Corporate Reputation Review

Volume 10 Number 4

Academic Research Online and Ofine Corporate Brand Images: Do They Differ?
Sharifah Faridah Syed Alwi Department of Marketing and Information Systems, Faculty of Business & Accountancy, University of Malaya (UM), Kuala Lumpur, Malaysia Rui Vinhas Da Silva Manchester Business School (MBS), University of Manchester, Manchester, UK
ABSTRACT

This study compares the corporate brand images (CBI) of two book retailers: one that sells exclusively online and the other that sells exclusively ofine. The study has adopted measurements developed by Davies et al. (2004) known as the Corporate Character Scale in order to measure the two bookstores CBI. The study is informed by 511 responses from experienced customers of these two bookstores. To identify the difference between ofine and online CBI, conrmatory factor analysis was used. Previous literature stresses that Agreeableness and Enterprise were the two most signicant variables in predicting online and ofine CBI. This study, however, found that consumers appeared to view online CBI as more Informal and Innovative than ofine CBI (which appeared to be expressed more in terms of Agreeableness and Competence). In general, the present study adds to the existing literature in branding and reputation in two main ways. First, it puts the concept of emotional brand attributes (the CBI) in a unique context (the internet), and compares it with the bricks-andmortar context. The present study provides empirical evidence in the context of internet and retail bookstores to enhance the understanding of branding and reputation. Secondly, the practical contribution of the study and its manage-

rial implications can be seen in the context of dening strategy and positioning the corporate brand in the online and ofine contexts. Corporate Reputation Review (2007) 10, 217244. doi:10.1057/palgrave.crr.1550056
KEYWORDS: corporate

branding; corporate brand image; online corporate brand image; ofine corporate brand image; emotional branding

INTRODUCTION

Much has been written about the impact of the internet revolution on corporate brands (Ind and Riondino, 2001; Phillips, 2001; Supphellen and Nysveen, 2001; Merrilees and Fry, 2002; Stuart and Jones, 2004) but empirical studies exploring the effect of this medium on a companys brand image (corporate brand image (CBI)) are very few (Merrilees and Fry, 2002). Yet this understanding is vital to marketers, because a favorable brand image for a company may inuence consumer patronage and consumer decision-making, while an unfavorable image may adversely inuence such decisions and behaviors (Porter and Claycomb, 1997). Furthermore, CBI, provided it is valued by stakeholders, can help companies to achieve a sustainable competitive advantage and hence contribute to a higher propensity

Corporate Reputation Review, Vol. 10, No. 4, pp. 217244 2007 Palgrave Macmillan Ltd, 1363-3589 $30.00

www.palgrave-journals.com/crr

217

Online and Ofine Corporate Brand Images

among consumers to buy from a particular source, consequently leading to higher levels of protability. For example, Schultz and de Chernatony (2002) suggest that a successful corporate branding strategy provides an opportunity for generating a signicant future income stream. When a company serves its constituents well, its company name becomes a valuable asset. Its constituents are likely to be satised, which may encourage repeat purchases and ultimately translate into tangible economic returns to the organization (Fombrun, 1996).When a competitive advantage has been achieved, it may then be able to charge a premium price (Starr and Robinson, 1978; Campman, 2001; Davies et al., 2003). Furthermore, Campman (2001) argues that when a company differentiates itself using a unique value that is widely valued by its customers, the reward is not only being able to enjoy the prots of a premium pricing strategy but also being seen as a market leader (p. 204). However, despite the importance of the corporate brand, much of the corporate brand literature has so far been conceptual rather than empirical (Balmer and Gray, 2003). A lack of empirical research in this subject has led to the ongoing debate as to whether it is strategically realistic to translate corporate brand value from an ofine to an online context. It is as yet not clear how a corporate brand thrives in both bricks and clicks environments. This pressing managerial problem raises questions in terms of the academic understanding of consumer perceptions of an online corporate brand, in comparison to an ofine corporate brand. Recognizing this need, it is thus this papers intention to investigate CBI in two different contexts, namely the online (or internet) and ofine (or High Street) contexts. Specically, the paper addresses the following questions: 1. Could online and ofine bookstores have differing CBI(s)? 2. If so, what would be the CBI of organizations selling strictly online and ofine, respectively?

3. Do the dimensions vary across the different settings? 4. Which CBI(s) is/are more important in the online and ofine contexts, respectively? This paper is divided into several sections. The rst section explores the CBI concept and the background literature concerning both contexts, online and ofine, and is where the research proposition is developed. After presenting the literature review and research proposition, research methods and results are described. Findings are then discussed and their implications for management and future research are explored.
CBI: WHAT IT IS? CBI as a Term and Element of Corporate Reputation

According to Fombrun (1996) and de Chernatony (1999), corporate brand is an intangible asset of a company.The terms corporate brand and CBI are new but their subject matter is not. For example, corporate brand and CBI have been studied in many disciplines such as accounting, economics and marketing (Fombrun and Van Riel, 1996). In the past, different terms have been used to mean corporate brand and CBI (eg, company and corporate brand are used interchangeably (Keller, 2000) along with company brand image, corporate or store image and personality to mean CBI) (Singh, 1991; de Chernatony and DallOlmo Riley, 1998; Keller, 2000; Franzen and Bouwman, 2001). Conceptually, they are all thought to overlap and be redundant (de Chernatony and DallOlmo Riley, 1998: 425; Ind, 1997: 2; Franzen and Bouwman, 2001: 328).This might be either because these studies had similar objectives or because they used the same fundamental terms (Patterson, 1999). Another important reason why these terminologies may overlap may be because the denition of CBI itself, which is quite broad, refers to consumer association. For example,

218

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

in developing corporate image, Keller (2000) points out that a corporate brand may evoke associations with common products and their shared attributes or benets (eg, the consumer may refer to the quality or innovativeness of the product); people and relationships (customer orientation); programs and values (how concerned the company is about the environment and being socially responsible) and corporate credibility (a company seen as expert, trustworthy and likeable). Any contact that a customer or any other constituent has with the company, directly or indirectly, may affect the existence or strength of the associations (Keller, 2000: 118). In line with this view, Keller (2000: 118) further indicates that CBI will depend on a number of factors, such as: (1) the products a company makes, (2) the actions it takes and (3) the manner in which it communicates to consumers. The reputation of the company is then reected in these behaviors and this is how CBI is linked to corporate reputation. Thus, consumers associations could range from common product attributes, benets or attitudes, people and relationships, values and programs, to the credibility of the company. This may lead to different terms having been used by previous researchers but ultimately, as pointed out by Keller (2000: 118), it represents a companys brand image. Another useful corporate branding concept, from an internet perspective, is suggested by de Chernatony and Christodoulides (2004: 239) where brand is a cluster of rational and emotional values that enable stakeholders to recognise a promise about a unique and welcome experience. This concept applies regardless of online or ofine branding (de Chernatony and Christodoulides, 2004). Explicitly, when trying to understand whether there is a need for a different concept of corporate brands in the online context, de Chernatony (2002) and de Chernatony and Christodoulides (2004: 249) propose that the concept of the brand

itself does not change (ie, the rational and emotional values), but what may differ between online and ofine is its enactment (eg, navigation in an online environment and layout in a store/ofine environment). One could argue however that these rational and emotional values could include any association or be attributable to anything emanating from a company, with these associations indirectly linking to the company reputation as discussed above. Thus, a CBI can be thought of as the associations in the consumers mind with the company or corporation making the product or providing the service as a whole (Keller, 2000). Consistent with this notion, Davies et al. (2003) explain that when a companys reputation and the corporate brand are viewed by an internal stakeholder (eg, the employee), then reputation is taken to mean corporate identity (this view is illustrated by the lefthand side of the model see Figure 1). On the other hand, when reputation is viewed by an external stakeholder (eg, the customer), reputation is then taken to mean corporate image or, CBI (this is illustrated by the right-hand side of the model see Figure 1). This notion is supported by Bromley (2000). CBI is the external stakeholders perceptions based on their cumulative experiences of an organization, where corporate reputation represents the whole or collective representation of internal and also external stakeholders experiences (image and identity) with the organization. These combinations of views (identity and image) represent corporate reputation. Corporate brand, provided that its stakeholders value it, can help the company not only to achieve and sustain competitive advantage, but, through stakeholder satisfaction, hopefully will lead to increased product sales and better company performance. Keller (2000) outlines that some marketing experts believe that consumer perceptions of a companys role in society and how a company treats its stakeholders is a contributory factor of

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

219

Online and Ofine Corporate Brand Images

Satisfaction

Employee The View Brand

Customer View

Satisfaction

Loyalty Retention Identity Image

Sales (Davies et al., 2003, p.76)

Figure 1: Corporate reputation chain

increasing importance to consumer purchasing decisions. Davies et al. (2003) suggest that customer satisfaction drives performance, and it in turn is driven by image (CBI) as was proposed in the corporate reputation chain (see Figure 1). Thus, corporate branding is very important as it is intimately connected with an organizations reputation. Furthermore, Bergstrom et al. (2002) suggest that corporate brand and reputation are both fundamentally intangible and enjoy a close relationship. Corporate branding, provided it is valued by stakeholders, will help companies to achieve sustainable competitive advantage and hence contribute to a higher propensity among consumers to buy from a particular source, consequently leading to higher levels of protability (Grifn, 2002). Schultz and de Chernatony (2002) suggest that a successful corporate branding strategy provides an opportunity for generating a signicant future income stream.
CBI as an Emotional/Affective Construct

Supphellen and Nysveen (2001) argued that in the case of a company website, consumers may evaluate a corporate brand emotionally rather than cognitively. As long as the site works reasonably well, the consumer may develop a positive evaluation (affective) of the company brand. That is, instead of investigating the specic functional aspects of the

company website (eg, personalization of a website, security, reliability and navigability), consumers may directly attach their affective positive evaluation of the brand to the company website. This process is known in cognitive psychology as top-down processing. Biel (1991) explains that consumers are faced with varieties of brands, all of which make functional promises. Lambin (1997) adds that it is difcult to sustain these functional promises in pace with the advances in technology; a fact which, according to de Chernatony and DallOlmo Riley (1997), is due to most brands competing in the same category having become functionally more similar. Thus, in order to differentiate their brands, McEnally and de Chernatony (1999: 10) point out that marketers focus on incorporating emotional values into their brands, portraying these through the use of metaphor in creating brand personality. CBI is also seen as a more emotional and affective construct involving consumer feelings and reactions resulting from their experience of product/brand consumption. One way of investigating the effectiveness of an advertisement is to measure consumer responses through brand personality/CBI research (ie, using human personality traits and projective technique). According to Batra et al. (1996), brand personality testing, involving user imagery or endorsers, is frequently employed to portray the image of the company or brand in advertisements. This is an emotional type of advertising that has been used widely as a marketing communications tool in the past. Indeed, the effect of such a technique (using endorser or user imagery) in advertisements has been proven both empirically and conceptually (Batra et al., 1996). Explicitly, the role of emotions in particular has been demonstrated in the past to have an inuence on attitudes to the brand (Ab). That is, attitudes towards an ad (Aad) have inuenced consumer beliefs about and evaluations of the brand (Ab). Zajonc (1980) explains that

220

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

affective reactions are important in an advertisement because feelings can emerge very quickly.
CBI as Value

According to Aaker (1991), brand image helps create value for companies. For example, as a result of using a Volvo car, consumers associated Volvo not only with the brand image safety but also caring (Urde, 2003). Thus, whenever consumers hear about Volvo, they may then associate to these brand images (safety, caring), which in turn helps to create value for the company. This in turn helps to increase the goodwill towards the company and ultimately its reputation. CBI in other words is seen as the additional brand values that are inherent in or associated with the corporation and its product and services (Balmer and Gray, 2003: 973). When these stakeholders buy the company product, they are also believed to be buying a set of values that are attached not only to the product but also to the company (Ind, 1997). Thus, corporate brand could be dened as a sum of values that represents the organization (Ind, 1997). These values are thought to be seen as a guarantee of quality, as an insurance against risk of poor performance or nancial risk (Balmer and Gray, 2003: 973).
CBI: The Studys Denition

used to differentiate between alternative offerings (Patterson, 1999). Thus, the CBI construct is seen as being more about consumer emotional responses/reactions after experience with the company/store over time. Specically, the stakeholder involved in this study is the customer in both contexts and consistent with Bromley (2000) and Davies et al. (2003). Where external stakeholders (such as the consumer) are studied here, these terms are used: online CBI in an online context, and ofine CBI in the ofine context.
WHY STUDY EXCLUSIVELY ONLINE AND OFFLINE CBI?

There are several reasons why the present study is particularly interested in examining online versus ofine CBI. These are to determine what these brand images are and whether they vary across contexts. The following paragraphs provide a justication for concentrating on these issues.
The Internet as a Medium: Does it Currently Fit with Marketing Strategy?

Based on the previous discussion, CBI comprises several concepts (such as corporate image, brand personality) and could be seen as an emotional element and values and is associated with ones experience of a company brand. Explicitly, the study denes CBI as the sum of values that represent an organization (Ind, 1997) and these values or perceptions held by stakeholders are based on their cumulative experiences of an organization (Goldsmith et al., 2000). Thus, CBI is about the consumers emotional response to a brand that leads to the personication of brand attributes, and this is then

Businesses can use the internet for several purposes: as an information retrieval source, as a sales tool (e-commerce), as a distribution channel and for customer support (Sandberg, 1998; Peterson et al., 1997). Perhaps because of this, consumer internet penetration is predicted to increase (Mintel Report, 2004) and consumer buying on the internet is also on the rise (The Marketing Pocket Book, 2005). For example, internet penetration in the UK increased from 33 per cent in 2001 to 51 per cent in 2004 and online shopping is likely to increase from 17bn to 80bn by 2009 (Mintel Report, 2004). Many businesses and academics have argued that this medium will change the way that business is conducted: for example, the web could take over traditional media (eg, Herbig and Hale, 1997; Thomas, 1998). As a result of the predicted rise in consumer internet penetration and consumer buying on the internet as highlighted

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

221

Online and Ofine Corporate Brand Images

above, and perhaps in response to competitive pressure, many companies have established corporate websites (Butler and Peppard, 1998; Korgaonkar and Wolin, 1999; Stuart and Jones, 2004). Nonetheless, previous studies suggest that disappointing corporate web experiences are also on the rise (eg Wolff, 1998), and generating a profitable business online is still an ongoing key issue (Heijden and Verhagen, 2004). For example, there have already been cases where companies faced difculties, losses and failures with this new medium (Butler and Peppard, 1998: 600). For example, the early fallout of IBMs sponsored web shopping malls was due, in part, to unrealistic expectations of the demand potential in the eld, technology problems and a rather nave understanding of the true nature of competition in the new world (Butler and Peppard, 1998). The practitioners felt that they knew their existing customers but they were not sure whether this knowledge was applicable to an electronic environment (Cowles et al., 2002). For example, established companies that are looking to establish an online presence are often faced with a dilemma: should they establish a new and distinct online brand or simply use the internet as a delivery channel for their ofine store (Clauser, 2001; de Chernatony and Christodoulides, 2004)? Perhaps one possible explanation as to why consumers may not be buying online as much as predicted is the lack of company understanding of brand values online. Hence, it appears that the issue is not about whether the internet replaces the conventional channel but rather about how the internet ts into companies overall marketing strategies (Cowles et al., 2002), as it has been suggested that this medium be seen as complementary to existing channels (Reedy et al., 2000). Therefore, the argument that the web has changed everything for brands (Ind and Riondino, 2001: 8) is not in actual fact a reality (Ind and Riondino, 2001).

According to Uncles (2001), there are more questions and speculations than answers. For example, questions like What is the impact of brands on the Internet?; What aspects of branding change? and Is corporate branding on the web different from normal corporate branding practices? (Uncles, 2001: 250), need rigorous and systematic empirical evidence. Until we have such information, development will be dictated largely by fear, confusion and a gold-rush mentality (Hoffman, 2000). In fact, it is proposed that in an online context, branding is even more crucial (Clauser, 2001). Thus, the main requirement for a company when using the internet as a medium is to understand how to t the internet into the overall marketing strategy, and this demands a thorough appreciation of how branding operates in online and ofine contexts.
Does a Bricks-and-Mortar CBI Migrate Well to the Web?

From the above discussion, it can be accepted that the argument that the web has changed everything for brands may not be true. The underlying argument is not about whether the internet has replaced the existing media but rather about how to incorporate the medium into the greater marketing communications strategy and exploit it. But in order to guide marketers in this respect, the question of whether a bricksand-mortar brand works the same in the clicks environment must be addressed. However, the limited empirical studies that have focused on this matter have not provided concrete answers to enable a formal conclusion to be drawn regarding the questions of whether there are conceptual differences between online and ofine branding, and if so, how these differences are manifested. Hence, at this point, it can be seen that there are more questions than answers, particularly pertaining to the area of online branding as explained in the previous section.

222

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

In fact, it is proposed that in an online context, branding is even more crucial (Clauser, 2001), and such an understanding is vital because a favorable brand image of a company will predispose consumer patronage and positive consumer decision-making, while an unfavorable brand image will adversely affect such decisions and behaviors (Porter and Claycomb, 1997). Porter and Claycomb (1997) point out that the value created by brand image is projected onto the image of retail stores that carry that brand, and thus a consumers description of the retail stores (ie, in terms of assessment of the brands carried) is one way in which the retail stores image or the corporate brand is evaluated. Furthermore, it is proposed that in order to stimulate the positive impact of brand loyalty on attitudes and intentions, the corporate brand name should be as salient and vivid as possible on company websites; and that these should be considered as a vehicle for strengthening the brand image. Managers thus need to highlight specic image elements in order to build brand awareness and stimulate brand loyalty through positive attitudes and intentions (Supphellen and Nysveen, 2001). In addition, a strong corporate brand will increase an organizations reputation, and help to encourage and increase consumer satisfaction and perhaps loyalty, as exhibited in the Corporate Reputation Chain (see Figure 1) proposed by Davies et al. (2003). Thus, because a virtual environment is different from that of the physical, the understanding of consumers attitudes and how this new medium challenges the conventional theories and models is crucial (Merrilees and Fry, 2002). De Chernatony and Christodoulides (2004) explain that it is unclear whether the existing theory of brand/corporate brand in the ofine context is applicable in the online setting. It is suggested that the way a visitor draws an inference about a brand online is different from the way this is done in the ofine environment

(de Chernatony and Christodoulides, 2004). Additionally, Hart et al. (2000: 956) state image dimensions that were once relevant to xed (conventional) location stores may no longer be applicable to the virtual retail stores. Yet it is vital to maintain the core brand value of a company in the online setting, particularly for companies that also exist ofine (Ind and Riondino, 2001; de Chernatony, 2001). Empirical research on the consumer aspects of online shopping is scarce (Balabanis and Vassileiou, 1999; Jarvenpaa and Todd, 1997). Indeed, with the exception of CBI empirical studies (eg, Supphellen and Nysveen, 2001; Merrilees and Fry, 2002), most of the understanding of CBI in an online setting is still in its infancy and remains, in the main, at the conceptual level (eg, see conceptual studies by de Chernatony, 2001; Ind and Riondino, 2001; Stuart and Jones, 2004; Phillips, 2001; Schultz and de Chernatony, 2002; Clauser, 2001; de Chernatony and Christodoulides, 2004).The appreciation of whether ofine corporate brand values could be translated directly into an online context is very limited, and as Stuart and Jones (2004: 84) state What has often been ignored by organisations is the effect of the Internet on their positioning, particularly in cases where they already have a strong corporate brand rmly established in the marketplace. It is still uncertain whether it is strategically realistic to translate corporate brand values from an ofine to an online environment, despite this being the main concern of previous branding literature. Having a clearer knowledge of how to manage brands online, through an understanding of what aspects of the core brand values are more signicant for predicting the brand image of the company trading online, would perhaps provide companies with a better appreciation of the potential uses of the new medium as part of their overall marketing communications strategies (Stuart and Jones, 2004; de Chernatony and

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

223

Online and Ofine Corporate Brand Images

Christodoulides, 2004). Thus, by exploring what might be the CBI for both settings, and examining which aspects of brand attributes have the most impact on online and ofine CBI, this study hopes to ll the current gap.
ONLINE AND OFFLINE CBI PREVIOUS EMPIRICAL RESEARCH

Empirical studies that examine company name as a brand, and more specically CBI that represents an organization using the metaphor of brand as a person, are limited in comparison to reputation studies. Furthermore, no known study has attempted to make a comparison between a strictly online context and strictly ofine context in the retailing sector. Spector (1961) was one of the earliest to propose that a company can be viewed as people. Early attempts to examine CBI or acknowledge a corporation as a brand name can be seen in the works of Markham (1972) and King (1973). The more recent studies in this area include the works of Davies and Chun (2002), Davies et al. (2004) and Rojas-Mendez et al. (2004) for CBI in an ofine context. By using metaphorical expression, specically the personication approach that views the company as a person, these studies have examined CBI in various settings. King (1973) suggested that the metaphor of brand as a person has become a common approach in studying the image of product and service brands. According to Markham (1972), a researcher ought to use a generic scale in order to measure corporate reputation or corporate brand. A widely accepted scale proposed by Aaker (1997), known as brand personality, which identies ve dimensions (Excitement, Sincerity, Competence, Sophistication and Ruggedness), and a newly published scale developed by Davies et al. (2001), have both contributed to the metaphorical expression of corporate brands and the generic personality scale.

A more relevant recent study into retail corporate branding was undertaken by Davies and Chun (2002). They compared the CBI of two British department stores (ofine context) with similar ranges of products as perceived by employees and customers. They used the seven-dimension Corporate Character Scale developed by Davies et al. (2001). Their study found that the CBI of the two department stores were similar but the gap found was between the employees identication with the brand and the consumers image of the brand, and not within the brand image itself. For example, employees appeared to perceive the company as being more Ruthless than did the consumers. However, as there is still a lack of research in this area, it is not possible to generalize these ndings.
Online and Ofine CBI: Are they Different?

Most past empirical studies in CBI have been conducted in the ofine setting. However, the empirical studies that examine company name as a brand in general, and more specically CBI that represents an organization using the metaphor of brand as a person, are limited in comparison to reputation studies. Furthermore, there are very few studies that attempt to make a comparison between a strictly online context and strictly ofine context in the retailing sector. Aaker (1997), Davies et al. (2003) and Keller (2003) indicate that recently increasing attention has been focused on the understanding of the more abstract, intangible aspects of consumer brand knowledge (such as brand personality or CBI). By using metaphorical expression, specically the personication approach (ie, by viewing the company as a person), these studies have examined CBI or brand personalities in various settings. However, the past studies of corporate branding have predominantly been conceptual rather than empirical (Balmer and Gray, 2003), and most of these studies were conducted in the

224

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

ofine setting. How consumers view an online company as a person raises an interesting issue, leading to the question: Are there are any differences in brand personality between the online and ofine setting? According to Aaker (1997) and Davies et al. (2003), different settings may result in different dimensions. Thus, this study proposes that: Corporate brand image may vary depending on which context it operates in (online or ofine).
METHODOLOGY Online and Ofine CBI Measures

This study has dened above the online and ofine CBI as an affective and emotional construct. This denition appears consistent with several academics (eg Davies et al., 2003; Franzen and Bouwman, 2001) who regard the CBI or brand personality variable as more about the intangible or affective and emotional side of the brand rather than the functional or tangible side of it. In order to understand how to select the type of scale to use to measure the CBI, Markham (1972) suggests that there is a need to use a generic personality scale to compare a company with its competitors. According to Davies and Chun (2002), generic measures are not only able to assess any target but they are also applicable to the assessment of both corporate brand and reputation. Examples of CBI scales of the generic type are those of Aaker (1997) namely The Brand Personality and of Davies et al. (2004) namely the Corporate Character Scale. The Aaker scale Brand Personality, identies ve dimensions (Excitement, Sincerity, Competence, Sophistication and Ruggedness, and consists of 42 items) and the Davies et al. Corporate Character Scale identies seven dimensions (Agreeableness, Enterprise, Competence, Chic, Ruthlessness, Informality, and Machismo, and consists of 49 items). Both of these scales have contributed to a generic personality scale, using the metaphor of brand as a person or personality

traits. While the Aaker scale is an established scale and widely used across different cultural settings, the Davies scale is relatively newly published, thus its usage has been limited in comparison to the Aaker scale. However, no known research has yet incorporated either the Aaker scale or the Davies et al. scale in the online context. In deciding whether to use the Aaker or the Davies et al. scale to measure online CBI for the current study, it was eventually thought that the latter would be preferable, and the next paragraph provides a rationale for that choice. Although the Aaker scale has a longer track record, and is widely accepted as reputable in the branding literature, the Davies et al. scale, while being newer, includes more dimensions (eg, Informality) and the existence of some negatively constructed items and dimensions (eg, Ruthlessness) which are not present in the Aaker scale, but which may be worthy of investigation; particularly in the context of the present study, where knowledge of online CBI is limited. Thus, the Corporate Character Scale was chosen to measure the CBI in both settings in the current study.
Description of the Davies et al. Corporate Character Scale

The personication approach was used to dene the Corporate Character scale with seven dimensions assessed by 49 items. The personication approach or metaphor of personality traits is a common approach used to measure CBI or corporate reputation (King, 1973). Specically, brand as a person is simply a useful metaphor, using the medium of something that is easily understood and familiar the personality of a human being to aid in the understanding of the complex issue of what a brand is (Davies and Chun, 2002). Using a projective technique (ie, using personality traits, similar parameters used to measure human personality), respondents were then asked to assess the companys personality by imagining that

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

225

Online and Ofine Corporate Brand Images

the company was a human being (Davies et al., 2001). According to Davies et al. (2003), the dimensions and items in their scale reect the content of human personality descriptors and comprise words that people are comfortable with in describing a corporate brand. It was expected that CBI would be multidimensional with at least ve main dimensions (Agreeableness, Enterprise, Competence, Chic, and Ruthlessness) and two minor dimensions (Informality and Machismo). The latter two were retained in their study on the basis that these minor dimensions might be relevant in other contexts, although they were not strongly dened, not generally predicted in the literature, and did not explain much of the variance in the data set (Davies et al., 2004). These two minor dimensions, together with the inclusion of a negatively constructed dimension, are therefore worthy of investigation in the context of this study.
Pilot Study of the Corporate Character Scale

Results at this stage indicated a high degree of internal consistency among the seven dimensions of the Corporate Character Scale for example: Agreeableness (0.80), Enterprise (0.89), Competence (0.92), Ruthlessness (0.67), Chic (0.61) Informality (0.54), and Machismo (0.59). These results appear consistent with previous studies that have adopted the same scale in other ofine settings in UK retail stores. For example, in a recent study by Davies et al. (2004), the internal consistencies were as follows: Agreeableness (0.91), Enterprise (0.89), Competence (0.92), Ruthlessness (0.67), Chic (0.61) Informality (0.54) and Machismo (0.59). Thus, because the scale demonstrated its validity (internal consistencies ranging from 0.92 for Enterprise to 0.54 for Informality), the full Corporate Character Scale was considered to be suitable for use in the current research, without the need for any alteration whatsoever.
SAMPLING AND RESEARCH PROCEDURES

A major concern with the Corporate Character Scale was its reliability and validity, because the appropriateness and validity of using this scale, particularly in the internet setting, has not been tested before. Consequently, a pilot study was conducted within the Manchester Business School and other schools in the University of Manchester in order to test the appropriateness and validity of the scale in the internet setting. A drop-off/pick-up method was employed in the participating schools and 187 questionnaires consisting of the 49 items of the Corporate Character Scale were distributed. The nal number of usable questionnaires from the pilot study was 53, giving a 28.3 per cent response rate. In order to assess the validity and reliability of the items, Cronbachs Alpha internal consistency measure was used, and the 49 items were analyzed using SPSS version 12.

The empirical context being studied here was two retail bookstores which operate in a strictly virtual environment (called Store A) and a strictly ofine environment (called Store B). These bookstores offer their customers a wide range of books, including academic books, childrens books and ction. The product books in bookstores was chosen for this empirical study because books are currently, and projected to be, among the most purchased products in the online environment in the UK, as reported by Burt and Sparks (2003). It should be noted that the actual survey was conducted not in fact in Store A and Store B but in an ofine/High Street bookstore with three branches in the UK. Specically, the three branches are in Manchester, Leeds and London. There are two main reasons why the actual survey was conducted in a different ofine bookstore rather than Store A or B; these being to reduce bias from the data, and

226

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

because it was difcult to access the sample in the chosen bookstore. The study population was obtained through convenience sampling, which occurred when the questionnaire was distributed in the bookstore. In using this type of sampling, it was suggested by McDaniel and Gates (2003) that initially, this may seem to be a highly biased approach. However, when respondents are not asked to provide an evaluation of the company or compare it with rival operators, convenience sampling is thought to be an efcient and effective means of obtaining the required information (McDaniel and Gates, 2003). As in this study, the respondents in the three branches of the bookstore where the survey was conducted were not asked to evaluate these branches, but rather a different bookstore that operated online (Store A) and ofine (Store B). Therefore, in this sense, these respondents were not asked to evaluate the bookstore they were in at the moment (ie, the ofine bookstore) when the survey was conducted, nor were they asked to compare the bookstore with its competitors. Therefore, this could help to reduce the bias while obtaining the required information. In operational terms, the respondents were approached before they left the bookstores and were screened before being handed the questionnaire to ensure they had sufcient experience with the chosen online bookstore being studied. Specically, they were stopped and asked if they could spare ve minutes to be screened. If the respondents agreed, they were screened immediately to ensure that they met the criteria for inclusion. The main purpose of the screening was compatible with the CBI denition, that is, corporate branding or corporate reputation is viewed through an accumulation of consumer reactions to the experiences they had with an organisation (Davies et al., 2003: 178) as explained earlier. Moreover, this appears to be consistent with Wolnbarger and Gilly

(2003). They explain that since the entire purchase experience is of interest in their study, respondents were selected not only according to whether they interacted with the site but also depending on whether or not they had bought products via the site. This is what is meant in this study by the criterion that respondents should have sufcient experience. Specically, in order to ensure respondents familiarity or sufcient experience of Store A and Store B, they had to meet the following criteria:
(1) use the online stores website at least once a month (2) have bought a book at least one to three times in the last year through the online stores website (3) have visited the ofine store at least once a month (4) have bought a book at least one to three times from the ofine store in the last year.

If the respondents met the above criteria (14), they were given a questionnaire to be completed on the spot, or in their own time to be returned in a self-addressed envelope.
The Questionnaire

The rst part of the questionnaire consists of two columns representing the online bookstore and the ofine bookstore and all 49 personality trait items (as shown in Table 1) proposed by Davies et al. (2004). Respondents were asked to circle their answers in a Likert scale format (15) with strongly disagree, disagree, neutral, agree and strongly agree in both contexts. Also included in the questionnaire were online, ofine, background and demographic information.
Characteristics of the Sample

The respondents in the current study were predominantly male (58.9 per cent) and the average age was between 35 and 44 years old (29.9 per cent). Most of the respondents

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

227

Online and Ofine Corporate Brand Images

Table 1: Items/Dimensions of Corporate Character Scale

Dimensions Agreeableness

49 items Friendly, Pleasant, Open, Straightforward Concerned, Reassuring, Supportive, Agreeable Honest, Sincere, Socially responsible, Trustworthy Reliable, Secure, Hardworking Ambitious, AchievementOriented, Leading Technical, Corporate Cool, Trendy,Young Imaginative, Up-to-date, Exciting, Innovative Extrovert, Daring Charming, Stylish, Elegant Prestigious, Exclusive, Rened Snobby, Elitist Arrogant, Aggressive, Selsh Inward-looking, Authoritarian, Controlling Masculine, Tough, Rugged Casual, Simple, Easy-going

Competence

Enterprise

Chic

or higher degree (18.6 per cent) in the study of (74). In addition, the characteristics of the respondents in the present study are representative of the UK internet population. For example, they compare with the recent internet Users Prole survey specically for bookselling in the UK, which indicated that 53 per cent were male with an average age of 3544 years old and came from the AB social grades (34 per cent), (66). Moreover, according to Keynote (2004), the average age prole for book purchasers in the UK is 3544 years old and, although book buying occurs across all social groups, the strongest representation was in the AB social grades. AB social grades are at the top of the UK social class system with the highest earnings and average age of 3544 years (based on the Keynote (2004) denition of the UK social class system).
DATA ANALYSIS AND RESULTS

Ruthlessness

Machismo Informality

Source: Davies et al (2004)

held university degrees and were employed in professional jobs (42.1 per cent). When comparing with other previous studies, particularly those dealing with online contexts, the characteristics of our sample in this study were more or less consistent with those in previous studies. For example, the average age was 32 years with 78.7 per cent being male in the study of (73); 37 years in the study of (74) and 44 years with 73 per cent being male in the study of (75). Furthermore, a large number of the respondents held a university degree (40.2 per cent)

In order to test the study proposition, conrmatory factor analysis (CFA) (rst and second order) was performed. The scale adopted in this study is proposed as being multidimensional (Davies et al., 2004: 131). However, when it is recommended by priori theoretical structure that CBI are expected to be multidimensional, or consist of at least ve dimensions (Davies et al., 2004), the second order needs to be performed for a thorough investigation of the constructs validity. As yet, however, it is unclear whether or not the scale has been empirically tested in the second-order form. The second-order form is important, particularly for a newly published scale, as it deals with the validation of the scale (Bagozzi and Heatherton, 1994; Hair et al., 1998; Graver and Mentzer, 1999). In addition, this technique is consistent with previous studies of CBI. For example, a recent study was carried out by RojasMendez et al. (2004) measuring the Ford brand personality using Aakers (1997) scale in Chile. In their study, the Aaker brand

228

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

personality scale was reduced to four dimensions in the rst-order phase: Sincerity, Competence, Excitement and Sophistication, with 16 items remaining from the original 32, in order to achieve reliability and validity. Since Aaker (1997) explained that there are ve brand personality dimensions that could potentially represent an organizations brand personality, Rojas-Mendez et al. (2004) empirically tested this assumption by performing the higher or secondorder CFA. Although not all ve dimensions were found to represent the brand personality of Ford, they did conrm that the four dimensions were indeed explained by the latent construct, namely brand personality, through the formation of a higher or second order as proposed by priori theoretical structure (Aaker, 1997). Therefore, using the technique consistent with Rojas-Mendez et al. (2004) to measure the online and ofine CBI, both rst- and second-order CFA were performed. Specically, rst-order CFA examines the CBI dimensions individually and second-order CFA examines brand image as a whole (Rojas-Mendez et al., 2004).
First-Order Model

model is tentative and regarded as only an approximation of reality (Jreskog and Srbom, 1982: 408) rather than an exact statement of truth and poor t is not surprising as often models need to be re-specied in CFA (Anderson and Gerbing, 1988: 412). The result clearly shows that the hypothesized models needed to be rened to achieve an acceptable level of t. Therefore, the priori model had to be rejected and the studys intention here is thus to nd two appropriate measurements or rst-order models (in both contexts) that are both substantively meaningful and statistically acceptable when using the conrmatory technique.
Assessment of Model Misspecication

The parameters of the Corporate Character Scale, which consists of 49 items as proposed by Davies et al. (2004) (see Table 1), were then sent for CFA, specically to perform the rst order analysis separately for the two contexts (see Appendices A and B) for both online and ofine CBI. The initial results for both contexts showed a poor t and needed to be re-specied. For example, the result of ( 2(1106) = 3,841, p < 0.001); GFI = 0.731; TLI = 0.657; CFI = 0.678; 2/d.f.= 3.5 for the online CBI store; and ( 2(1106) = 3,831, p < 0.001); GFI = 0.735; TLI = 0.684, CFI = 0.703, 2/d.f.= 3.5 for the ofine CBI store; suggests that there were mists in the two models. Only RMSEA indicated a marginal t for both contexts at 0.07. It should be understood that in most empirical work, a

Items that are cross-loaded in more than one dimension were relaxed one at a time as proposed by Long (1983) and insignicant parameters were excluded from the study. For example, two items (Inward Looking and Corporate) for the online store and one item (Corporate) for the ofine store were excluded in the early stage of this analysis due to being insignicant. Besides relaxing parameters, moving or adding items (or parameters) from one dimension to another dimension that were highly cross-loaded, was also performed in this study. However, any moving, excluding or adding of parameters needs to be performed based on theoretical, statistical and practical considerations (Bagozzi and Heatherton, 1994; Byrne, 2001; Anderson and Gerbing, 1988). For example, in the online context, two items called Straightforward and Open in the Agreeableness dimension were highly cross-loaded with the items Simple and Easy Going in the Informality dimension; and Technical in the Competence dimension was highly crossloaded with all of the items representing the Enterprise dimension. These cross-loadings indicate that two items Open and Straightforward from the Agreeableness dimension and one item, Technical, from

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

229

Online and Ofine Corporate Brand Images

the Competence dimension appeared to make up the dimensions Informality and Enterprise, respectively, rather than Agreeableness and Competence, to which they originally belonged. Theoretically, as highlighted by Davies et al. (2003), the item Simple in Informality means it is straightforward and clear, easily understood and lucid (p. 253). Hence, the inclusion of the item Straightforward in the Informality dimension is deemed to be appropriate. The second item, Open was originally associated with warmth and is now more associated with easy-going in the online environment.The company may be seen as more open, relaxed, unconstrained in their thinking and in the way they deal with customers as explained by Davies et al. (2003: 243), as life here is not about corporatism. Practically speaking, the online store may be seen as more open in its thinking when it is associated with technology, and thus may be seen as representing a more informal style rather than a corporate and formal style. The third item, Technical, was moved to the Enterprise dimension from Competence. Technical here should not be seen as technocratic or inexible in orientation by the company as highlighted by Davies et al. (2003), but in contrast, means a rather more exible approach. This point is supported by the company being seen as more informal (Open and Easy Going) as explained earlier. In addition,Technical may be associated with books and with the technology itself, where customers might associate this technology with modernity (eg, Cool, Exciting and Trendy) and also creative and adventurous (Imaginative and Innovative). It is worth noting at this stage that it appeared to be more relevant to view the online CBIs Enterprise dimension as more about the Innovation dimension than the Enterprise dimension; since here consumers may perceive the company that trades on the internet to be more innovative, imaginative, cool, trendy and exciting; and this is

related to the technicality of technology the internet itself. For this reason, it was thought appropriate that Innovation should replace the Enterprise dimension in the online CBI. On the other hand, in the ofine store, the item Leading in Competence has been moved to Chic, not only for statistical reasons (such as high modication indices, standardized residuals and higher standardized loading, which reect the items convergent validity when removed), but because it makes practical sense when it is associated with the items Prestige and Elegance in the Chic dimension. Both models were then revised until they attained acceptable values in the goodness-of-t indices. As a result, the hypothesized model lost more than 60 per cent of its items for both contexts (29 items for online and 28 items for ofine). The nal results show 20 items were retained for the online store and 21 items for the ofine store (as shown in Figures 2 and 3). Specically, ve dimensions (Agreeableness, Enterprise, Competence, Chic and Informality) represent the online store and four dimensions (Agreeableness, Enterprise, Competence and Chic) represent the ofine store. Goodness-of-t statistics show that the online model ( 2(160) = 337.84, p < 0.001); GFI = 0.936; TLI = 0.925; CFI = 0.937; RMSEA = 0.047; 2/d.f.= 2.1) and ( 2(183) = 497.73, p < 0.001); GFI = 0.915; TLI = 0.900; CFI = 0.913; RMSEA = 0.058; 2/d.f.= 2.7) ts the data very well. Convergent validity for the online store and ofine CBI were all supported in this study with all parameter estimates, that is, the items standardized loadings >0.5 (Kline, 1998) (with one exception the item Simple in the online context was retained for theoretical reasons) and all items were statistically signicant at p < 0.001 (Anderson and Gerbing, 1988) (see Figures 2 and 3). It is also worth noting from Figures 2 and 3 that the Agreeableness dimension differs in the two contexts. For example, for the online

230

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

Pleasant Friendly Agreeable Supportive Trendy Cool Imaginative Daring Exciting Innovative Technical Straightforward Open Easy Going Simple Refined Prestige Elegant Secure Reliable
0.67* 0.53* 0.65*

0.74* 0.57* 0.75* 0.65*

Agreeableness

0.52
0.54*
0.62*

Innovation

0.81

0.67* 0.49*

0.52 0.58 0.67

0.59* 0.59* 0.64* 0.36*

0.31 Informality 0.29 0.26

0.64* 0.72* 0.79*

Chic

0.61

0.41
0.71* 0.64*

Competence

Note: * indicates all loadings were significant at p<0.001 First Order Online CBI: Fit Indexes
Fit Indices 337.84 p>0.001 /df 2.1 GFI 0.936 TLI 0.925 CFI 0.937 RMSEA 0.047

Figure 2: Online corporate brand image first order

CBI Agreeableness, which consisted of four items, described more about being pleasant and friendly than supportive and agreeable. However, in the ofine CBI, nine items were found to represent the dimension and these were reected mostly by warmth (friendly, pleasant, open and straightforward), empathy (reassuring, agreeable and supportive)

and integrity (sincere and socially responsible). This perhaps is reective of the different nature of the two contexts. In the online context, the consumer may perceive the website under investigation as more user-friendly, being a pleasant website to use and providing sufcient technical and moral support during use of the site. In the

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

231

Online and Ofine Corporate Brand Images

e1 e2 e3 e4 e5 e6 e7 e8 e9 e10 e11 e12 e13 e14 e14 e15 e16 e17 e18 e19 e20

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Pleasant Friendly Socially Responsible Agreeable Reassuring Straightforward Open Sincere Supportive Trendy Cool Imaginative Daring Extrovert Refined Prestige Elegant Leading Secure Reliable Hardwork
0.63* 0.70* 0.58* 0.65* 0.63* 0.69* 0.80* 0.62* 0.67* 0.76* 0.53* 0.52*
0.62* 0.51*

0.74*

0.76* 0.72* 0.53* 0.57* 0.65* 0.73*

Agreeableness

0.57

0.82 Enterprise 0.53 0.41

Chic

0.44

0.54

Competence

Note: * indicates all loadings were significant at p<0.001

First Order Offline CBI: Fit Indexes Fit Indices /df

497.73 p<0.001 2.7

GFI 0.915

TLI 0.900

CFI 0.913

RMSEA 0.058

Figure 3: Offline corporate brand image first order

bricks-and-mortar context, the bookstore may be seen as not only having a friendly and pleasant environment with supportive and reassuring treatment in the personal interactions between staff and customer but also may be viewed as having more integrity (such as sincerity and social responsibility) due to the bookstore itself sponsoring an event or simply being seen as a shop selling books and therefore fullling a moral obligation to society. Two issues remain unresolved at this point however. The rst issue relates to high correlation. Since there is covariance (a correlation) between Agreeableness and Informality = 0.81 in the online store, and

Agreeableness and Competence = 0.82 in the ofine store (as shown in Figures 2 and 3, respectively), the question of discriminant validity has yet to be dealt with. In other words, discriminant validity will be achieved if the degree of association is weak and this means that the CBI dimensions are distinct in concept (Bagozzi and Heatherton, 1994). However, this had yet to be shown in our case. Thus, a chi square differences test (SCDT- 2 test) was conducted to compare a two-factor model with a one-factor model solution to see whether the factors would hold up as separate factors despite being highly correlated. The test result indicated that the differences were signicant in the

232

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

two-factor model ( 2(1) = 34.7, p < 0.001) for the online store and ( 2(1) = 40, p < 0.001) for the ofine store, thus conrming that these factors were indeed separate factors. Besides that, other correlation results for both stores were low (ranging from 0.26 to 0.67 in the online store and 0.41 to 0.57 in the ofine store) as depicted in Figures 2 and 3, respectively. In this sense, discriminant validity was supported for both contexts.
Online and Ofine: Second/Higher-Order Model

As explained earlier, the second order was performed not only because the scale (Corporate Character Scale) adopted in this study had been specied in priori theoretical structure but also because it will aid an understanding of the dimensions representing online and ofine CBI. Results for both stores in the rst-order CFA were then sent for second-order CFA for further analysis.
Online and Ofine CBI: Alternative Models

However, before a decision is made to select the Second-/Higher-Order model for both contexts, there is a need to compare the second-order model (Model 1 as shown inFigure 4 and 5) with two other alternative models. This comparison is important as it provides an additional test for common method bias (Davy et al., 1997). The rst alternative model is the rst-order model (as shown in Figure 4 or Model 2 in Figure 5). The second alternative model is a singlefactor model (presented in Figures 4 and 5 as Model 3). The results of the second-order model (Model 1) were then compared with the two alternative models in both contexts. As shown in Tables 2 and 3, in Figures 4 and 5, respectively, the results of the singlefactor model (Model 3) did not acceptably t the data ( 2 = 1,130, p < 0.001; 2/d. f.= 6.6, GFI = 0.768,TLI = 0.619; CFI = 0.659; RMSEA = 0.11) for online (Table 2) and 2 = 1,162, p < 0.001; 2/d.f.= 6.2, GFI = 0.79,

TLI = 0.69; CFI = 0.731; RMSEA = 0.10) for ofine (Table 3). Furthermore, according to the Chi-Square Test Statistic ( 2 test) comparing Model 3 and Model 1, 2 (5, N = 511) = 757.56, p < 0.001 for the online and 2(23, N = 511) = 702.44, p < 0.001 for ofine contexts, the differences between the two models (Model 1 and Model 3) were signicant. The signicant 2 test would indicate that Model 1 is being supported in its multidimensionality and not Model 3 (the single-factor Model). Therefore, Model 3 was rejected for both contexts. When comparing the rst-order model (Model 2) and the second-order model (Model 1) for both contexts, both performed similarly where the second-order model produced near-identical results to the rstorder model (see Tables 2 and 3 and compare their t indices). However, where both models show acceptable t indices as in this case, there are two ways in which a decision concerning which model to choose for further analysis can be made. First, it can be based on the priori status of the CBI structure. As is recommended by priori theoretical structure, whereby CBI consists of multidimensional constructs with at least ve personality dimensions (Davies et al., 2003), this suggests that the second-order model may be preferable. Secondly, as explained earlier, the use of a second-order model increases the validity of the construct (Graver and Mentzer, 1999; Hair et al., 1998: 626). If the priori structure demonstrates multidimensionality, then all dimensions should measure the same thing and should covary at a higher level if they are good measures of the underlying variable (Bagozzi, 1994: 331). In other words, if the model could be empirically tested in a second-order form, this would allow a stronger statement: While there may be some overlap between the dimensions of corporate brand image, the dimensions are to some extent distinct from each other (Hair et al., 1998; Schmidt, 2005). As indicated in Figures 6 and 7, for

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

233

Online and Ofine Corporate Brand Images

Model 1: Second-Order Model Pleasant, Friendly, Agreeable, Supportive Daring, Trendy, Cool, Exciting, Imaginative, Innovative, Technical Reliable, Secure Refined, Prestige, Elegant Agreeableness Innovation Competence

ONLINE CBI

Chic

Open, Easy Going, Simple, Straightforward

Informality

Model 2: First-Order Model Pleasant, Friendly, Agreeable, Supportive Daring, Trendy, Cool, Exciting, Technical, Imaginative, Innovative Reliable, Secure Agreeableness

Innovation Competence

Refined, Prestige, Elegant

Chic

Open, Easy Going, Simple, Straightforward Model 3: Single-Factor Model Pleasant, Friendly, Agreeable, Supportive

Informality

Daring, Trendy, Cool, Exciting, Technical, Imaginative, Innovative

Open, Easy Going, Simple, Straightforward

Refined, Prestige, Elegant

Reliable, Secure

ONLINE CBI

Figure 4: The alternative models: Online context

online and ofine contexts, respectively, the structural relationships (or factor loadings) covaried from one dimension to another when they were tested in a higher/secondorder form. Therefore, on the basis of the priori status of the scale and construct validity (that the measures indeed covary as shown in Figures 6 and 7, respectively), a decision to select the second-order (or Model 1) was made for both contexts (online and ofine).

Online CBI: Reliability Analysis

CA (the Cronbach Alpha) and CR (the Construct Reliability) are all above the recommended level for the online and ofine CBI constructs (as shown in Table 4 for both contexts).
The Importance of CBI Dimensions

This study has earlier stated the research proposition: Corporate brand image may vary

234

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

Model 1: Second-Order Model Pleasant, Friendly, Agreeable, Supportive, Sincere, Straightforward, Open, Socially Responsible, Reassuring

Agreeableness

Daring, Trendy, Cool, Imaginative, Extrovert

Enterprise

OFFLINE CBI

Reliable, Secure, Hardworking

Competence

Refined, Prestige, Elegant, Leading

Chic

Model 2: First-Order Model Pleasant, Friendly, Agreeable, Supportive, Sincere, Straightforward, Open, Socially Responsible, Reassuring

Agreeableness

Daring, Trendy, Cool, Imaginative, Extrovert

Enterprise

Reliable, Secure, Hardworking

Competence

Refined, Prestige, Elegant, Leading

Chic

Model 3: Single-Factor Model Pleasant, Friendly, Open, Agreeable, Supportive, Reassuring, Sincere, Straightforward, Socially Responsible Daring, Trendy, Cool, Imaginative, Extrovert

Reliable, Secure, Hardworking

Refined, Prestige, Elegant, Leading

OFFLINE CBI

Figure 5: The alternative models: Offline context

Table 2: Comparative Fit Statistics and Indices for Competing Models of Online CBI

Fit indices Second-order model First-order model Single-factor model

d.f. 165 160 170

/d.f.

GFI 0.931 0.936 0.768

TLI 0.915 0.925 0.619

CFI 0.926 0.937 0.659

RMSEA 0.047 0.05 0.11

372.44 337.84 1130

2.3 2.1 6.6

depending on which context it operates in (online or ofine). The above results have shown that the online store was represented

by ve dimensions and the ofine store by four (as shown in Figures 6 and 7). While Informality was found to be important in

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

235

Online and Ofine Corporate Brand Images

Table 3: Comparative Fit Statistics and Indices for Competing Models of Ofine CBI

Fit indices Second-order model First-order model Single-factor model

2 459.56 497.73 1162

d.f. 166 183 189

/d.f.

GFI 0.917 0.915 0.789

TLI 0.902 0.900 0.694

CFI 0.915 0.913 0.731

RMSEA 0.059 0.058 0.10

2.7 2.7 6.2

Pleasant Friendly 0.57* 0.75* Agreeable 0.65* Supportive

0.74*

Agreeableness

Trendy Cool Imaginative Daring Exciting Innovative Technical Straightforward

0.54* 0.62* 0.67* 0.53* 0.65* 0.67* 0.49* 0.56* Innovation 0.98*

Online CBI Second Order 0.59* 0.59* 0.64* Informality 0.50* 0.83*

Open

Easy Going

0.36*

Simple 0.70* Refined 0.64* 0.72* Chic Prestige 0.79*

Elegant

Secure

0.71* 0.64* Competence

Reliable

Figure 6: Online CBI: Higher/second-order construct

explaining the CBI in the online context, it was not so in the ofine context. This result simultaneously supported the postulation that

CBI could be explained by multidimensional constructs and that the dimensions vary depending on in which context they are

236

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

Pleasant Friendly
Socially Responsible

Agreeable Reassuring Straightforward Open Sincere Supportive Trendy Cool Imaginative Daring Extrovert Refined Prestige Elegant Leading Secure Reliable Hardwork

0.74 0.62 * * 0.51* 0.76* 0.72* Agreeableness 0.53* 0.57*

0.65* 0.73*

0.94*

0.62* 0.67* 0.76* 0.53* 0.52*

Enterprise

0.58*

Offline CBI Second Order

0.36*
0.69* 0.80* 0.65* 0.63*

Chic

0.86*

0.63* 0.70* 0.58*

Competence

Figure 7: Offline CBI: Higher/second-order construct

operating. Secondly, as shown in Table 5, online CBI explains 95, 31, 49, 25 and 69 per cent of the variance associated with the dimensions Agreeableness, Enterprise, Competence, Chic and Informality, respectively, and this represented the online store. In contrast, ofine CBI comprises 88, 33, 74 and 36 per cent of the variance associated with the dimensions Agreeableness, Enterprise, Competence and Chic, respectively, and this represented the ofine store.
DISCUSSION AND CONCLUSION

(2) If so, what would be the CBI of organizations selling strictly online and ofine, respectively? (3) Do the dimensions vary across different settings? (4) Which dimension or factor is more important in the online and ofine contexts, respectively?

The following questions were investigated:

(1) Are the CBI represented by multidimensional constructs in both contexts of the bookstores?

In response to the above questions, the following has been proposed: Corporate brand image may vary depending on which context it operates in (online or ofine). The dimensions that represented both contexts were: 1. Online CBI was represented by ve dimensions (Agreeableness, Innovation, Competence, Chic and Informality), and

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

237

Online and Ofine Corporate Brand Images

2. Ofine CBI was represented by four dimensions (Agreeableness,Enterprise,Competence and Chic). Although not all of the seven dimensions appeared in both contexts, as different study environments may result in different dimensions (Aaker et al., 2000; Davies et al., 2003), the differences were apparent between the two contexts. While Innovation is a new dimension representing the online CBI, Agreeableness (with 95 per cent in variance) and Informality (69 per cent) were found to be the rst and second most important dimensions in explaining the CBI in the online context; while Agreeableness (88 per cent) and Competence (74 per cent) were the two most important dimensions found in the ofine context. These results simultaneously support the proposition that CBI dimensions vary depending upon the context in which they are operating. In relation to ofine contexts, it appears that this study has conrmed the earlier results of Davies et al. (2004) where the two most important brand image dimensions were Agreeableness and Competence. However, Informality, which was regarded as a minor and weak factor in past research in the ofine context (eg Duarte and Davies, 2002; Davies and Chun, 2002; Davies et al., 2003, 2004) was, in contrast, found to be the second most important factor for the CBI in the online context in this study. Some studies have shown that the electronic environment may be more exible and informal (Clauser, 2001), but this is more of a conceptual proposition than an empirical one. Studies involving the comparison of High Street and online stores are vast in the areas of marketing and retailing, but there are no known examinations of the comparison of CBI in online and ofine settings using the personality trait approach.Thus, comparisons of this studys results with past research in which personality traits are represented, or have been shown to be more signicant in the

online context, have been limited. However, this nding perhaps gives some insight to both academics and practitioners into which dimensions of brand image are considered to be more important in an online context; and simultaneously helps to address whether the corporate brand is similarly perceived in the online and ofine contexts, respectively. Since brand personality is heavily built on advertising (McEnally and de Chernatony, 1999), the company may want to address or emphasize its symbolic value (being informal and agreeable) more in the online context. The way a visitor draws an inference about a brand online is different from in the ofine environment, where conventional factors such as staff and the store environment have a notable impact (de Chernatony and Christodoulides, 2004), and this may explain why the Informality dimension is more signicant in the online context than in the ofine. Specically, customers may associate the company with qualities of being Simple and Straightforward in the online environment probably because the site recommends a book that matches the customers wants or needs or simply because it has an easy-tonavigate website (eg, for ease in nding book information or completing a purchase transaction online, etc), whereas in the ofine context, if a book could not be found in the store, customers might ask the employees themselves. Open, on the other hand, has been highly correlated to easy going in the sense that the online environment may be seen as more open and easy going in dealing with customers. Davies et al. (2003) suggest that easy going infers a company that is able to be more informal and relaxed in its thinking and in its dealings with customers. This image is indeed perceived as favorable or positive by several emergent companies in the high-tech sector (Davies et al., 2003). Furthermore, although Agreeableness dimensions were found to be related in explaining the CBI in both contexts, the items representing the Agreeableness dimension

238

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

differ between the two contexts. For example, as discussed earlier, the dimension Agreeableness, which consisted of four items in the online CBI, was described as being more about pleasant and friendly than supportive and agreeable, while in the ofine CBI, nine items were found to represent the dimension and they were reected mostly by warmth (friendly, pleasant, open and straightforward), empathy (reassuring, agreeable and supportive) and integrity (sincere and socially responsible). This perhaps is reected in the different nature of the contexts where, in the online context, the consumer may perceive the website under investigation as more user-friendly and being a pleasant website to use, and providing sufcient technical and moral support in using the site. The bricks-and-mortar bookstore may be seen as not only having a friendly and pleasant environment with supportive and reassuring treatment in the personal interaction between staff and customer, but also as having more integrity (such as sincere and socially responsible) because the bookstore itself may be sponsoring an event or simply be seen as a shop selling books and thus fullling a moral obligation to society. Besides this, it is worth noting that, as discussed in the previous section, it appeared to be more relevant to view the online CBIs Enterprise dimension as more about the Innovation dimension rather than the Enterprise dimension. This is due to the consumer possibly associating the company that trades behind the internet or the internet itself as more innovative, imaginative, cool, trendy and exciting, and is related to the technicality of technology and the internet itself. For this reason, it was thought appropriate that Innovation should replace the Enterprise dimension in the online CBI. Additionally, other dimensions of online CBI explain 31, 49 and 25 per cent of the variances associated with the dimensions Innovation, Competence, and Chic, respec-

tively. In contrast, ofine CBI comprises 33 and 36 per cent of the variances associated with the dimensions Enterprise and Chic, respectively. The CBI of Chic in both contexts was low in the associated variance, with 36 per cent in the ofine store and 25 per cent in the online store. Its low importance may be due to the fact that customers may be associating Chic with luxury, exclusive products or companies (Davies et al., 2003) rather than with a bookstore. The inclusion of a negative dimension in this scale was not relevant in this studys context. Ruthlessness was found to be inappropriate in explaining CBI for both stores, as was Machismo. Ruthlessness was probably less important here because stores operate within a retailer-controlled campaign, as suggested by Burt and Sparks (2002) and Davies et al. (2004). Machismo, on the other hand, may not be associated with either the contexts or the bookstore itself. This was consistent with the ndings of Davies et al. (2001) and RojasMendez et al. (2004) on the Ruggedness dimension from the Aaker scale. However, the inclusion of more dimensions in the Davies et al. (2004) scale (specically the Informality dimension) appears to have been useful in this studys context. These results thus supported the above research proposition that corporate brand image dimensions vary depending on in which context it is operating. Explicitly, while the Agreeableness, Innovation and Informality dimensions have been found to help explain the CBI in the online context, Agreeableness, Enterprise and Competence have been found to be the three main dimensions of ofine CBI. However, this study is not without its limitations. The present study was conducted in two bookstores in two different contexts. Thus, the ndings of this study cannot be generalized to other industries. Crossvalidation to other industries could not be performed in this study as this was outside of its scope. Replicating this research in similar or other industries is therefore vital in order to be able to generalize the ndings of this

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

239

Online and Ofine Corporate Brand Images

study. Another future study might compare the results when a corporate brand migrates from a bricks to clicks environment. This would identify whether there is any consistency when transferring the brand values from an ofine to an online setting, or to identify whether there is any need for them to be consistent after all, due to the different nature of their settings, as has been found in the present study. Finally, identifying the differences would be interesting, but the ndings do not provide much insight as to why online is more informal than ofine or ofine is more competent than online. Gaining an understanding of consumer cognitive thinking, for instance by integrating store attributes (eg, products or services of the company) might be useful in aiding a deeper understanding. By integrating and investigating their relationships (the products/services and brand image/symbolic value) in one model, one might be able to understand what the drivers of perceived brand image might be. This would aid an understanding of why the CBI is perceived in that sense.Therefore, investigations into understanding why these gaps occur, by adopting a qualitative or/and quantitative approach, are a suggestion for future research.
REFERENCES
Aaker, D.A. (1991) Managing Brand Equity, The Free Press, New York. Aaker, J.L. (1997) Dimensions of brand personality, Journal of Marketing Research, 24, 347356. Aaker, J.L., Benet-Martinez, V. and Garolera, J. (2000) Consumption symbols as carriers of culture: A study of Japanese and Spanish brand personality constructs, Journal of Personality and Social Psychology, 81, 492508. Ailwadi, K.L. and Keller, K.L. (2004) Understanding retail branding: Conceptual insights and research priorities, Journal of Retailing, 80, 331342. Anderson, G.J. and Gerbing, W.D. (1988) Structural equation modelling in practice: A review and recommended two-step approach, Psychological Bulletin, 107, 238246. Arbuckle, L.J. (1999) AMOS 4.0 Computer Software, Smallwater, Chicago.

Bagozzi, R.P. (1994) Structural equation model in marketing research, in R. Bagozzi (ed.), Basic Principles in Principles of Marketing, Blackwell, Oxford, pp. 317385. Bagozzi, R.P. and Heatherton, T.F. (1994) A general approach to representing multifaceted personality constructs: Application to state self-esteem, Structural Equation Modelling, 1, 3567. Balabanis, G. and Vassileiou, S. (1999) Some attitudinal predictors of home-shopping through the Internet, Journal of Marketing Management, 15(5), 361385. Balmer, J.M.T. (1998) Corporate identity and the advent of corporate marketing, Journal of Marketing Management, 14, 963996. Balmer, J.M.T. and Gray, E.R. (2003) Corporate brands: What are they? What of them, European Journal of Marketing, 37(7/8), 972997. Batra, R., Myers, G.J. and Aaker, A.D. (1996) Advertising Management, 5th edn, Prentice-Hall, UK. Beren, G. and Van Riel, C.B.M. (2004) Corporate associations in the academic literature: Three main streams of thought in the reputation measurement literature, Corporate Reputation Review, 7, 161178. Bergstorm, A., Blumenthal, D. and Crothers, S. (2002) Why internal branding matters: The case of saab, Corporate Reputation Review, 5(2/3), 133142. Berry, L.L. (1969) The component of department store image, Journal of Retailing, 45, 311. Biel, A. (1991) The brandscape: Converting brand image into equity, ADMAP, 26(10), 4146. Bromley, D.B. (2000) Psychological aspects of corporate identity, image and reputation, Corporate Reputation Review, 3, 240252. Burt, S.L. and Sparks, L. (2002) Corporate branding, retailing and retail internationalisation, Corporate Reputation Review, 5, 194212. Burt, S.L. and Sparks, L. (2003) E-commerce and retail process: A review, Journal of Retailing and Consumer Services, 10, 275286. Butler, P. and Peppard, J.C. (1998) Consumer purchasing on the Internet: Processes and prospects, European Management Journal, 16(5), 600610. Byrne, B.M. (1994) Structural Equation Modelling with EQS and EQS/Windows: Basic Concepts, Applications and Programming, Sage, Beverly Hills, CA. Byrne, B.M. (2001) Structural Equation Modelling with AMOS: Basic Concepts, Applications, and Programming, Lawrence Erlbaum Associates, London, UK. Campman, M. (2001) Branding.com: Building brand leadership in the new economy, Corporate Reputation Review, 2(3), 200208. Cheng, E.W.L. (2001) SEM being more effective than multiple regression in parsimonious development research, Journal of Management Development, 20, 650667.

240

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

Clauser, R.C. (2001) Ofine rules, online tools, Brand Management, 8(4), 270287. Cowles, D.L., Kiecker, P. and Little, M.W. (2002) Using key informant insights as a foundation for e-retailing theory development, Journal of Business Research, 55(8), 629636. Davies, G. (1991) Retailer advertising strategy, International Journal of Advertising, 10, 189204. Davies, G. and Chun, R. (2002) Gaps between the internal and external perceptions of the cor porate brand, Corporate Reputation Review, 5, 144158. Davies, G., Chun, R., Da Silva, R. and Roper, S. (2003) Corporate Reputation and Competitiveness, Routledge, London, UK. Davies, G., Chun, R., Da Silva, R. and Roper, S. (2004) Corporate character scale to assess employee and customer views of organisation reputation, Corporate Reputation Review, 7(2), 125146. Davies, G., da Silva, R., Chun, R. and Roper, S. (2001) The personication metaphor as measurement approach, Corporate Reputation Review, 4, 113127. Davy, J.A., Kinick, A.J. and Scheck, C.L. (1997) A test of job securitys direct and mediated effects on withdrawal cognitions, Journal of Organisational Behaviour, 18, 323349. De Chernatony, L. (1999) Brand management through narrowing the gap between brand identity and brand reputation, Journal of Marketing Management, 15(1/3), 3956. De Chernatony, L. (2001) Succeeding with brands on the Internet, Journal of Brand Management, 8(3), 186195. De Chernatony, L. and Christodoulides, G. (2004) Taking the brand promise online: Challenges and opportunities, Interactive Marketing, 5, 238251. De Chernatony, L. and DallOlmo Riley, F. (1998) Dening a brand: Beyond the literature with experts interpretation, Journal of Marketing Management, 14(4/5), 417443. De Chernatony, L. and Harris, F. (2000) Developing corporate brands through considering internal and external stakeholders, Corporate Reputation Review, 3, 268274. De Chernatony, L. (2002) Would a brand smell any sweeter by a corporate name, Corporate Reputation Review, 5(2/3), 114132. Duarte, M. and Davies, G. (2002) The anatomy of the ford brand, Paper presented at the 6th International Conference on Corporate Reputation, Sheraton Hotel, Boston, MA, 2325 May. Fombrun, C. (1996) Reputation: Realising Value from the Corporate Image, Harvard Business School, Boston, MA. Fombrun, C. and Van Riel, C.B.M. (1996) The reputational landscape, Corporate Reputation Review, 1(2), 513.

Franzen, G. and Bouwman, M. (2001) The Mental World of Brands, World Advertising Research Centre, Henley-on-Thames. Goldsmith, E.R., Lafferty, A.B. and Newell, J.S. (2000) The inuence of corporate credibility on consumer attitudes and purchase intent, Corporate Reputation Review, 3, 304318. Gray, E.R. and Balmer, J.M.T. (1998) Managing corporate image and corporate reputation, Long Range Planning, 31, 695702. Graver, M.S. and Mentzer, J.T. (1999) Logistics research methods: Employing structural equation modelling to test for construct validity, Journal of Business Logistics, 20, 3357. Grifn, J.J. (2002) To brand or not to brand? Tradeoffs in corporate branding decisions, Corporate Reputation Review, 5(2/3), 228240. Hart, C.A., Doherty, N. and Ellis-Chadwick, F. (2000) Retailer adoption of the Internet Implications for retail marketing, European Journal of Marketing, 34(8), 954974. Hair, F.J., Anderson, E.R., Tatham, L.R. and Black, C.W. (1998) Multivariate Data Analysis, Prentice-Hall Inc., New Jersey. Heijden, der van H. and Verhagen, T. (2004) Online store image: Conceptual foundations and empirical measurement, Information & Management, 41, 609617. Herbig, P. and Hale, B. (1997) Internet: The marketing challenge of the twentieth century, Internet Research Electronic Networking Applications and Policy, 17(2), 95100. Hoffman, D. (2000) The revolution will not be televised: Introduction to the special issue on marketing science and the Internet, Marketing Science, 9(1), 13. Ind, N. (1997) The Corporate Brand, Great Britain, Macmillan Press Ltd, London. Ind, N. and Riondino, M.C. (2001) Branding on the web: A real revolution, Brand Management, 9, 819. Jarvenpaa, S.L. and Todd, P.A. (1997) Consumer reactions to electronic shopping on the world wide web, International Journal of Electronic Commerce, 1(2), 5988. Jarvenpaa, S.L. and Tractinsky, N. (1999) Consumer trust in an Internet store: A cross-cultural validation, Journal of Computer Mediated Communication, 5(2). [Online] Available: www.ascusc.org/jcmc/vol5/ issue/jarvenpaa.html. Jreskog, K.G. and Srbom, D. (1982) Recent developments in structural equation modelling, Journal of Marketing Research, 19(4), 404416. Keller, K.L. (2000) Building and managing corporate brand equity, in M. Schultz, M.J. Hatch and M.H. Larsen (eds.), The Expressive Organisation: Linking

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

241

Online and Ofine Corporate Brand Images

Identity, Reputation and the Corporate Brand, Oxford University Press, Oxford, UK, pp. 115138. Keller, K.L. (2003) Brand synthesis: The multidimensionality of brand knowledge, Journal of Consumer Research, 29, 595600. Keynote (2004) Bookselling, Market Report, 12th edn., May 2004. King, S. (1973) Developing New Brands, Pitman, London, UK. Kline, R.B. (1998) Principles and Practices of Structural Equation Modelling, The Guildford Press, New York. Korgaonkar, P.K. and Wolin, L.D. (1999) A multivariate analysis of web usage, Journal of Advertising Research, 39(2), 5368. Lambin, J-L. (1997) Strategic Marketing Management, McGraw-Hill, Maidenhead, UK. Lindquist, D.J. (1974) Meaning of image, Journal of Retailing, 50, 2938. Long, S.J. (1983) Conrmatory Factor Analysis: A Preface to Lisrel, Sage, Newbury Park, CA. Markham, V. (1972) Planning the Corporate Reputation, George Allen and Unwin, London, UK. Martineau, P. (1958) The personality of retail store, Harvard Business Review, 36, 4755. McDaniel, C. and Gates, R. (2003) Marketing Research Essentials, College Publishing (Thomson Learning), South-Western, USA. McEnally, M.R. and de Chernatony, L. (1999) The evolving nature of branding: Consumer and managerial considerations, Academy of Marketing Science Review, 1999, 124. Merrilees, B. and Louise Fry, M. (2002) Corporate branding: A framework for E-retailers, Corporate Reputation Review, 5, 213225. Mintel Report (2004) Leisure shopping-UK, Leisure Intelligence, October 2004. Murphy, J.M. (1987) Branding A Marketing Tool, The Macmillan Press Limited, London. Parasuraman, A., Zeithaml,V.A. and Malhotra, A. (2005) E-SQUAL, Journal of Service Research, 7, 213233. Patterson, M. (1999) Re-appraising the concept of brand image, The Journal of Brand Management, 6, 409426. Peterson, R.A., Balasubramanian, S. and Bronnenberg, B.J. (1997) Exploring the implications of the Internet for consumer marketing, Academy of Marketing Science, 25(4), 329346. Phillips, D. (2001) Managing reputation in cyberspace, Journal of Communication Management, 5(3), 300304. Porter, S.S. and Claycomb, C. (1997) The inuence of brand recognition on retail store image, The Journal of Product and Brand Management, 6(6), 373384.

Reedy, J., Schullo, S. and Zimmerman, K. (2000) Electronic Marketing: Integrating Electronic Resources into the Marketing Process, Dryden, USA. Rojas-Mendez, J.I., Podlech, I.E. and Olave, E.S. (2004) The ford brand personality in Chile, Corporate Reputation Review, 7, 232251. Sandberg, J. (1998) It isnt entertainment that makes the web shine; Its dull data, Wall Street Journal, 20 July, p. 1. Schmidt, P. (2005) SEM and longitudinal Data, The 38th Essex Summer School in Social Science Data Analysis and Collection, University of Essex, Essex, 812 August. Schultz, M. and de Chernatony, L. (2002) Introduction: The challenges of corporate branding, Corporate Reputation Review, 5, 105112. Singh, J. (1991) Redundancy in constructs: Problem, assessment, and an illustrative example, Journal of Business Research, 22(3), 255280. Spector, A.J. (1961) Basic dimensions of the corporate image, Journal of Marketing, 25(6), 4751. Starr, M.K. and Robinson, J.R. (1978) A loyalty group segmentation model for brand purchasing simulation, Journal of Marketing Research, 15(3), 378383. Stuart, H. and Jones, C. (2004) Corporate branding in marketspace, Corporate Reputation Review, 7(1), 8493. Supphellen, M. and Nysveen, H. (2001) Drivers of intention to revisit the websites of well-known companies, International Journal of Market Research, 43(3), 341352. Syzmanski, D.M. and Hise, R.T. (2000) E-Satisfaction: An initial examination, Journal of Retailing, 76, 309322. The Marketing Pocket Book (2005) The Advertising Association (AA), NTC Publication, UK. Thomas, J.W. (1998) The brave new world of Internet marketing, Direct Marketing, 6(9), 4042. Uncles, M.D. (2001) Interactive electronic marketing and brand management, Journal of Brand Management, 8(4/5), 245254. Urde, M. (2003) Core value-based corporate brand building, European Journal of Marketing, 37(7/8), 10171040. Van Riel, C.B.M. and Van Bruggen, G.H. (2002) Incorporating business unit managers perspectives in corporate branding strategy decision making, Corporate Reputation Review, 5, 241251. Wolnbarger, M. and Gilly, C.M. (2003) eTailQ: Dimensionalising, measuring and predicting eTail quality, Journal of Retailing, 79, 183198. Wolff, M. (1998) Burn Rate: How I Survived the Gold Rush Years on the Internet, Simon and Schuster, New York. Zajonc, R.B. (1980) Feeling and thinking preferences need no inferences, American Psychologist, 35, 151175.

242

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Syed Alwi and Da Silva

APPENDIX A: ONLINE CBI [THE FIRST ORDER]

Honest, Sincere, Straightforward, Open, Trustworthy, Reassuring, Agreeable, Social-responsible, Friendly, Pleasant, Supportive, Concerned, Pleasant, Friendly, Agreeable, Supportive

Agreeableness

Daring, Trendy, Exciting, Cool, Young, Extrovert, Imaginative, Up-to date, Innovative Secure, Technical, Corporate, Achievement-oriented, Ambition, Reliable, Hardworking, Leading

Enterprise

Competence

Selfish, Authoritarian, Controlling, Arrogant, Inward-looking, Aggressive

Ruthlessness

Charming, Exclusive, Refined, Stylish, Prestige, Elegant, Snobby, Elitist

Chic

Casual, Simple, Easy-going

Informality

Masculine, Tough, Rugged

Machismo

Fit Indices Online

3841 P<0.001

Df 1106

/df 3.5

GFI 0.731

TLI 0.657

CFI 0.678

RMSEA 0.07

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00

Vol. 10, 4, 217244

Corporate Reputation Review

243

Online and Ofine Corporate Brand Images

APPENDIX B: OFFLINE CBI [THE FIRST ORDER]

Honest, Sincere, Straightforward, Open, Trustworthy, Reassuring, Agreeable, Social-responsible, Friendly, Pleasant, Supportive, Concerned, Pleasant, Friendly, Agreeable, Supportive

Agreeableness

Daring, Trendy, Exciting, Cool, Young, Extrovert, Imaginative, Up-to date, Innovative Secure, Technical, Corporate, Achievement-oriented, Ambition, Reliable, Hardworking, Leading

Enterprise

Competence

Selfish, Authoritarian, Controlling, Arrogant, Inward-looking, Aggressive

Ruthlessness

Charming, Exclusive, Refined, Stylish, Prestige, Elegant, Snobby, Elitist

Chic

Casual, Simple, Easy-going

Informality

Masculine, Tough, Rugged

Machismo

Fit Indices Offline

3831 p<0.001

df 1106

/df 3.5

GFI 0.735

TLI 0.684

CFI 0.703

RMSEA 0.07

244

Corporate Reputation Review

Vol. 10, 4, 217244

2007 Palgrave Macmillan Ltd. 1363-3589 $30.00