*3.2. Measures*

#### 3.2.1. Religiosity

Religiosity was measured using four aspects: the cognitive, intrinsic, extrinsic affective, and behavioural aspects of religiosity, developed by Mazereeuw-van der Duijn et al. [9]. The cognitive aspect of religiosity was measured using five questions concerning the respondents' conception of God, views on individuals, and expectations regarding eternity. The answers to these questions were recalculated to the same scale. Since the internal consistency of the answers proved to be good (Cronbach's alpha = 0.94), the answers were taken together to form the variable 'cognitive aspect of religiosity'. The effective element of religiosity was measured using the intrinsic/extrinsic religiousness scale [43]. The scale consists of 16 statements (8 statements for each), measuring the intrinsic and extrinsic inspiration of the participants toward their religious beliefs. The internal consistency of the intrinsic motivation was 0.84. Averaging the scores on eight statements regarding

participants' extrinsic enthusiasm toward their religious beliefs produced the extrinsic measure. The internal consistency of both intrinsic and extrinsic motivation was 0.73.

The behavioural aspect of the religiosity of the respondents was assessed using five questions including the attendance of religious ceremonies, participation in other activities of the religious community, and time spent in private prayer, work-related prayer, and meditation. Since these five items strongly correlate with each other, we created one measure, named the intensity of religious behaviour, based on the average score of these five questions [9]. The internal consistency of this measure was 0.69 [43]. A correlation analysis was conducted on the four measures of religiosity to control any potential crossrelationships between the various aspects of religiosity. Consistent with the findings of Mazereeuw-van der Duijn et al. [9], Table 1 demonstrates a positive, weak, but significant correlation between the cognitive and extrinsic effects. The Intrinsic affective aspect of religiosity also proves to be a positive and significant correlation with the extrinsic effect and the behavioural aspects of religiosity. The Extrinsic effect also indicates a positive and significant correlation with the behavioural aspects of religiosity. This implies that there are no strong and significant correlations between the aspects of religiosity.

**Table 1.** Means, Standard Deviations, and Correlations of the Aspects of Religiosity.


Note: \*\* *p* < 0.01.

#### 3.2.2. Attitudes toward CSR

The attitudes toward CSR were measured as financial, legal, ethical, and philanthropic responsibilities, respectively. The study adopted a scale developed by Aupperle et al. [44]. The participants were asked to allocate 10 points to each of the five groups of four statements measuring the importance the participants attributed to each of these four CSR attitudes. Following Aupperle et al.'s [44] recommendation, we used this forced-choice approach to minimise a social desirability response bias. We also created four reduced-scale items for financial, legal, ethical, and philanthropic orientation and subjected the reduced-scale items to confirmatory factor analysis. The internal consistencies of the factors were 0.87 for financial orientation, 0.71 for legal orientation, 0.65 for ethical orientation, and 0.76 for philanthropic orientation.
