The moderating effect of brand image on public relations perception and customer loyalty

The Authors

An-Tien Hsieh, Da Yeh University, Taiwan, Republic of China

Chung-Kai Li, Ling Tung University, Taiwan, Republic of China

Abstract

Purpose – The purpose of the paper is to investigate the relationship between customer perceptions of public relations (PR) and customer loyalty; to test for the moderating role of brand image in that relationship.

Design/methodology/approach – Data were collected in a survey of customers of the insurance industry in Taiwan, using a questionnaire designed on the basis of focus-group discussions with 30 consumers. Hierarchical regression analysis of data from 367 respondents was used to test two hypotheses.

Findings – The results show that consumers' perception of an organisation's PR practice is an antecedent of loyalty. The impact of public relations perception (PRP) on customer loyalty is stronger and more significant when the brand image is favourable. If it is unfavourable, the effect of PRP on customer loyalty is negligible.

Research limitations/implications – This study extends previous research by examining the moderating role of brand image. Further research is indicated, to identify the key moderators of the driving force of PR in relation to customer relationship marketing.

Originality/value – This paper proposes an original eight-item scale for the assessment of customer PRP activity, which can be applied in practice to measure its effectiveness under different brand-image conditions.

Article Type:

Research paper

Keyword(s):

Public relations; Customer relations; Brand image; Customer loyalty; Insurance services; Taiwan.

Journal:

Marketing Intelligence & Planning

Volume:

Number:

Year:

2008

pp:

26-42

Copyright ©

Emerald Group Publishing Limited

ISSN:

0263-4503

Introduction

How to develop, maintain, and enhance customer loyalty toward a firm's products or services is generally seen as the central thrust of marketing activities (Dick and Basu, 1994). Higher customer loyalty implies a higher market share and an ability to demand relatively higher prices compared to those of competitors (Chaudhuri and Holbrook, 2001). This increased customer loyalty can also help lower marketing costs, solicit more customers, and effectively operate trading leverage (Aaker, 1997). Additionally, loyal customers foster positive word-of-mouth promotion, defy competitors' strategies (Dick and Basu, 1994) and generate higher corporate profits (Fornell and Wernerfelt, 1988; Reichheld et al., 2000). Loyalty factors are an organization's most reliable success indicators (Zeithaml et al., 1996). However, what are the factors influencing consumer loyalty? The antecedent factors of customer loyalty have been discussed extensively in numerous works, but the relationship between public relations (PR) and customer loyalty needs further work because environmental changes have led businesses to concentrate as much on societal orientation (Kitchen, 1996) as on the traditional orientation of product and marketing. This in turn highlights the increased importance of PR.

Cultip et al. (1985) defined PR as “the management function that identifies, establishes, and maintains mutually beneficial relationships between an organization and the various public on whom its success or failure depends.” Moreover, according to a summary of empirical results from several studies, it is consumer perception of these organization-public relationships that influences customer satisfaction evaluations, behavioural intent and actual behaviour. Such a result has been shown for school-student relationships (Bruning, 2002; Bruning and Ralston, 2001; Hon and Brunner, 2001) and bank-customer relationships (Bruning, 2000; Bruning and Ledingham, 2000) and indicates that customers that stay possess a higher level of PR perception and satisfaction compared with those that leave. Ledingham and Bruning (1998) note that the consumer's awareness of organization-customer relationships could raise their loyalty toward the corporation, which would in turn increase company income, enhance market share, and achieve other corporate objectives. Coombs (2001) also proposes that when corporations have plans to cultivate PR and fulfil commitment, consumer's loyalty to corporations would be higher. Although past scholars have not produced consistent findings on the dimensions of establishing relationship quality (Wulf et al., 2001), empirical results indicate that the relationship between corporations and customers is increasingly intimate, and this would benefit the enhancement of customer loyalty in the long run (Crosby et al., 1990; Morgan and Hunt, 1994).

However, previous research (Brown and Dacin, 1997; Ellen et al., 2000; Sen and Bhattacharya, 2001) has not produced consistent results regarding the factors that positively affect consumer perception of public relations (PRP) practice. Some studies show that the inconsistent results might be caused by a moderator such as brand association. However, how existing brand image especially negative brand image impacts on the effectiveness of PR has received very little attention. To find out, it is necessary to take into account the separate effects of positive and negative brand images when analysing PR results.

Take, for example, the actions of tobacco giant Philip Morris. The company spent $75 million on its charitable PR contributions in 1999, and then launched a $100 million advertising campaign to publicize them. Not surprisingly, there are genuine doubts about whether such approaches actually work or just breed public cynicism about the company motives (Cone et al., 2003; Porter and Kramer, 2002). In the above case, consumers' unfavourable image of the firm caused PR to have a negative effect on their attitude. It can be suggested, therefore, that brand image played a key moderator role.

Consequently, the purpose of this research study is to clarify the relationship between PRP, brand image and customer loyalty. We first develop a conceptual model with supporting research hypotheses. Next, we describe the research method. Finally, the study findings are presented, and the managerial and research implication are discussed.

Literature review and hypotheses

Figure 1 shows the research model underlying our study. The model begins with customer PRP provided by an insurance company. PRP is presented as an antecedent to customer loyalty. Brand image is investigated as having a potential moderation role between PRP and customer loyalty relationships. We present below the literature review and the hypotheses developed.

Keller (2003) reported that in an increasingly networked economy, understanding the consumers' tendency to link a brand to other entities such as a person, place, thing, or brand is crucial. In terms of linking a brand to a product, PR strategy can enhance brand knowledge and establish brand awareness through recall and recognition. PR can further enhance the brand associations of brand image, draw brand emotions, and create brand attitude and experience. As customer loyalty is often viewed as resulting from brand knowledge (Keller, 1993, 1999), it follows that PR can raise consumer loyalty through the above strategies.

In addition, the self-congruence theory states that the ways consumers evaluate products match their self-image (Belk, 1988; Sirgy, 1982). Consequently, a higher congruence between self-image and product image would influence consumers' attitudes or behaviour regarding brand preference, brand attitude, product purchase decisions, customer satisfaction, and repurchasing intentions (Ekinci and Riley, 2003; Graeff, 1996; Hong and Zinkhan, 1995; Sirgy, 1985). In contrast, in the case of incongruence, consumers could refuse or avoid purchasing products and services, which transmit negative intrinsic messages about themselves (Hogg and Banister, 2001). On the other hand, a person's self-concept is comprised of a number of self-identities, each varying along a continuum ranging from personal identity at one end to social identity at the other (Tajfel and Turner, 1985). When social categorisation is made salient through PR, consumers' demand for social-identity increases (Mardigal, 2001), leading to higher consumer self congruence.

PR usually strive to create two kinds of brand associations (Hoeffler and Keller, 2002):

user profiles, where customers (users) are portrayed as kind and generous, and as doing good things; and
brand personality, where the sincerity dimension of brand personality is stimulated (Aaker, 1997).

Consumers perceive the people behind the brand as caring and genuine. Conversely, consumers' reaction to PR depends on their levels of self-congruence and how their expectations match the company's properties presented through PR efforts (Sen and Bhattacharya, 2001). If consumers' self-concept and perception of characteristics, associations and values are identical to the company conducting PR activities, consumers attach a higher degree of acceptance of the company. The phenomenon of high self-congruence implies that consumers' values and beliefs are mixed with those of the company (Dutton et al., 1994). Higher consumer self-congruence enhances the establishment of commitment and meaningful relationships with the organization and intensifies customer loyalty. Therefore, organizations' PR practice can raise consumers' self-congruence; the higher the self-congruence, the higher the consumer loyalty. Thus, the first hypothesis is as follows:

H1. Customer PRP will be positively related to customer loyalty.

Cognitive psychologists believe that memory is extremely durable, so that once information becomes stored in memory its strength of association decays very slowly. Therefore, if consumers have committed a favourable brand image to memory, such a perception is expected to guide the integration of new information (Petty and Cacioppo, 1986). This is also called the halo effect; there are two main types of halos in marketing literature. The first type states that the response to a particular attribute can be influenced by the general impression of the overall object (Beckwith and Lehmann, 1976). The second type asserts that the evaluation of a dominant attribute can influence the response to other attributes. When consumers have a favourable brand image, this could lead to a positive influence over other messages related to the brand. In other words, the halo effect reflects the individual's tendency to maintain cognitive consistency and to avoid cognitive dissonance (Beckwith and Lehmann, 1976; Holbrook, 1983; Wirtz and Bateson, 1995).

When brand image is favourable then proactive PR perceived by customers will be more congruent to corporate reputation, consumers are more able to maintain positive beliefs, attitudes and behaviours. As a result, favourable brand image can enhance the PR effects and boost customer loyalty.

On the contrary, when consumers commit an unfavourable brand image to memory, the organizations' aggressive pursuit of PR activities can conflict with the consumers' perceptions about the corporate' reputation. Inconsistencies in the cognitive system lead to consumer nervousness, and consumers develop a psychological tendency of balanced differences (Leuthesser et al., 1995). At this time, consumers attempt to identify a reasonable explanation for this phenomenon. The attribution theory, for example, describes how people make causal inferences about the behaviour of others. As noted, consumers may use aspects of an offer to make inferences about the motives of the company. These inferences may in turn affect their evaluations of the offer. The theory predicts a relationship between attributions and subsequent attitudes and behaviours (Dean, 2002). In other words, consumers' causal attributions of PR motivation can influence their attitudes or behaviours. Unfavourable brand image leads to a negative halo effect and also negatively influences other brand associations (Wirtz and Bateson, 1995).

Those results could be due to the consumers' lack of trust in organizations, which can lead to consumers being sceptical about the sincerity of the organizations' actions (Coombs, 2001). Consumers also explore the motivation behind PR. If a consumer perceives the organizations' motives are egoistic or that PR is a means to attain profits or avoid punishment, this impression will lead to harmful results for the organizations' reputation (Sallot, 2002). Similar results were found in a qualitative study by Webb and Mohr (1998). They discovered that about 50 per cent of the respondents believed that organizations have selfish motives in conducting PR initiatives; the remaining half believed that organizations' motives are simultaneously selfish and altruistic.

A negative attitude can also arise from scepticism over an organization's honesty and fairness in executing PR activities. Sceptical consumers doubt the integrity and fairness of organizations' PR initiatives, and tend to distrust organizations. In this case, purchasing behaviour is not influenced by PR activities. Unfavourable brand image reduces the positive effects brought about by PR, and thus the contribution toward raising consumer loyalty is negligible. Therefore, the second hypothesis of this study can be stated as follows:

H2. The relationship between PRP and customer loyalty is moderated by brand image, and the tendency for PRP to be positively related to customer loyalty will be significantly more pronounced when brand image is high rather than low.

Research method

Sample and data collection

Samples for this study were taken from consumers in Taiwan with actual experiences in purchasing insurance. The study focused on the insurance industry because Wee et al. (1996) suggested that the service industry and particularly insurance is more concerned with and would be more willingly to participate in PR activities than the manufacturing industry. In this research we take the broad view that an unfavourable brand image is generally considered to be undesirable for the brand to be associated with, from the consumers' perception. Adopting convenience sampling, we asked consumers to respond to questions according to the insurance companies they are familiar with. Finally, 203 valid questionnaires were collected indicating companies with a good brand image and 164 indicating companies with a poor brand image. The total sample of 367 represented an effective response ratio 61.17 per cent, with male (n=135) accounting for 36.8 per cent of the total sample. The majority of respondents (41.5 per cent) were 20-30 years old, while those between 30 and 40 years old made up 35.2 per cent of the total sample. The study used a seven-point Likert scale, with scale anchors ranging from “1” (strongly disagree) to “7” (strongly agree), all scale items are listed in the Appendix.

Measures

PRP is the independent variable of this study. Perception is defined as the process by which an individual selects, organizes and interprets stimuli into meaningful and coherent pictures of the word. Therefore, this study defines PRP as “consumers' overall perception of corporations' dedication of time, efforts, and resources to public relations practice.”

The scale employed was developed by the authors through a process as follows (Devellis, 1991):

items were first collected through interviews with practitioners of professional PR, and 14 items were generated;
we pre-tested our questionnaire with a convenience sample of 30 consumers; and
a scale validation procedure was established using item analysis, item-to-total correlation, and exploratory factor analysis (EFA).

In the end, eight items remained in the PRP scale.

Customer loyalty is the dependent variable for this study. Related literature provided different definitions and associated measurements for customer loyalty. Dick and Basu (1994) suggested that loyalty is influenced by related forces of attitude and behaviour. This study operationally defines customer loyalty as:

… a deeply held commitment to re-buy or re-patronise a preferred product or service consistently in the future, thereby causing repetitive same brand set purchasing, despite situational influences and marketing efforts having the potential to cause switching behaviour (Oliver, 1999).

This study measured customer loyalty using a five-item scale adapted from Zeithaml et al. (1996).

Brand image was used as a moderator variable for this study. Keller (1993) defined brand image as a set of perceptions about a brand as reflected by the brand associations held in consumer memory. However, a consensus and immutable definition of brand image has not been established. We measured brand image using a three-item scale adapted from Park et al. (1986), which covered functional benefits, symbolic benefits, and experiential benefits. Functional needs refer to the intrinsic features possessed by the product when consumers attempt to solve purchasing decisions; symbolic needs are related to consumers' self-concept and whether the product could satisfy self-esteem needs; and experiential needs refer to issues of stimulation, sensory pleasure, or novelty linked to products.

Control variables

We controlled for several variables to rule out alternative explanations. Past literature has cited control variables like sex (male = 1, female = 0) (Ndubisi, 2006), age and disposable income (Homburg and Giering, 2001), all of which we used. We also included altruism, which refers to the behaviour of reducing personal consumption for the sake of increasing others' consumption. Altruism is generally seen as positive but does not always denote self-sacrifice; instead it sometimes refers to a form of benevolence toward others. This study defines altruism as a form of utilitarianism, with the purpose of helping others or considering others' welfare. Webb and Mohr (1998) and Sallot (2002) both pointed out that altruism influences consumers' judgments of PR and loyalty. This study measured altruism using a five-item scale adapted from Webb and Mohr (1998). Coefficient α was 0.9278.

Reliability and validity

In order to test the validity and reliability of each concept, a three-step approach to data analysis was adopted in this study:

exploratory factor analysis;
confirmatory factor analysis; and
reliability analysis.

We first conducted an EFA, using principal component analysis with varimax rotation. A factor loading of greater than or equal to 0.5 was used as the benchmark to include items in a factor. The percentage of the variance explained for PRP was 63.404 per cent, that for brand loyalty was 90.299 per cent, and that for brand image was 85.779 per cent (Table I). In social science, it is common to consider a solution that accounts for 60 per cent of the total variance as satisfactory (Hair et al., 1998). We then conducted a confirmatory factor analysis (CFA) to determine the construct validity. In Table I, the chi-square [χ ² (df)=268.231(93), p<0.001] was not satisfactory which was expected given the sample size; other indices [GFI = 0.912, RMSEA = 0.072, χ ²/df = 2.884, AGFI = 0.87, NFI = 0.96, CFI = 0.97, IFI = 0.97] indicated that the model fits the data reasonably well. Convergent validity was assessed by examining the indicator loadings. Table I shows that all standardized loading of the manifest variables on the indicators in the CFA were significant (p<0.001). Thus, convergent validity was supported.

A common test of discriminant validity is determining by testing additional models that constrain the association between two latent constructs to 1 and using a χ ² test of the difference in fit between the model with the unconstrained association and the model with the constrained association (Anderson and Gerbing, 1988; Chiou and Droge, 2006; Smith and Barclay, 1997). Discriminant validity is inferred if the model with the constrained association provides a significantly worse fit to the data than the model with the unconstrained association; in all three cases, Δχ _1,0.05 ²>3.84 and the overall fit significantly decreased. Thus, discriminant validity was supported.

The reliability of multi-item scales was determined by computing Cronbach αs. All scales have acceptable α values ranging from 0.9136 to 0.9735 (Table I).

Analysis and results

We tested the hypotheses using hierarchical regression. As shown in Table II, in step 1, four control variables, namely altruistic motives, gender, age, and disposable income, were entered into the regression. Overall, the model accounted for 37.5 per cent of the variance in customer loyalty. Altruistic motives and disposable income were significantly related to customer loyalty (β=0.602 and 0.092, p<0.05, respectively). In step 2, the independent variable, PRP, was added to the regression, the overall explanation power increased by 6.8 per cent. PRP has positive and significant influences on customer loyalty (β=0.305, p<0.01). This result supports H1 of this study. This result indicates that the higher the level of consumer perception of an organization's dedication to PR activities, the more willing they are to purchase the organization's products and to recommend those products to friends and relatives.

H2 postulated that the impact of PRP is contingent on the level of brand image. Brand image moderated the relation between PRP and customer loyalty. To test the effect of the moderator variable on brand image, this study performed a split-group analysis. The sample was initially arranged in ascending order of brand image but split along the median to form two groups: favourable brand image and unfavourable brand image (Hsieh and Hsieh, 2001).

Table III indicates that brand image moderated the relation between PRP and customer loyalty. For the favourable brand image group (n=203), the standardized regression coefficient was significant for PRP (β=0.231, p<0.01, R ²=0.133). But for the unfavourable brand image group (n=164), the standardized coefficient for PRP was not significant (β=0.151, p>0.05, R ²=0.122). Results indicate that only the favourable brand image has a positive influence over customer loyalty, while the unfavourable brand image has no significant influence. Therefore, H2 of this study is also supported. This result demonstrates that the effect of PRP on customer loyalty differs according to the acceptance level of brand images. Relative to organizations with unfavourable brand images, in organizations with favourable brand images, the PRP effect is more significant and greater.

Examining potential confounding factors

We further performed a t-test to explore the influence of differences in levels of trust on consumer attitude and behaviour toward different brand images. According to Table IV, means of altruistic motives, trust and customer loyalty are higher in organizations with favourable brand images than in organizations with unfavourable brand images and the t-values are all significant. This result implies that favourable and unfavourable brand images have different degrees of influence on consumers' cognitive power. Therefore, H2 of this study is again supported.

Discussion

This study contributes to helping to explain the influence of PRP on customer loyalty by investigating whether attitudes towards brand image affect this relationship. It was found that the effect of PRP on loyalty is stronger and more significant when brand image is favourable.

In the first research problem, we find that PRP has a positive influence on customer loyalty. However, whether PR is a form of strategic relationship management (Ferguson, 1984) or a tactical publicity function is still under scholastic debate (Moss et al., 1996). This study is interested in consumers' perception of PR. Empirical results indicate that the higher the consumers' perception of PR, the higher the customer loyalty. In today's highly competitive and rapidly changing environment, how to maintain or even enhance consumer loyalty has become an important success indicator for companies (Zeithaml et al., 1996). Empirical results of this study show that company's dedication to the practice of PR is worthwhile because PR can definitely enhance customer loyalty.

In the second research problem, results indicate that the impact and effect of PRP on customer loyalty varies according to the level of brand image acceptance. Fredericks et al. (2001) suggested that understanding consumer needs has already transcended the issues of products and services, and that the concept of image exploration is of growing importance. Both corporate image and brand image could support or destroy consumers' beliefs about value gained, thereby indirectly influencing customer attitude and behaviour (Abdullah et al., 2000). Hence, brand image may function as a precursor or gatekeeper to more extensive ramifications. When brand image is favourable, this positive impression would enhance the customers' interest in PR information (Poisz, 1989), and consumers would tend towards a positive cognitive, attitude and behaviour toward the company. So PR associated with favourable brand image has an increased effect on customer loyalty with little extra effort.

Relatively, the negative impression associated with an unfavourable brand image could lead to selective attention and distortion of PR information (Poisz, 1989). Consumers might tend to display a negative attitude and behaviour toward the company. Webb and Mohr (1998) noted that half of the respondents in their study had negative attitudes toward companies engaged in PR activities and regarded a company's motives as egoistic. According to Table IV of this study, when brand image is favourable, consumers are willing to believe companies' PR motives are altruistic (mean = 4.94), on the contrary, companies with unfavourable image, consumer doubt their altruistic motives (mean = 3.24). Consumers adopt the attribution theory to deduce the company's motives, and this in turn influences the evaluation of PR activities (Dean, 2002; Sallot, 2002), and further impacts on consumer loyalty. This study extends Webb and Mohr's research on the subject that, “Motives make a difference in how people view public relations.”

From empirical results, we discovered that the level of brand image can influence consumers' trust in companies, which would in turn influence consumers' inferences of PR motives. Sen and Bhattacharya (2001) have stated that consumers believe that PR has trade-off effects. This implies that PR prompts the redistribution of resources. When brand image is unfavourable and company's resources are limited, heavy investment in PR efforts can reduce resources dedicated to product-related activities. Since, PR activities cannot help raise product value, it is possible that consumers would interpret companies as not attending to their proper duties. As a result, PR associated with unfavourable brand image has a decreased effect on customer loyalty despite the extra effort.

Grunig and Hunt (1984) regarded PR as a form of communication management, and the management of brand image is a very important factor determining the success of consumer communications (Kitchen, 1997). For the communication to be effective, trust is particularly critical. However, unfavourable brand image would decrease the level of consumer trust (Swan et al., 1999). As with the Philip Morris case, the company should first establish more consumer trust because current consumers refuse to accept the messages disseminated by the company through charity PR, and that attitude toward the company is rooted in the company's long-term history, politics, social behavioural performance, and product nature (Clark, 2000). PRP can boost the effect of company messages but cannot change its essential value. After building consumer trust, consumers lower their perceived risks and information uncertainty of companies (Morgan and Hunt, 1994); the consumers feel satisfied about results of PR activity and possess greater motivation to realise decisions, and this make communications more effective. Moreover, regular and high-quality communication elevates consumer trust of companies (Farrelly et al., 2003; Geyskens et al., 1998; Morgan and Hunt, 1994; Sharma and Patterson, 1999). The mutual interaction can help accomplish the objective of enhancing consumer loyalty (Garbarino and Johnson, 1999; Sirdeshmukh et al., 2002; Swan et al., 1999).

Implications and limitations

This study has made several contributions to the literature. First, in this study we measured PR from the consumer-oriented perspective. This is in contrast to “craft” PR (Grunig and Grunig, 1992) which measures the effectiveness of PR from a source-oriented perspective (Pinkleton and Austin, 1999). The shortcomings of this approach are:

PR messages may not have influence on consumers' knowledge; and
the influence of the messages might not be positive on consumer's attitude or behaviour.

Second, relationship marketing has been adopted extensively and enthusiastically in “professional” PR (Grunig and Grunig, 1992) literature, but how PR programs affect the quality of relationships is rarely discussed. This study found that customers' overall perception of PR practice can enhance one of the relationship marketing dimensions (namely trust) and influence customer loyalty, so the authors suggest the co-existence of a tactical and strategic approach in PR practice. Third, this study could enrich a body of literature in customer loyalty by adding to the small number of studies examining moderators of the PR and customer loyalty.

Managerial implications for the practice of PR have emerged from the study. In today's globally competitive market, insurance companies have to seek ways to enhance the company's brand image and to gain customer loyalty. This study clearly proves that customer loyalty is quantitatively and qualitatively influenced by PRP, however the effect varies widely, and the halo effect of brand image appears to carry over to the evaluation. When brand image is favourable, a company's disseminated messages have higher congruence with consumer cognitions so company objectives are easier to accomplish. In contrast, when brand image is unfavourable, a company's messages differ greatly from consumer cognitions and consequently the company's objectives are more difficult to achieve. Although in general a company's investments in PR activities enhance company-consumer relationships, the motives behind such dedication can prompt consumer scepticism and can reduce PR credibility. Larsson (2007) found that the trust in PR agents is very low; therefore the trust to company becomes an important factor for consumers in appraising PR motives (Sallot, 2002). Moreover, the establishment of a good brand image helps with building consumers' trust in evaluating company's PR motives which are beneficial in elevating future organization-public relationships and customer loyalty.

Based on extant research about how to obtain a favourable brand image, it appears that perceived positive quality may yield incremental benefits in brand evaluations. Other research suggests that positive interactions and communication build favourable image while unpleasant interactions lead to unfavourable image (Coombs and Holladay, 2006). In conclusion, the results imply that PR alone is not enough to generate customer loyalty thus brand image should be managed more carefully to produce desired results.

Some limitations of this research should be mentioned. First, an issue to be resolved is the need for replicating the PRP scale among consumers in different industries and countries in order to confirm the validity of the concept used in this research and refine its measurement. Second, we unintentionally discovered that the relationship between PRP and customer loyalty may be nonlinear. This nonlinear relationship, which in our results demonstrates a positive correlation but a decreasing trend, is a good subject for future research. Third, we adopted a cross-sectional approach in our study. However, due to the dynamic nature of the studied relationships, using a longitudinal method to analyze and clarify the relationships among the three variables would be an approach that merits further examination. Fourth, further research is encouraged to identify the key driving force of PR as it relates to relationship marketing as well as company performance.

In summary, the impact of PRP on customer loyalty is undisputed. Ideally, through PR, companies can demonstrate altruistic and non profit-oriented objectives, indicate sincere concern for consumer benefits, and establish consumer trust in order to enhance customer loyalty. Nonetheless, we have shown that the impact of PRP on customer loyalty is moderated by brand image and that that impact can have varying effects. Thus, when consumers perceive an organizations' brand image as unfavourable, companies should focus on implementing strategies aimed at enhancing their brand image to capture the complete trust from customers. Keller (1999) suggested that we can accomplish this through brand reinforcement or revitalisation strategies. Only after implementing those measures, should companies engage in PR activities to deliver timely and adequate messages to raise consumer's loyalty. By doing things in this order, companies can achieve “the multiplying effect of yielding double the results with half the efforts” according to a Chinese old proverb; as Cone et al. (2003) noted, “PR is a way of making a strong brand even stronger, but it can never turn a brand into something it's not.”

Figure 1Research model

Table IResults of EFA and CFA

Table IIRegression analysis of the relationship between PRP and customer loyalty

Table IIIModerating effects of brand image

Table IV t-Test results

References

Aaker, J.L. (1997), "Dimensions of brand personality", Journal of Marketing Research, Vol. 34 No.3, pp.347-57.