Abstract: Reputation Quotient (RQ) is one among the most prominent instrument that is proposed to measure corporate reputation. The RQ have suggested that more robust measure than other instruments due to it can be used elicit the perceptions of multiple stakeholders. However, yet few study validate the RQ scale based on human resource management view. Present study attempted to fulfill this gap. With data collected in banking industry, 339 branch managers, this study contribute to the existing corporate reputation literature by provide empirical evidences of validity. Results also reveal that most of the RQ dimensions were strongly associated with important outcome variables with a few exceptions. Contribution and future research also is presented.

INTRODUCTION

A company’s reputation is critical to sustainability in a globalized economy. Cross-disciplinary theorists view corporate reputations as strategic assets and as a source of economic value (Fombrun, 2001). A good feeling about company, its activities, workplace, past performance and future prospects by key stakeholders can leads to positive stakeholder’s decision about company (Fombrun et al., 2000).

Research have demonstrated that more favorable reputation increase organization performance (Return on Average Assets: ROA) (Deephouse, 2000) and related to investor’s and customer’s behavioral loyalty (Helm, 2007; Walsh and Beatty, 2007; Walsh et al., 2009) and employee’s turnover intention (Riordan et al., 1997). Thus, in order to manage their reputation, company need to continuously measure and evaluate it.

Whereas, the topic of measuring corporate reputation has receiving increased attention in the multi-disciplinary literatures, Fombrun et al. (2000) founded that measures of corporate reputation currently in widespread use suffer from fundamental methodological and conceptual weakness.

To date, most of the popular measurement instruments have a tendency to define reputations on the basis of the perceptions of restricted set of financially oriented stakeholders (Fombrun et al., 2000). To address these limitations, Fombrun and Van Riel (1997, 2004) and Fombrun et al. (2000) define corporate reputation based on integrative view and proposed new instruments, the Reputation Quotient (RQ) to measure corporate reputations. By integrative view definition, RQ scale measure corporate reputation on the basis of the perceptions of multiple stakeholders. This RQ scale is tested and demonstrated empirical validity and reliability in the 30 best regarded companies in the Unite State. However, they suggested that further study is needed to validate the RQ measure of corporate reputation cross-culturally. Research on the empirical correlates of the RQ will also help to establish the measure and to strengthen its underpinnings (Fombrun, 2001).

The literature reviews indicate that although, recent researches interest in validate the Fombrun et al. (2000)’s RQ measure of corporate reputation, yet few studies have directly validate it in the Human Resource Management (HRM) context.

Most of recent researches do not directly validate the RQ scale but rather adapt and used the RQ scale as the start point for developing their instrument (Schultz and Nielsen, 2002; Walsh and Wiedmann, 2004; Aperia et al., 2004; Walsh and Beatty, 2007). This study contributes to this gap in the research by directly validated the RQ measure of corporate reputation in HRM view and Thai context.

The concept of corporate reputation: In the corporate reputation literature, there are three main conceptual streams (Berens and van Riel, 2004). The first streams define corporate reputation based on social expectation, the expectations that people have regarding the behavior of companies. The second main stream is based on the concept of corporate personality; the personality traits that people attribute to companies. The third main stream use the concept of trust as its starting point that is the perception of a company’s honesty, reliability and benevolence. Based on Baren and van Riel’s study, the researchers noted that the first main stream is more frequently used than the second and the third main streams. Therefore, the researchers interest in the background of the first conceptual stream. Many researchers cluster different corporate perception under the different expectations that stakeholders have regarding the behavior of companies in society. Of these studies the Reputation Institute’s Reputation Quotient (Fombrun et al., 2000) is among the most prominent. Based on analysis of similarities of the reputation concept across fields, integrative view, Fombrun and van Riel (1997) define corporate reputation as a collective representation of a firm’s past actions and results that describes the firm’s ability to deliver valued outcomes to multiple stakeholders. A corporate reputation, it is suggested, calibrate a firm’s relative standing internally with employees and externally with its other stakeholders in its competitive and institutional environments (Fombrun et al., 2000).

Underlying integrative view (Fombrun et al., 2000) proposed a new instrument the Reputation Quotient (RQ) to measure corporate reputation. Based on the focus groups and pilot study, they found that the people justify their feeling about companies on one of 20 attributes that can be grouped into six dimensions (Table 1). The first dimension, emotional appeal refers to good feeling and degree of trust the company inspires in people. Second, products and services refer to the perception of the value, quality, innovation and reliability of the company’s products and services.

Third, vision and leadership captures perceptions of a clear vision and strong leadership. Fourth, workplace environment refers to the perception of how well the company’s managed, how it is to work for and the quality of its employees. Fifth, social and environmental responsibility captures perceptions of the company as a good citizen in its dealing with communities, employees and the environment. Finally, financial performance describes perceptions of the company’s profitability, prospects and risk (Fombrun and van Riel, 2004). The results indicated that the RQ 6 sub scale is a valid instrument for measuring corporate reputations and can be used to benchmark companies across industries (Fombrun et al., 2000).

Corporate reputation and important employee-outcome variables: Fombrun and Shanley (1990) indicate that corporate reputation is a signals that an organization transmits to its various stakeholder groups. These signals influence making choice decisions of potential stakeholders. Previous study consistently demonstrated that corporate reputation is related to stakeholder decisions about organization such as customers’ loyalty and word of mouth behaviors (Walsh and Beatty, 2007; Walsh et al., 2009), investors’ behavioral loyalty (Helm, 2007) employee’s attitudes and behaviors toward their organization (Dutton and Dukerich, 1991).

From the HRM perspective, organizational loyalty and turnover intention are potentially consequences of corporate reputation. Organizational loyalty refers to the behaviors reflect the identification with and allegiance to an organization, demonstrated by defending the organization against threats conducing to its good reputation and cooperating with others to serve the interest of the whole (Van Dyne et al., 1994).

According to Fombrun (1996) not only does a good reputation attract loyal customers, it also makes loyal employees. Fombrun (1996) have stated that a good reputation builds employee’s loyalty by increasing the willingness of employees to corporate with unusual requests and by fostering teamwork and a sense of shared destiny. Similarly, Chun (2005) argued that when managers and employees perceive that the organization has a good reputation, they feel more security and happy which leads to organizational loyalty.

In addition, Martin and Hetrick (2006) indicate that employees’ view of positive perception is likely to influence their willing to remain with the company. Especially, the talented people are more likely attracted by high performing and reputable companies. Consistently with Riordan et al. (1997) have argued that the employee’s reputation perceptions of the organization’s actions toward its various stakeholders influences the employee’s intention to remain or leave the firm.

If the employee has negative perceptions of these actions, disassociation will likely follow. If perceptions are positive, the employee will likely remain with them. The relationship between corporate reputation and employee’s turnover intention are supported by several previous study (Riordan et al., 1997; Turban and Cable, 2003; Cole and Bruch, 2006; Helm, 2007).

Thus in this study, we examine the relationship between RQ, branch managers’ organizational loyalty and turnover intention.

We anticipate that branch managers’ organizational loyalty will be positively correlated with RQ while turnover intention will be negatively correlated. That is the researchers believe that banks with better reputations will have more loyal and less turnover intention branch managers.

Table 1:	Corporate reputation factors
AVE: Average Variance Extracted; cr: composite reliability aItems removed after data screening; bItems were developed in this study; γc = eigenvalues

MATERIALS AND METHODS

Sample: The data were collected from branch manager sample in banking business in Thailand. The branch managers were chosen as a target population because they are valuable employees in banking business. Financial performance of branch bank mostly depends on the capability of branch manager. Previous research indicates that firm performance was significantly and negatively related to managers’ turnover (Kang and Shivdasani, 1995).

In addition when a branch manager leaves, organization must incur, time and uncertainty on recruit and training to replace capable branch manager. Overall, 234 participants (69%) were men and 105 participants (31%) were women. One hundred and eighty five participants (54%) had a college degree and 118 participants (33.7%) had a master degree. The experience working in the bank ranged from 1-34 years (mean = 20.12, SD = 8.43). About 114 (33.6%) had a salary >1,560 dollar, 92 participants (27.1%) and 90 participants had a salary 938-1,250 dollar and 1,250-1,560 dollar, respectively.

Instrumentation: In this study, the corporate reputation was measured using Fombrun et al. (2000)’s 20-items Reputation Quotient (RQ) and two items were added to the emotional appeal dimension as shown in Table 1. However, two items were deleted from RQ scale before the factor analysis is run. Because of the primary data screening shown that it correlates very highly with other items in its dimension (r>0.80) (Field, 2000).

The first item (i.e., stands behind its products and services) in products and services dimension and the second item (i.e., has a strong record of profitability) in financial performance dimension. Therefore, corporate reputation in this study was measured using 20 items. About 18 items from Fombrun et al. (2000)’s RQ scale and 2 items were developed in the study.

Data analysis: The data were split in two samples for two-step analysis. First sample (n = 170) was analyzed by Exploratory Factor Analysis (EFA) with principle components extraction and direct oblimin rotation. This makes fair the number of component and proves the psychometric properties of the reputation quotient. Direct oblimin was chosen due to we expected the dimensions to be correlated. As Stevens (1992)’s recommended factor loading with an absolute value >0.4 was chosen as the cut off determining whether an item defined a factor/components. In addition, the eigenvalue >0.7 was used to define factor (Jolliffe, 2002). Because the Kaiser’s guideline (eigenvalue >1) is accurate when the sample size exceeds 250 (Stevens, 1992). In the first sample, the researchers also conducted first-order Confirmatory Factor Analysis (CFA) in order to assess the fit of the factor structure model identified from EFA.

As the recommendation of Hu and Bentler (1995, 1998) and Sharma et al. (2005), the following fit indexes were used to assess the fit of the models: Chi-square/df ratio, Root Mean Square Error of Approximation (RMSEA), Standardized Root Mean Square Residual (SRMR), the Non-Normal Fit Index (NNFI), the Incremental Fit Index (IFI) and the Comparative Fit Index (CFI). The use if multiple fit indexes is generally advisable in order to provide convergent evidence of model fit. However, the Goodness of Fit Index (GFI) was not chosen as fit index because it is more affected by sample size (Bentler, 1990; Hu and Bentler, 1995, 1998). The Chi-square/df ratio is within range of 2-5 indicating an acceptable fit model. The values of NNFI, IFI and CFI >0.95 indicating a well-fitting model (Schermelleh-Engel et al., 2003). The values of RMSEA and SRMR were <0.05 indicating a well fitting model but <0.10 may be interpreted as acceptable (Browne and Cudeck, 1993).

Second sample (n = 169) was used to validating the measurement model of reputation quotient identified with CFA in the first sample. In addition, the second order confirmatory factor analysis of reputation quotient was used to assess the fit if the theoretically-based hierarchical model of corporate reputation.

RESULTS

Sample 1: Table 1 shows the factor loading of all items and coefficient alphas for each of the six RQ sub-scales. As demonstrated in the Table 1, EFA results indicated six interpretable factors (eigenvalue >0.7). All factors loading exceed 0.4. Only item 8 loaded ambiguously on the factors product/service and factors vision/leadership with cross-factor loading >0.4 (0.41 and 0.46, respectively). This may be occurred because of the relationship between factors product/service and factors vision/leadership. However, the difference of factor loading value is relatively few. The researchers consider the importance of a theoretical or face validity more than a statistical level (factor loading). Therefore, item 8 is considered as an indicator of product and service scale.

Next, the six-factor structure identified with EFA was tested using CFA. First-order CFA was conducted by using LISREL. The model fit indexes revealed an adequate overall fit (Chi-square/df ratio = 2.53, RMSEA = 0.09, SRMR = 0.05, NNFI = 0.96, IFI = 0.97, CFI = 0.97). All 20 items had standardized loading ranged from 0.77-0.88 as in Table 1. Furthermore, the Average Variance Extracted (AVE) of each dimensions (ranged from 0.62-0.71) was above 0.50 hence, indicating convergent validity (Fornell and Larcker, 1981). The AVE for each dimensions also exceeds the squared correlation between both dimensions (ranged from 0.36-0.69), suggesting the discriminant validity of the dimensions in the employee-based corporate reputation (Fornell and Larcker, 1981).

Sample 2: The six factor structure identified with first sample was validated by using CFA. Results in Table 1 reveal that all items had standardized loading above threshold of 0.40 (ranged from 0.76-0.93). The fit indexes suggest that the measurement model adequately represents the input data (Chi-square/df ratio = 2.31, RMSEA = 0.08, SRMR = 0.05, NNFI = 0.97, IFI = 0.98, CFI = 0.98). Furthermore, all six reputation dimensions had AVE exceeds 0.50 (ranged from 0.63-0.82). Thus, indicating convergent validity.

It also had good reliability, with composite reliabilities >0.60 (Bagozzi and Yi, 1988). These provide validity evidence in favor of the indicators used to represent the employee-based corporate reputation construct. In addition, the second-order factor model fit the data adequately well (Chi-square/df ratio = 2.39, RMSEA = 0.09, SRMR = 0.06, NNFI = 0.97, IFI = 0.98, CFI = 0.98). However, it provides a poorer fit than the first order or multidimensional model. Result of second-order CFA suggests that the emotional appeal, products and services, vision and leadership, workplace environment, social and environmental responsibility and financial performance factors have a higher order factor; namely corporate reputation.

Based on sample 2, the researchers also assess nomological validity of RQ in the organizational view. Nomological validity was established by examining the correlation between the measure and other related constructs based on the theoretical view (Churchill, 1995). Corporate reputation is associated with employee’s organizational loyalty and turnover intention.

Table 2:	Organizational loyalty and turnover intention factors

Table 3:	Correlation coefficients of RQ scales and related constructs
All Pearson’s correlation coefficients are significant at 0.05; ns = non-significant. Correlation matrix of sample 2 presented in parentheses

Therefore, organizational loyalty and turnover intention were chosen as employee outcome scales. The organizational loyalty and turnover intention were operationalized with 4 and 5 items, respectively. Items for the both outcomes measure were based on prior items in the past literature (Van Dyne et al., 1994; Cole and Bruch, 2006). Table 2 shows the standardized loading of both outcome constructs (ranged from 0.64-0.87), AVE (0.53 and 0.59) and composite factor reliability (0.85) based on CFA both sample 1 and 2 which clearly confirmed the appropriateness of the operationalization.

Correlation analysis was conducted to assess nomological validity. Table 3 shows that 27 of the 28 correlations between the reputation dimensions and the employee outcome variables were significant at p<0.05. The measures correlated in a manner predicted by theory, thus supporting nomological validity. Result by dimensions reveal that the RQ sub scales correlated with turnover intention more than organizational loyalty. The dimension emotional appeal had the strongest and most consistent correlations for the RQ dimensions with the outcome variables.

DISCUSSION

Results indicate that the Fombrun et al. (2000)’s RQ is a valid instrument for measuring corporate reputation based on internal stakeholders’ perception and in Thai context. This study serves as one of the first studies to validating the RQ six-sub scales in the HRM view. The RQ scale enables human resource managers to study and measure employees’ perception of corporate reputation at different levels of abstraction. According to Walsh and Beatty (2007), using an overall scale without considering the individual dimensions of corporate reputation would be to lose the ability to see how these different dimensions work. Considering both individual dimensions and overall perceptions can leads to more understand how company stands on the different dimensions of reputation in order to know how to improve its corporate reputation.

The finding support Fombrun et al. (2000)’s arguments that corporate reputation is a multidimensional construct that affects employees’ reaction to the company. The researchers found that the RQ dimensions are associated with important employee outcome variables. Results support our view that employees’ view of positive perception about their company’s reputation are influence their attitude, willingness and behaviors. Furthermore, the finding also indicated that emotional appeal dimension had the strongest and most consistent correlation for the RQ dimensions with the outcomes variables. Hence, present study extended the existing literature by provide a greater understanding of the dimensionality and employee outcomes of corporate reputation.

CONCLUSION

Reputation quotient is one among most instruments that is frequently used in corporate reputation research. However, few study has validate it based on the perceptions of internal stakeholders, based on HRM view. This study attempted to fulfill this gap. By validating the RQ scale in banking industries and relating RQ to relevant employee outcome variables. As Fombrun et al. (2000)’s expectation, the researchers hope that the results would help further the study of corporate reputations in HRM perspective by fostering systematic research on the correlations and consequences of corporate reputation.

Although, the researchers have achieved some useful results, a few limitations of the present study should be noted. First, the sample was limited to a single organization within the banking industry. In order to continue exploring the properties of the Fombrun et al. (2000)’s RQ and its generalizability.

RECOMMENDATIONS

Future research should investigate multiple organizations within different industries. Second, the sample size in this study (170 and 169 in sample 1 and 2, respectively) is relative small for factor analysis. As Tabachnick and Fidell (1996)’s rule of thumb, future research should investigate based on the sample size of 300 that as the good sample size which correlations are more reliably estimated.