3.2.1. Corporate Culture

Outcome orientation and stability culture dimensions were measured with seven and three items, respectively, adapted from (O'Reilly et al., 1991) [18] the Organisational Culture Profile (see Appendix A). The confirmatory factor analysis (CFA) was conducted with ten items under two culture dimensions. The CFA results indicate a poor model fit to the dataset (CMIN/DF = 3.57; GFI = 0.888; AGFI = 0.818; CFI = 0.882; RMSEA = 0.113; SRMR = 0.066). After re-specifications of the model by connecting error terms based on modification indices, we found a final model with an acceptable level of the goodness of fit indices(the recommended threshold scores for the assessment of good SEM model fit to the data set are CMIN/DF < 5.0; GFI > 0.90; AGFI > 0.80; CFI > 0.90 [52]): CMIN/DF = 2.180; GFI = 0.936; AGFI = 0.887; CFI = 0.951; RMSEA = 0.077; SRMR = 0.048. The Cronbach alpha reliability scores (α) of the seven-item outcome orientation and three-item stability culture measures are 0.824 and 0.745, and thus exceeded the minimum cut-off of 0.70 [47].

#### 3.2.2. Combating the Bribery Aspect of Corporate Sustainability

A six-item construct based on the OECD (2011) [6] seven principles relating to the combating bribery was used to measure the extent to which combating the bribery aspect of corporate sustainability was being practiced within the sampled organisations (see Appendix A). The respondents were asked to indicate the extent to which each item explained the current organisational practices on a five-point Likert scale, with the anchors of 1 'Not at all' and 5 'To a great extent'. In order to validate this measure, we conducted CFA. The initial model CFA results (CMIN/DF = 2.331; GFI = 0.967; AGFI = 0.924; CFI = 0.977; RMSEA = 0.082; SRMR = 0.030) indicate a good model fit to the dataset. The Cronbach alpha score (α = 0.866) of the six-item measure exceeded the minimum cut-off of 0.70.

#### 3.2.3. Organisation Performance

"Organisation performance" was measured from the perspective of both financial and non-financial aspects. A three-item construct for each of the financial and non-financial performance was adapted from Kaynak and Kara, 2004 [53]. The results of the initial model CFA (CMIN/DF = 3.446; GFI = 0.956; AGFI = 0.886; CFI = 0.963; RMSEA = 0.111; SRMR = 0.037) indicated a poor model fit to the dataset. Re-specification of the model was then taken after deleting one item: "we have a lower employee turnover rate than our competitors" from the non-financial performance measure because its factor loading score (0.38) was less than the acceptable limit of 4.0 to be considered for the CFA analysis. The revised CFA model with the five items under both measures provided a good model fit to the dataset: CMIN/DF = 5.791; GFI = 0.957; AGFI = 0.841; CFI = 0.962; RMSEA = 0.155; SRMR = 0.031. The Cronbach alpha reliability scores (α) of the 3-item financial performance measure and the two-item non-financial performance measure were 0.892 and 0.71, respectively, and hence met the minimum cut-off score of 0.70.

#### *3.3. Convergent and Discriminant Validity*

In addition to the content validity of the measures of this study tested by the CFA, we tested the convergent and discriminant validity of the measures. The convergent validity of the measures was tested with respect to their path coefficients, standard errors (S.E.), and *t*-values [54]. The estimated factor loadings of all the measures are more than twice their respective S.E., and their *t*-values (*t* > 2) are significant at the level of 0.01 (see Appendix A for all measures), and composite reliability scores of all measures are more than 0.90, thereby supporting their convergent validity of the measures. The discriminant validity of the measures was tested by comparing the measures' (Cronbach's, 1951) [55] alpha reliability scores with their correlations, with the other scales of the study. The scales' reliability scores are higher than their correlations with other scales (see Table 2), and thus provide evidence of the presence of the discriminant validity of the measures.


**Table 2.** Inter-construct correlation, descriptive statistics, and Cronbach alphas and composite reliability scores.

n = 201. \*\* and \* correlation is significant at the 0.01 and 0.05 levels (2-tailed).

#### *3.4. Common Method Bias Test*

As suggested by Podsakoff et al., 2003 [50], this study followed numerous pre-survey techniques to minimise the likelihood of common method bias problems, including an extensive review of the literature to come up with established and validated scales to measure the variables of this study, drafting a questionnaire with a simple and easy language with the necessary clarification that helps minimise ambiguity, and mentioning that the respondent's identity will be kept confidential. This entire process indicates that social desirability bias is less likely of an issue in this study [32].

In addition, this study employed a post-survey technique to check common method bias issues, if any were present. In this regard, we conducted Harman's (1967) one (single) factor test [40]. The exploratory factor analysis (EFA) (varimax rotation) was resulted in

four factors having eigenvalues of more than 1, and explained 61.02% of the total variance of 100%. While no factor (31.58% maximum) was explained by more than or equal to 50% of the total variance, thereby indicating that a common method bias is less likely an issue in this study.

#### **4. Results and Discussion**

#### *4.1. Structural Equation Model*

We developed structural equation modelling (SEM) by utilising the AMOS software, as suggested by Hair et al., 2010 [52]. We conducted CFA models first and then the SEM. We first developed a base model. The base model results are reported under "Model A" in Table 3. The base model's goodness-of-fit indices indicated a poor model fit to the data set: CMIN/DF = 2.874; GFI = 0.819; AGFI = 0.772; CFI = 0.807.

**Table 3.** Summary of SEM results.


Note: *p*-values having \*\*\* and \*\* indicate statistically significant at 0.01 and 0.05 levels, respectively (2-tailed).

The initial model was revised by connecting error terms based on the highest possible modification indices recommended by Anderson and Gerbing, 1988 [54], and resulted in a good model fit to the dataset: (CMIN/DF = 2.467; GFI = 0.874; AGFI = 0.803; CFI = 0.953) (see Figure 2 and Model B in Table 3).

**Figure 2.** Results of the structural equation model among the study variables. Note: \*\*\* and \*\* indicate statistical significance at 0.01 and 0.05 levels, respectively (2-tailed).

4.1.1. Association between Outcome-Orientated Corporate Culture and the Bribery Combating Aspect of Corporate Sustainability

The SEM results reported in Figure 2 and Table 3 (Model B) reveal that the outcomeoriented culture is significantly and positively associated with the bribery combating aspect of corporate sustainability (β = 0.490; *p* = 0.000). Hence, hypothesis H1 is supported. The findings indicate that the culture of promoting competitiveness among employees, responding to their high expectations of performance, as well as providing them rewards based on their individual performance may refrain them to take bribes from and/or offer bribes to business partners. Hence, organisations that want to combat bribery practices, and thus promote corporate sustainability practices, are suggested to pay attention to develop an outcome-oriented culture by developing a culture of respecting employees' high expectations of performance, providing them rewards based on performance, and motivating them to be action-oriented. Our findings are in line with what other researchers discourse about the role of organisational outcome orientation culture on corporate sustainability practices [41,42]. Researchers [48,56] argue that the existence of outcome orientation culture is likely to promote employee sustainability practices. Our empirical findings contribute to the existing sustainability literature, with insights into the role of suitable organisation culture towards enhancing sustainability practices.

4.1.2. Association between Organisation Stability Culture and the Bribery Combating Aspect of Corporate Sustainability

The SEM results reported in Figure 2 and Table 3 (Model B) highlighted that organisational stability culture is significantly and positively associated with the bribery combating aspect of corporate sustainability (β = 0.224; *p* = 0.003). Hence, hypothesis H2 is supported. Furthermore, the associations between the stability culture and organisational financial (β = −0.006; *p* = 0.945) and non-financial performance (β = −0.050; *p* = 0.571) are statistically insignificant. The findings indicate that organisational stability culture has a positive influence on combating bribery practices in organisations. More specifically, if organisations provide employment security, bribery and small facilitation payment seems to be less likely to occur in organisations. This is because job security may lead employees not to be worried too much about the future, and thus there may have less tendency to make money illegally and abruptly by using organisational identity. Accordingly, organisations are advised to build a stability culture by providing employees with job security, and thus enhance corporate sustainability practices by reducing illegal and unethical practices, including taking and offering bribes. These findings contribute to the very limited corporate sustainability literature with the role of stability culture towards promoting corporate sustainability practices. Thus, the findings could be useful for organisational decision makers and policymakers, to devise suitable polices to combat bribery practices.

4.1.3. Association between the Bribery Combating Aspect of Corporate Sustainability and Organisational Financial and Non-Financial Performance

As reported in Figure 2 and Table 3 (Model B) and summarised in Table 4 Panel B, the bribery combating aspect of corporate sustainability practices is significantly and positively associated with financial performance (β = 0.262; *p* = 0.007) and non-financial performance (β = 0.307; *p* = 0.004). Hence, both hypotheses H3 and H4, are supported. The findings indicate that the bribery combating aspect of corporate sustainability practices drives organisations towards achieving their short-term financial goals, such as profit, sales, and ROI goals, as well as long-term strategic (non-financial goals) goals, such as the goal of ensuring high-quality product and service offering and the goal of achieving a higher customer retention rate than competitors. These findings conform with what previous studies found (see [2,8–10,41,52]). For example, Bhuiyan et al., 2020 [12], found a positive association of corporate sustainability practices (minimising the illegal activities aspect of sustainability practices) with organisations' financial performance. While others [46,47,57] found a positive association between corporate sustainability practices and organisation non-financial performance, such as reputation, innovation and differentiation, customer

satisfaction, and employee commitment. Therefore, the findings of the study may be useful for attracting the attention of policymakers and organisational decision makers in order to minimise the likelihood of bribery practices, and thus enhance organisational performance.

**Table 4.** Summary of SEM results with control the organisation size (number of employees) as the control variable.


Note: *p*-values having \*\*\* and \*\* indicate statistical significance at 0.01 and 0.05 levels, respectively (2-tailed).
