*4.2. Assessment of the Structural Model*

Once the validity and reliability of the measurement model were assured, the structural model was assessed. Possible collinearity problems between variables were checked through the VIF. Additionally, the coefficient of determination (R2) and the predictive relevance (Q2) of the model were evaluated. The constructs did not have collinearity problems as all VIF values were below the acceptable threshold of 5. The R2 value results indicated that 35% of the variance in SCB was explained by EK, MAT, EI, PSC, and SCBI, and 15% of the variance in SCBI was explained by EK, MAT, EI, and PSC. The predictive relevance of the structural model was established since the Q<sup>2</sup> values of SCB and SCBI were greater than zero.

The results revealed that ten of the formulated hypotheses were supported, whereas twelve were not supported (Table 5 and Figure 2). EK (β = 0.214, *p* = 0.000) was positively related to SCBI, while MAT (β = −0.250, *p* = 0.000) was negatively related to SCBI. Thus, H1 and H2 were supported. EI (β = 0.014, *p* = 0.774) and PSC (β = 0.004, *p* = 0.925)

had no significant effect on SCBI. Hence, H3 and H4 were not supported. The H5a– H5e hypotheses were accepted, as EK (β = 0.186, *p* = 0.000), EI (β = 0.182, *p* = 0.000), PSC (β = 0.196, *p* = 0.000), and SCBI (β = 0.160, *p* = 0.000) were positively related to SCB, and MAT (β = −0.171, *p* = 0.001) was negatively related to SCB. The mediating hypotheses were tested using the indirect effect approach. The results showed that SCBI significantly mediated the relationship between EK and SCB (β = 0.034, *p* = 0.004) and between MAT and SCB (β = −0.040, *p* = 0.001). Therefore, H6a and H6b were supported. Hypotheses H6c and H6d were not supported since the indirect effects between EI and SCB (β = 0.002, *p* = 0.780) and between PSC and SCB (β = 0.001, *p* = 0.927) were insignificant.

**Figure 2.** Structural model.

The moderation outcomes revealed that age moderated only the relationship between SCBI and SCB (β = −0.104, *p* = 0.008). A positive relationship was found between SCB and SCBI for younger respondents (Figure 3). Thus, H8e was supported. Other moderating effects (H7a–H7d, H8a–H8d) were not supported (Table 5).

With respect to the influence of environmental knowledge on sustainable consumption behavioural intention (H1), our findings align with prior studies (Lin et al. 2022; Wang et al. 2014). The indirect effect of EK on SC-related behaviour was also supported, in consonance with past literature (Dhir et al. 2021; Kollmuss and Agyeman 2002; Saari et al. 2021). The negative association of materialism with consumption behavioural intention (H2) confirmed by us is in concurrence with previous studies (Alzubaidi et al. 2021). That is, materialists are not willing to make certain compromises and adopt SCB. However, to form more conclusive comments, this relationship needs to be tested further since significant differences will probably be observed in the willingness for sustainable consumption among highly materialistic consumers and among consumers with low materialism. Besides, differences may also occur regarding consumers' intentions for SC depending on the gender, age, and other demographic variables of the respondents. Thus, Alzubaidi et al. (2021) made an interesting discovery in their study; namely, that younger consumers' materialism did not have a significant effect on behavioural intention, whereas, for older consumers, materialism had a strong negative effect on behavioural intention.


**Table 5.** Structural model estimates.

R2 SCB = 0.352, R<sup>2</sup> SCBI = 0.154, Q<sup>2</sup> SCB = 0.291, Q2 SCBI = 0.120.

**Figure 3.** Simple slope analysis for the interaction effect of age and SCBI on SCB.

The findings related to H3 contradicted those of past studies (e.g., Bruno et al. 2022). Practically, our results did not confirm the positive effect of environmental influences on the intention for individual sustainable behaviour. That was a surprising discovery for us but still, there could also be other influences that might cause consumers to refrain from motivating people's desire for making sacrifices or paying more to consume sustainably. For instance, the different aspects of culture, such as "individualism" and "collectivism", which

reflect the differences in the cultural values of western and eastern countries (Hofstede 2001), could play a determining role in this result.

Our study did not confirm the assumption of the existence of a link between sustainable consumption promotion and sustainable consumption behavioural intention (H4). This result contradicted prior studies that reported an indirect effect of PSC on green product buying (Piligrimiene et al. 2020 ˙ ). A plausible explanation for our findings may be that the public institutions, non-governmental organisations, retailers, and other stakeholders in Bulgaria still fail to fully utilise the promotion possibilities for the purpose of raising consumers' awareness regarding various SCB-related issues. As has already been stressed, the present study is among the few that present empirical evidence of the effect of the SC construct on SCB. Therefore, there is a great need for further research that would enhance the current understanding of this issue.

On the other hand, the tested direct effect of the five factors identified (EK, MAT, EI, PSC, and SCBI) on sustainable consumption behaviour was confirmed. Therefore, the data found support for Hypotheses 5a, 5b, 5c, 5d, and H5e. These findings were also in agreement with past research (Figueroa-García et al. 2018; Wang et al. 2014).

H6a, H6b, H6c, and H6d assumed that SCBI mediated the relationship between EK, MAT, EI, PSC, and SCB. The data supported the SCBI mediating effect between environmental knowledge, materialism, and SCB but not between environmental influences, sustainable consumption promotion, and SCB. Past studies also found support for the mediating effect of behavioural intention in the link between environmental value, environmental knowledge, environmental responsibility, and sustainable consumption behaviour (Sheoran and Kumar 2020), environmental concern and sustainable consumption (Saari et al. 2021), and between consumer attitude, subjective norm, perceived behavioural control, and sustainable consumption behaviour (Matharu et al. 2021). The unsupported mediating role of SCBI in the relationship between EI, PSC, and SCB leads to the conclusion that EI and PSC only affect SCB directly, not indirectly.

In nine hypotheses (H7a, H7b, H7c, H7d, H8a, H8b, H8c, H8d, and H8e), it was assumed that age would moderate the relationships of the influencing factors with sustainable consumption behaviour. The results showed that age moderated only the effect of SCBI on SCB (H8e), which is positive for younger respondents. Our study indicated that younger respondents were more motivated to make certain sacrifices to demonstrate sustainable consumption behaviour. This finding is in contrast with previous literature, which contended that older consumers were more environmentally friendly and aware of environmental issues than young consumers (Witek and Ku ´zniar 2021; Dhir et al. 2021). A partial explanation of the result we obtained could be sought in the increasing knowledge levels of young people having public consciousness, who obtain and distribute information using different modern communication channels. As a recommendation in this respect, a proposal could be addressed to the policymakers that would include, for instance, the targeted distribution of information using social networks aimed at promoting the benefits of the transition towards more sustainable consumption models among these individuals.
