength of the message and by extension the strength of its influence on the receiver. Also, strong and significant relationship was found in the case of strength of delivery of negative word of mouth and perceived value although its impact was not as strong as positive word of mouth (Sweeney, Soutar, & Mazzarol, 2012). This means that while positive and strongly delivered word of mouth can result in positive perception of value of a restaurant, negative word of mouth delivered strongly can have the opposite effect. However, other studies suggest that while manner of delivery of word of mouth is important other factors such as richness of content and reputation of the messenger are critical to the persuasiveness of word of mouth messages (Wang, Sun, & Peng, 2013).

2.3. Conceptual framework

Although much literature has been written on the word of mouth communication, most of these studies centred on Western countries with very limited application to dissimilar cultures such as found in Malaysia. Most of the research conducted on word of mouth did not consider the reserved and non-confronting disposition of Malaysians. Thus, there is a need to develop a new framework that is applicable to the Malaysian context. Hence, this research fills this gap in the literature by studying the impact of word of mouth on perceived customer value in the Malaysian context. The main factors of word of mouth (independent variable) discussed in this research include frequency of message, reputation of messenger, richness of message, dispersion of conversations and manner of message delivery. It is shown below as follows:

Figure 1 indicates that frequency of message, reputation of messenger, richness of message, dispersion of conversations and manner of message delivery are the independent variables while student customer perceived value is the dependent variable. The conceptual framework involves determining the impact or effect of the independent variables on customer perceived value of restaurants in Malaysia.

3. Research Methodology

Research Design: Quantitative research method is used in this causal research (also called explanatory research). This research is designed as causal research because it is seeking to establish a causal relationship between word of mouth and customer perceived value and test the extent to which word of mouth influences customer perceived value in the Malaysian restaurant industry (Saunders et al, 2009). Moreover, a quantitative causal research design allows for rigorous statistical analysis to be applied to the research data (Saunders et al., 2007). Sample size:Three restaurants have been chosen for this study. All selected restaurants are strategically located and accessible to local Malaysians and serving variety of Malaysian food. Selected restaurants frequently facilitate the use of word of mouth in their advertisement strategies. For example, when incentives are announced, customers are asked to inform friends and family members. Thus, it will provide a more representative assessment of customer perceived value of restaurants in selected geographical area. The sample size of this research is 150. This sample size is suitable for this research because a study that investigated the complexities of the WOM concept, the triggers that motivate people to offer WOM and the conditions that enhance the chance of WOM occurring used 100 responses Mazzarol, Sweeney, & Soutar, 2007) while an internet study that investigated the influences of electronic word of mouth message appeal and message source credibility on brand attitude involved surveyed 211 subjects (Wu & Wang, 2011).Sampling Technique: Convenience sampling is chosen as the sampling method for this research. This is because convenience sampling allows the researcher to meet customers that are easiest to find in the restaurant area (Saunders et al., 2007). Also, convenience sampling is suitable because the total number of restaurants in Kuala Lumpur is unknown (difficult to find exact number) which means probability sampling methods will not be suitable. Thus, restaurant customers are approached when leaving selected restaurants as agreed by the restaurant management and asked to participate in the research by filling out the research instrument.

Data Collection Method: Questionnaires is used to collect primary data. A questionnaire is mainly selfadministered and responses collected on the spot. The questionnaire was made up of personal profile questions and closed-ended questions that apply a 6-point Likert scale. Instrument Design: The questionnaire is designed with five short direct statements about each the independent variables and dependent variable based on the conceptual framework. Data of participant profile was also collected consisting gender, frequency of restaurant visits, age, monthly income, occupation, food preferences and reason to eat out and mode of travel. Data Analysis Plan: Raw data from the questionnaire is checked for completeness before coded using Statistical Package for Social Science (SPSS) software, version 22. Thereafter, descriptive analysis is carried out on ordinal and nominal data (mainly for participants’ profile) using frequency tables while descriptive analysis on ordinal data (Likert-scale type questions) is carried out using mean and standard deviation. Statistical analysis is carried out on ordinal data to test the hypothesis and frame a decision. Also, multiple regression analysis is used to determine the extent to which the independent variables (frequency of message, reputation of messenger, richness of message, dispersion of conversations and manner of message delivery) impact customer perceived value of restaurants in Malaysia. Validity, Reliability and Pilot test: This research satisfies measurement validity as the same questionnaire is presented to all respondents which increases objectivity and eliminates bias. External validity is satisfied as the research sample includes customers which mean that the research findings can be generalised to other restaurants in Kuala Lumpur. Moreover, reliability bothers on the extent a research can be repeated in the future (Saunders, 2007). In terms of reliability, a pilot test is carried out to ensure the questionnaire is reliable.

The questionnaire is administered tao 20 students and the reliability score represented by the Cronbach Alpha statistic is determined. The reliability of the overall questionnaire is 0.933 which indicates a high degree of reliability of the questionnaire as a data collection instrument. Individual factor reliability measurements show that frequency of message, reputation of messenger, richness of message, dispersion of conversation, manner of delivery and customer perceived value has a reliability score of 0.828, 0.697, 0.844, 0.789, 0.807 and 0.666 respectively. This indicates the questionnaire measures the intended variables overall and measures the factor constructs at an acceptable level of internal consistency thus, reliable. Ethical Considerations: Some ethical issues include confidentiality, privacy and anonymity. Privacy issues are addressed in this research as restaurant patrons will not be forced to participate in the survey. Confidentiality issues are addressed in this research as the collected research data is used only for research purposes. Anonymity is addressed in this research as personal identifiable information of respondents will not be collected. Thus, this research meets acceptable ethical requirements.

4. Results

This section shows the sample results, analysis and interpretation. It includes descriptive and statistical analysis of collected data. Assumptions for using the chosen statistical technique are discussed.

4.1. Descriptive analysis of participant profile

Approximately 40 percent of respondents were females while 60 percent were males. Out of all, 32 percent visit the restaurant twice to four times per week while 31 percent visit the restaurant five times to nine times per week representing the two top groups by frequency of visits. The largest groups of respondents were between 18 years and 25 years of age and between 25 years to 29 years of age with each group making up approximately 33 percent of the respondents. In terms of earning, the largest group of respondents earned RM 3000 or less per month making up approximately 57 percent of the respondents. The two largest groups of respondents were students and business people by profession with students making up approximately 33 percent while business people making up 19 percent of the respondents. Approximately 29 percent preferred Malay menu while approximately 25 preferred western menu making up the top two categories by food preference. The two largest groups that visited the restaurant did so for time-savings and habit reasons with time-savers making up approximately 37 percent while habitual patrons making up 30 percent of the respondents. The largest group of respondents transported to the restaurant by car making up approximately 61 percent of the respondents.

4.2. Descriptive analysis of the responses

Frequency of messages factor: Considering only whole numbers of the mean, respondents preferred restaurants that are broadcasted to the public (Mean=4.01, Std. Dev=1.196), think a positive buzz increases customer value perception of the restaurant (Mean=4.14, Std. Dev=1.187), would generally request more information about a new restaurant after hearing praises about it (Mean=4.05, Std. Dev=1.161) and prefer restaurants whose services are communicated in various languages (Mean=4.29, Std. Dev=1.183). However, respondents did not agree being attracted to restaurants that are regularly praised by friends (Mean=3.87, Std. Dev=1.359). Since the standard deviation on all the items is greater than 1, it cannot be argued that respondents did not form opposite opinions on the measured items.

Reputation of messenger factor: Considering only whole numbers of the mean, respondents agreed that restaurant recommendations from reputable sources influence restaurant value perception (Mean=4.23, Std. Dev=1.232), preference to visit restaurants regulated by governmentapproved agencies (Mean=4.21, Std. Dev=1.229), would rarely visit unknown restaurant except recommended by someone trusted (Mean=4.15, Std. Dev=1.239), would visit a restaurant if the person recommending can provide proof (Mean=4.09, Std. Dev=1.198) and would be willing to pay more at restaurants recommended by credible sources (Mean=4.16, Std. Dev=1.1342). However, since the standard deviation on all the items is greater than 1, it cannot be argued that respondents did not disagree on the measured items.

Richness of message factor: Considering only whole numbers of the mean, respondents agreed on looking for specifics in any restaurant recommendations before visiting the restaurant (Mean=4.03, Std. Dev=1.231) and that detailed word of mouth message should be the basis for restaurant choices (Mean=4.27, Std. Dev=1.215). However, respondents disagreed on considering nationality biases in restaurant recommendations before visiting the restaurant (Mean=3.79, Std. Dev=1.329), not visiting a restaurant when friends do not have enough details about it (Mean=3.73, Std. Dev=1.1258) and not visiting a restaurant based on one line messages online (Mean=3.95, Std. Dev=1.228). However, since the standard deviation on all the items is greater than 1, it cannot be argued that respondents did not form opposing views on the measured items.

Dispersion of conversations factor: Considering only whole numbers of the mean, respondents preferred visiting restaurants that people are talking about (Mean=4.28, Std. Dev=1.296), like restaurants that are famous (Mean=4.28, Std. Dev=1.275), prefer restaurants whose website include different testimonials from past customers (Mean=4.17, Std. Dev=1.140) and prefer restaurants with widely dispersed information reach (Mean=4.26, Std. Dev=1.201). However, respondents did not agree on using online recommendations in making restaurant choices even when the restaurant is praised (Mean=3.61, Std. Dev=1.365). Since the standard deviation on all the items is greater than 1, it cannot be argued that respondents did not form opposite opinions on the measured items.

Manner of delivery factor: Considering only whole numbers of the mean, respondents liked visiting a restaurant friends talk about passionately (Mean=4.38, Std. Dev=1.224), agreed on scrutinising restaurant recommendations by sponsored agents (Mean=4.07, Std. Dev=1.201), preferred unfamiliar restaurants provided people talk about it excitedly (Mean=4.35, Std. Dev=1.111), would visit a restaurant if people praised it with menu pictures online (Mean=4.45, Std. Dev=1.138) and agreed that if people do not praise a restaurant passionately, it reflects on the restaurant’s level of service (Mean=4.19, Std. Dev=1.358). However, since the standard deviation on all the items is greater than 1, it cannot be argued that respondents did not disagree on the measured items.

Customer perceived value: Considering only whole numbers of the mean, respondents agreed that frequency of word of mouth messages (Mean=4.15, Std. Dev=1.241), reputation of word of mouth messenger (Mean=4.33, Std. Dev=1.168), detailed messages recommending a restaurant (Mean=4.45, Std. Dev=1.053), popularly known restaurants (Mean=4.50, Std. Dev=1.163) and restaurants people are excited to recommend (Mean=4.57, Std. Dev=1.113) directly influences positive customer perceived value of the restaurant. However, since the standard deviation on all the items is greater than 1, it cannot be argued that respondents did not disagree on the measured items.

4.3. Sample Adequacy & Normality Test

As the sample adequacy statistic of 0.823 which confirms that this research sample is suitable and adequate for research analysis. Skewness and Kurtosis can be used to test for normality of the research observations. Thus, kurtosis values between -10 and +10 and skewness values between -3 and +3 can be acceptable for assuming normality in social science research (Hair et al, 2010). Skewness and kurtosis values all fall within acceptable limits for this research. Therefore, it can be seen this research sample is normally distributed.

4.4. Hypothesis Statements

[Hypothesis one] Null (H₀): Frequency of message does not significantly influence customer perceived value of restaurants in Malaysia. Alternative (H_A): Frequency of message significantly influences customer perceived value of restaurants in Malaysia.

[Hypothesis two] Null (H₀): Reputation of messenger does not significantly influence customer perceived value of restaurants in Malaysia. Alternative (H_A): Reputation of messenger significantly influences customer perceived value of restaurants in Malaysia.

[Hypothesis three]Null (H₀): Richness of message does not significantly influence customer perceived value of restaurants in Malaysia. Alternative (H_A): Richness of message significantly influences customer perceived value of restaurants in Malaysia.

[Hypothesis four] Null (H₀): Dispersion of conversations does not significantly influence customer perceived value of restaurants in Malaysia. Alternative (H_A): Dispersion of conversations significantly influences customer perceived value of restaurants in Malaysia.

[Hypothesis five] Null (H₀): Manner of delivery does not significantly influence customer perceived value of restaurants in Malaysia. Alternative (H_A): Manner of delivery significantly influences customer perceived value of restaurants in Malaysia.

4.5. Decision criteria

Significance (alpha) level: 0.05 (95 percent confidence level). Reject all null hypotheses if the significance alpha score is less than 0.05 and accept the alternative hypothesis.

4.6. Multiple Regression Analysis

The significance value in the regression modal fit is less than the alpha value of 0.005 (F: 5, 144 = 15.612, p < .005. Hence, the regression model has significant predictive capacity thus, a good fit for the data. As per regression modal summary, the R-Square value 0.352 means that the independent variables (frequency of message, reputation of messenger, richness of message, dispersion of conversation and manner of delivery) explain 35.2 percent of variability of the dependent variable (customer perceived value).

In order to ensure all the variables (independent and dependent) of this research actually measure different constructs, the correlation coefficients between all the variables should be less than 0.7 since correlations greater 0.7 may indicate strong correlation between the variables which means that the variables are measuring the same thing or construct (Hair et al, 2010). The row “Pearson correlation” shows that all the correlations between the dependent variable “customer perceived value” and independent variables and between the five independent variables are less than 0.7. For example, the correlation between customer perceived value and frequency of message is 0.244, reputation of messenger (0.404), richness of message (0.403), dispersion of conversation (0.369) and manner of. Also, the correlation between “manner of delivery” and customer perceived value is 0.569, frequency of message is 0.267, reputation of messenger (0.511), richness of message (0.600) and dispersion of conversation (0.441). All correlations measuring 1.000 are ignored as it indicates the correlation between the same variable. Since all the variables have correlations with each other less than 0.7, it can be said that the variables sufficiently measure different constructs for regression analysis results to be valid.

The significance column shows that only ‘manner of delivery’ factor has statistically significantly relationship to customer perceived value (p-value<0.05). Thus, manner of delivery can be said to be a factor that is a statistically significant predictor of customer perceived value in this model. In other words, manner of delivery in word of mouth communication significantly impacts customer perceived value of restaurants in Malaysia. The column ‘VIF’ with values less than 5 and ‘Tolerance’ with values above 0.1 shows no multi-collinearity issues with the data (meaning there is no two or more independent variables that are highly correlated) (Hair et al, 2010).

4.7. Regression Equation

Based on Table14, the regression equation for this research is stated below:

Y = 1.310 +0.043X₁ + 0.107X₂ + 0.030X₃ +0.091X₄+ 0.461X₅+ 0.836

Where, Y = Customer perceived value, X₁ = Frequency of message, X₂ = Reputation of messenger, X₃ = Richness of message, X₄ = Dispersion of conversation, X₅ = Manner of delivery, e = total error = 0.836

(0.396+0.081+0.090+0.089+0.088+0.092), Constant = 1.310

Hence, it can be observed that a unit increase in manner of delivery of word of mouth messages will deliver the highest impact (0.461 or 46.1%) or increase on customer perceived value in this model.

4.8. Discussion on regression analysis

Table 3 shows that one null hypothesis was accepted associated with “manner of delivery” factor while four null hypotheses was not accepted associated with “frequency of message”, “reputation of messenger”, “richness of message” and “dispersion of conversation” factors. The regression analysis shows that the regression model made up of the variables: frequency of message, reputation of messenger, richness of message, dispersion of conversation and manner of delivery are good predictors of customer perceived value of restaurants in Malaysia. However, these variables only explained 35.2 percent of the variances in the dependent variable which means that other factors outside the five independent variables of this research also contribute to customer perceived value of restaurants in Malaysia. Moreover, since “manner of delivery” is a statistically significantly factor in the regression model, it means that customers perceived value of a restaurant will increase when the manner of delivery of word of mouth communication is more effective.

[Table 1] Correlations Coefficients

[Table 2] Regression Coefficients

[Table 3] Summary of Hypothesis testing

5. Findings

Frequency of word of mouth message did not significantly influence customer perceived value in Malaysia. A possible reason could be customers’ underestimating the frequency of message received. This means that when customers underestimate the frequency of WOM message, a high frequency message can become a low frequency message that fails to significantly influence customer perceived value of restaurants in Malaysia especially when the messages are not from the Malaysian’s circle of friends. Reputation of word of mouth messenger did not significantly impact customer perceived value of restaurants in Malaysia. A possible reason could be customers’ subjectivity in the customer’s assessment of the reputability of WOM message. This means that if one customer assesses a WOM message as reputable another customer may assess it as not reputable. Consequently, if a large proportion of customers perceive WOM as coming from a perceived less reputable source when others perceive it highly, an otherwise reputable WOM source may not significantly impact customer perceived value of restaurants in Malaysia. Richness of message did not significantly impact customer perceived value of restaurants in Malaysia. A possible reason could be customers’ subjectivity in the customer’s assessment of richness of WOM message. This means that if one customer assesses a WOM message as rich another customer may assess it as not rich enough. Consequently, if a large proportion of customers perceive WOM message as vague when others perceive it as rich, an otherwise rich WOM message may not significantly impact customer perceived value of restaurants in Malaysia. Dispersion of word of mouth conversations did not significantly influence customer perceived value of restaurants in Malaysia. A difficulty in distinguishing real and faked restaurant agents may make it difficult for restaurant patrons to increase the perceived value of a restaurant as a result of not being convinced of the credibility of the agency. In Malaysia, this can especially be true if such agency is online and its message is unsolicited and outside the person’s close associates. Manner of delivery was found to significantly impact customer perceived value of restaurants in Malaysia. Statistical significance of manner of delivery factor in predicting customer perceived value of restaurants in Malaysia may be traced to increasing availability of and reliance on offline and online social networks for purchase recommendations and availability and easy internet access.

6. Conclusions

The statistical analysis revealed that manner of delivery factor significantly impacts customer perceived value of restaurants in Malaysia. The possible reasons discussed as being responsible for this research findings is traced to increasing availability of and reliance on offline and online social networks for purchase recommendations and availability and easy internet access that have made it possible for people to form social groups where information is shared from personal to business and purchase recommendations. Also, increasing reliance on these offline and especially online social networks would contribute to an individual’s restaurant choice since a friend’s passionate and excited recommendation in one’s social group can easily encourage others to try out the restaurant. Important implications found from this research include the need for restaurant marketers engaging in word of mouth campaigns to be passionate and excited about the restaurant offerings as Malaysian customers will significantly respond to restaurant offerings communicated in an excited and passionate manner. Moreover, commercial marketing organizations in Malaysia contracted by restaurants to implement word of mouth marketing campaigns should employ high emotional appeal in its campaign delivery in order to successfully attract customers to Malaysian restaurants. Finally, restaurant managers and marketing organizations can spend less (not absence of) resources and effort on other factors of word of mouth communication such as frequency of message, reputation of messengers and dispersion of message in order to maximise the returns on implemented word of mouth campaigns. The research recommends that restaurant managers in Malaysia engaging in word of mouth campaigns should encourage marketers to be passionate and excited about the restaurant offerings as Malaysian customers will significantly respond to restaurant offerings communicated in an excited and passionate manner. Emotional appeal should be highlighted in any word of mouth implementations by restaurant managers and marketing organizations to ensure marketing and business success.