Introduction

Planet Earth is currently experiencing a generalized crisis that must be addressed. The environmental problem is manifested in the intensification of precipitation and droughts, the continuous increase in sea level, the impact on ecosystems, floods, and protracted and intense heat waves in cities (^{Intergovernmental Panel on Climate Change, 2021}). Likewise, several studies have addressed the crisis from a social and economic perspective, which is global and systemic, where growing inequalities and widespread poverty occur as a characteristic of contemporary societies and recurrent global economic crises, social, political, and gender violence, famine, lack of health care, wars, militarization of the population, forced displacement, and organized crime, among other situations are observed (^{Hernández, 2018}; ^{Curiazi & Cajas Guijarro, 2019}; ^{Lang, 2021}).

Given this discouraging context, sustainability, understood as the way to meet the needs of the present without compromising the ability of future generations to meet their own needs (^{Brundtland, 1987}), has emerged as an alternative for action to counteract this civilizational crisis. One must possess broad, deep, and holistic knowledge to understand how economic, social, and environmental systems interact (Stibbe, 2009; Roberston, 2014, cited in ^{Heeren et al., 2016}). To this end, it has become important to develop research on the subject, which is approached from a multi- and transdisciplinary perspective and addresses contents from diverse approaches, ranging from environmental to social, economic, and behavioral issues (^{Acuña-Moraga & Severino-González, 2018}). Thus, sustainable action is recognized as a tripartite responsibility among governments, companies, and consumers that enables real changes in people’s lifestyles (^{Franco Crespo, 2017}). Considering that such collaborative participation should be studied separately, this paper refers only to the participation of consumers in the transformative drive toward sustainability and particularly to the social factors that affect this task.

Background

The sustainable consumer has been defined from various angles, recognizing, for example, their growing demand for information on sustainable products, the origin of raw materials and their forms of distribution, concern for the environmental and social impacts generated by their consumption, and their trend to purchase ecological products and services, even at higher prices than traditional ones (^{Shao et al., 2016}). Sustainable consumption is related to how a person thinks about their purchasing decisions, reflecting on the social and environmental impacts of such decisions and recognizing that the current production and distribution of goods and services are inadequate for a sustainable world. Therefore, this consumer also accepts that the changes in their consumption patterns transform their surroundings’ social and environmental aspects (^{Villa Castaño et al., 2018}). Engagement occurs at emotional, cognitive, and behavioral levels; greater commitment is associated with more desirable behavioral change (^{Piligrimiene et al., 2020}).

From a more orthodox perspective, the sustainable consumer is located within the solidarity economy, respectful of the environment and local culture, which promotes situations of fair trade, cooperation, community work, and sustainable consumption, seeking forms of production that do not harm people or animals (^{Gadotti, 2016}). This current involves collective action for constructing an alternative economic system governed by ethical principles for the development of life (^{Coraggio, 2016}). Likewise, an almost utopian alternative is the theory of Good Living, where the consumer is open to new forms of relations of life in the community, where individualism has no place, consumption is determined solely by need, and the recovery of cultural identity and age-old heritage is privileged (^{Schlemer Alcántara, 2017}).

Recently, ^{Salgado Beltrán (2019)} identified two distinct segments declared the existence of two differentiated segments in the sustainable consumption behavior of people: environmentally participatory and environmentally consistent. In the first case, a group is categorized as mainly composed of middle-income young people who, although carrying out environmentally positive activities, are not consistent due to poorly consolidated environmental values and beliefs. Conversely, the environmentally consistent group, which is smaller, is fully aware of how their actions affect the environment and act accordingly. They are economically active and of medium-high socioeconomic level. In recent decades, various studies have prior studies identify internal and external determinants of sustainable consumption behavior (SCB). To this end, it has been shown that behavior can be affected by external factors that transform or limit behaviors derived from a change in people’s values and beliefs. Therefore, it is important to consider that individual factors also include environmental variables and the context in which a person decides (^{Jagers et al., 2016}; ^{Piligrimiene et al., 2020}). To explain SCB, as early as 1974, Kinnear, Taylor, and Ahmed determined two ecological and conscious behavior dimensions: attitudes and buying behaviors (^{Yarimoglu & Binboga, 2018}).

In terms of what this study aims to demonstrate, the authors present different approaches to explain sustainable consumption behavior derived from factors external to the individual, in this case, social factors: the influence of the social environment was found in the literature as a trigger for more responsible consumption behaviors (^{Villa Castaño et al., 2018}; ^{Wang et al., 2019}; ^{Lubowiecki-Vikuk et al., 2021}). Likewise, social pressure is a factor that forces consumers to act according to politically correct premises regarding their consumption decisions (^{Heeren et al., 2016}; ^{Yarimoglu & Binboga, 2018}). On the other hand, it has also been claimed that government policies on sustainability promotion positively impact consumers (^{Heeren et al., 2016}; ^{Wang et al., 2019}; ^{Garcia-Machado et al., 2020}; ^{Piligrimiene et al., 2020}; ^{Lubowiecki-Vikuk et al., 2021}). Likewise, market conditions such as price, for example, turn out to act as motivating or limiting factors for consumers of green, organic, fair trade, or sustainable products, so they are considered to be determinants of this behavior (^{Acuña-Moraga & Severino-Gonzalez, 2018}; ^{Villa Castaño et al., 2018}; ^{Yarimoglu & Binboga, 2018}; ^{Miranda-de la Lama et al. 2019}; ^{Salgado Beltrán, 2019}; ^{Piligrimiene et al., 2020}; ^{Lubowiecki-Vikuk, et al., 2021}).

In addition, it has been concluded in several studies that education is a generator of habits and interests in favor of sustainability (^{Salgado Beltrán, 2019}; ^{Wang et al., 2019}; ^{García-Ruiz et al., 2022}). In the case of demographic variables such as age, gender, and educational level, although they do not always manifest the same behavior as this depends on the context studied, it has been shown that they do influence people’s sustainable behaviors (^{Tabernero et al., 2015}; ^{Acuña-Moraga & Severino-González, 2018}; ^{Acuña-Moraga et al., 2022}).

With this theoretical foundation and to identify the social factors that determine the sustainable behavior of consumers in Mexico, this paper, which is the last in a series of three investigations conducted in Poland (^{García-Machado et al., 2020}) and Spain (^{Figueroa-García et al., 2018}), aims to define how these factors contribute to and promote the sustainable consumption behavior of their populations in different economic, social and cultural contexts, which will be analyzed in a fourth study that will bring together the results.

Methodology

We conducted an explanatory quantitative study in the Querétaro metropolitan area, formed by the municipalities of Querétaro, Corregidora, and El Marqués. The instrument was a 26-item questionnaire operationalizing the endogenous construct sustainable consumption behavior (SCB) and the exogenous constructs: demographic factors (DEM_VAL), environmental influence (ENV_INF), education and information (E&I), social pressure (SO_PRE), market conditions (MKT_C), and government actions (GOV_A). The study’s hypotheses were built from the literature review carried out on this occasion and the two previous ones (^{García-Machado et al., 2020}; ^{Figueroa-García et al., 2018}), also shown in Figure 1. The definition of the indicators can be found in Table 1.

Source: created by the authors

Figure 1 References of the theoretical model proposal 

For the application of the instrument, simple random probability sampling was used to determine a sample of 244 respondents. The sample distribution was 66% female and 34% male; the age range of the respondents was 18 to 69 years, with a mean of 37 years and a mode of 19. 25% had a high school or technical education, 43% had a university degree, and 30% had a postgraduate degree. Two percent reported another level of education. The survey was carried out by 6 interviewers who conducted the questionnaire in public squares, universities, parks, and homes.

Theoretical model estimation

As mentioned above, the starting point is a theoretical model based on the existing literature, with references in Figure 1.

Based on the authors referred to in Figure 1, who explain sustainable consumption behavior based on social factors, a theoretical model is proposed that presents the exogenous latent variables mentioned as causal constructs of the endogenous construct of sustainable consumption behavior (SCB), forming the PLS pathway model explained in Figure 2.

Source: created by the authors with SmartPLS

Figure 2 Initial theoretical model 

In this context, Table 2 shows the hypotheses on the possible social determinants of a person’s sustainable consumption behavior, according to the model in Figure 2.

Table 2 Initial hypotheses 

H1: Demographic variables affect SCB. H1: DEM_VAL → SCB.

H2: Education and information on sustainability are decisive for SCB. H2: E&I → SCB

H3: The influence of the social environment is decisive in the development of SCB. H3: ENV_INF → SCB

H4: Market conditions have a direct relation with SCB. H4: MKT_C → SCB

H5: Social pressure has a positive and relevant relation with SCB. H5: SO_PRE → SCB

H6: Government actions are determinant in the development of SCB. H6: GOV_A → SCB

Source: created by the authors

Results

For the presentation of the results, once the PLS path model is determined, the evaluation of the measurement models begins. Figure 3 shows the loadings and external weights of the indicators for the measurement models in Mode A and Mode B, the path or regression coefficients for the relationships of the structural model, and the R² values for the endogenous latent variable. Preliminary estimates suggest that ENV_INF (0.410) has the strongest effect on SCB, followed by E&I (0.216) and MKT_C (0.210). On the other hand, 52% of the variance of the dependent variable is explained by the 6 exogenous constructs. It will be necessary to determine whether the connection between the constructs is statistically significant, for which it is important to consider the path coefficient’s size. Notwithstanding these first observations, evaluating the measurement and structural models is necessary to make definitive statements about the coefficients’ meanings and the overall model’s goodness of fit.

Source: created by the authors using SmartPLS (^{Ringle et al., 2015})

Figure 3 PLS-SEM first results 

Evaluation of A-mode measurement models

The evaluation of reflective measurement models is performed by testing the reliability and validity of each construct, which is necessary to confirm its inclusion in the model (^{Hair et al., 2019}). The indicators observed in this assessment are indicator reliability, composite reliability, convergent validity (average variance extracted, AVE), and discriminant validity. For a construct indicator to be reliable, external loadings above 0.708 are sought; nevertheless, keeping indicators with loadings below this figure is subject to consideration since if their permanence implies an increase in composite reliability and AVE remains above the suggested threshold value, it is advisable to keep them (^{Hair et al., 2019}). To improve the initial pathway model shown in Figure 3, and after several tests and adjustments, 7 of the 26 indicators originally proposed were eliminated since they did not reach an acceptable level of reliability (Figure 4). Therefore, from the initial hypotheses, H₃:GOV_ASCB was eliminated since its only two indicators had undesirable composite reliability values, constituting an invalid measure of the construct as they are redundant (^{Hair et al., 2019}). Likewise, new links were created between the constructs, generating new hypotheses that will be tested: H₂:E&I→ENV_INF, H₃:E&I→MKT_C, H₆:MKT_C→ENV_INF, and H₈:SO_PRE→MKT_C

Source: created by the authors using SmartPLS (^{Ringle et al., 2015})

Figure 4 PLS-SEM Enhanced model 

Regarding the reliability and validity results obtained through the evaluation of the reflective measurement models, Table 3 shows that almost all the external loadings of the reflective constructs are above the threshold value of 0.708. On the other hand, the reliability indicators reveal what part of the variance of an indicator is explained by the construct and considering that, as a general rule, a latent variable should explain at least 50% of the variance of each indicator (^{Hair et al., 2021}), it can be said that, except for ENVI_1 and SCB_3 whose reliability value is below this threshold, all the results of the analysis meet this criterion. In order to determine convergent validity at the construct level, the average variance extracted (AVE) was used, equivalent to a construct’s commonality (^{Hair et al., 2021}). The requested threshold value is 0.50; in the proposed model, the results are above that value in all constructs, i.e., more than half of the variance of their indicators is explained: DEM_VAL (0.757), ENV_INF (0.527), SO_PRES (0.789) and SCB (0.540), which means that the measures for the four constructs have high levels of convergent validity.

Table 3 Summary of results for reflective measurement models 

	Convergent validity				Internal consistency
	Indicators	Loads	Reliability indicator	AVE	Composite reliability	Cronbach’s alpha
DEM_VAL	DEMVAL_1 DEMVAL_3	0.879 0.861	0.773 0.741	0.757	0.862	0.679
ENV_INF	ENVI_1	0.706	0.498	0.527	0.817	0.702
	ENVI_2	0.746	0.557
	ENVI_3	0.713	0.508
	ENVI_4	0.740	0.548
SO_PRES	SOPRES_1	0.835	0.697	0.789	0.882	0.745
	SOPRES_2	0.939	0.901
SCB	SCB_1	0.733	0.537	0.540	0.823	0.713
	SCB_3	0.612	0.375
	SCB_4	0.778	0.605
	SCB_5	0.801	0.642

Source: created by the authors

In addition to using Cronbach’s alpha, composite reliability was used to evaluate internal consistency, which assumes that the indicators have different weights (^{Chin, 1998}). For reflective constructs, Cronbach’s alpha was ≥ 0.70 in almost all cases, which is the recommended value (^{Cronbach, 1951}). Reliability values between 0.70 and 0.90 can be qualified as satisfactory (Nunally & Bernstein, 1994, cited in ^{Hair et al., 2014}; ^{Frías-Navarro, 2022}). As seen in Table 3, the reliability values of all reflective measurement model constructs show values above the suggested lower limit. This means all indicators for the three latent variables are above the minimum level required for external loads and composite reliability. Regarding the variable DEM_VAL that shows a Cronbach’s alpha coefficient below the threshold value, it was decided to keep it in the model since it has been shown that Cronbach’s alpha underestimates the magnitude of reliability when there are items with less than five indicators as is the case of this variable (Zumbo et al., 2007, cited in ^{Contreras & Novoa-Muñoz, 2018}). It is not always effective when used exclusively (^{Cervantes, 2005}), so, supported by the value obtained by the composite reliability indicator, it was decided to retain it (^{Contreras & Novoa-Muñoz, 2018}).

In order to confirm the validity of the measures of a construct, the criteria of convergent validity and discriminant validity are used. In the first case, a high correlation is sought between measures of the same construct, while for the second, this correlation must be higher than that existing between measures of a different construct (Campbell & Fiske, 1959, cited in ^{Martínez-García & Martínez-Caro, 2009}). In the evaluation of the measurement models in Mode A, the discriminant validity will be reviewed, for which the Heterotrait-Monotrait correlations ratio (HTMT) is used, whose threshold value, according to ^{Henseler et al. (2015)}, is 0.90. As can be seen in Table 4, the discriminant validity criterion is satisfied in all the constructs, being presented below the threshold value.

Table 4 Discriminant Validity (HTMT Criterion) 

	DEM_VAL	ENV_INF	SCB	SO_PRE
DEM_VAL
ENV_INF	0.210
SCB	0.358	0.884
SO_PRE	0.131	0.456	0.220

Source: created by the authors

Discussion

The study concluded that the demographic variables education and information, social pressure, social environment influence, and market conditions explain 50.8% of the endogenous latent variable SCB variance. Likewise, new hypotheses were accepted that establish other relationships: on the one hand, 43.2% of the variance of the ENV_INF variable is explained by the E&I and MKT_C constructs. On the other hand, the E&I and SO_PRE variables explain 36% of the variance of the MKT_C variable. According to the results reviewed, all the hypotheses tested were accepted with a confidence level of 99%. The model demonstrates that the variables posed in the corrected model are positively and significantly related to people’s sustainable consumption behavior. In the case of education and information, it was possible to affirm that this social factor does determine SCB (hypotheses E&I→ENV_INF; E&I→MKT_C and E&I→SCB). Although it has been concluded that, by itself, educating on sustainability issues does not necessarily imply an increase in pro-environmental behaviors, it has also been stated that the lack of knowledge on the subject is one of the main causes of unsustainable behaviors since people do not know how such behaviors affect society (^{Heeren et al., 2016}).

Moreover, it is considered an opportunity for environmental education to be the means to raise people’s awareness, especially of the effects of their consumption on environmental deterioration (^{Salgado Beltrán, 2019}). On the other hand, as a measure to respond to the planet’s environmental problems, the generation of knowledge must be guaranteed, integrating the issue of sustainability into educational and research programs (^{Geng et al., 2013}). Consumers can modify their consumption behaviors if the awareness of sustainability is developed in them through educational programs (^{Wang et al., 2019}).

It is also confirmed that the influence of the social environment is a determinant of SCB (ENV_INF(SCB hypothesis). On the subject, ^{Villa Castaño et al. (2018)} also presented evidence of the influence of socioeconomic and cultural environments on responsible consumer behavior. Likewise, the authors report that sustainable consumption is possible, from which cultural values, which focus on satisfying basic needs, are transformed into values that seek the common welfare. Similarly, it is recognized that the political factor can influence the degree of responsibility of a consumer (^{Lubowiecki-Vikuk, et al., 2021}). On the other hand, there are negative environmental factors, such as corruption, inequality, and poverty, which limit sustainable consumption and, in this context, economics, politics, and cultural diversity are also elements that intervene in the definition of sustainable consumption patterns in countries (^{Wang et al., 2019}). Undoubtedly, it can be stated that the social environment strongly influences consumer awareness regarding their responsibility to create a more sustainable society.

Likewise, it is demonstrated that market conditions greatly influence consumers to be more responsible and aware, recognizing that there is a direct and meaningful cause-and-effect relation (hypotheses: MKT_C→ENV_INF and MKT_C→SCB). Thus, the strategies used by marketing to educate and inform consumers about responsible consumption are essential to achieve the purchase of products that correspond to their values. Likewise, evidence has been found that such consumers encounter limitations deriving from poorly made marketing decisions, such as high prices, incorrect labeling, and certification not known or not supported by the government (^{Miranda-de la Lama et al., 2019}). Given this reality, markets are challenged to create less harmful products and services, which are communicated to consumers so that they change their consumption habits in favor of a sustainable lifestyle (^{Acuña-Moraga & Severino-González, 2018}). It has also been recognized that marketing factors reinforce sustainable lifestyles (^{Lubowiecki-Vikuk et al., 2021}).

In another aspect, social pressure is a predictor of sustainable consumption behavior. Through social norms, this variable has been recognized as a determinant of sustainable behavior in several areas (consumption behavior, waste management, and environmental risks) (^{Heeren et al., 2016}). Also, social pressure, from the perception of subjective norms, is part of the indicators used to measure behavioral intention (^{Yarimoglu & Binboga, 2018}). Given this situation, social pressure was expected to influence sustainable behavior positively (SO_PRE→SCB hypothesis); nevertheless, this model produced a negative value, considered an important finding. As social beings, individuals respond to contextual norms, so they tend to act consistently with the norms of the groups to which they belong, so the pressure can be positive or negative, as the case may be (^{Tabernero et al., 2015}). In this regard, ^{Alonso and Ildefonso (2012)} state that if the consumer does not belong to a group with values and symbols compatible with their own, the pressure this group exerts on them may be contrary or non-existent.

Regarding demographic variables (DEM_VAL→SCB hypothesis), it has been shown that gender or age plays a role in responsible consumer behavior, causing meaningful differences in the measured dimensions in certain contexts (^{Acuña-Moraga & Severino-González, 2018}). Nevertheless, some contradictions indicate that these variables are unrelated to such behavior (^{Tabernero et al., 2015}). The present study concluded that age and educational level have a positive and significant relation with SCB, but gender does not exhibit a significant association with SCB in this sample.

On the other hand, it has been recommended that governments, at all levels, should be involved in promoting responsible behaviors in consumers based on information through incentives for purchasing sustainable products (^{Lubowiecki-Vikuk et al., 2021}). Likewise, it has been concluded that, both in developed and developing economies, governments, through their policies, strategies, and programs, should actively participate in achieving sustainable consumption and production systems (^{Wang et al., 2019}). Notwithstanding the above, this study could not prove that government actions determine sustainable consumption behavior in Mexico since there was redundancy in its indicators, and the construct had to be eliminated. Nevertheless, studies in other countries, such as Poland, conclude that government actions explain sustainable consumption behavior (^{García-Machado et al., 2020}).

Conclusions

This research was based on a theoretical model that proposed that 6 social factors are determinants of people’s sustainable consumption behavior. This research concludes that these factors are crucial in determining sustainable consumption behavior. The influence of these social factors on how people purchase and use products and resources cannot be underestimated. As society continues toward greater environmental and social awareness, it is essential to recognize and encourage these factors as key drivers to promote more responsible and sustainable consumption. Collaboration between individuals, communities, and governments plays a key role in creating an enabling environment for more ethical and equitable consumption behavior.

On the other hand, sustainable consumption behavior is an increasingly important trend in today’s society. As environmental and social concerns continue to grow, people are taking a more conscious and responsible approach toward what they buy and how they use it. This approach benefits the environment by reducing waste and ecological footprint and promotes a more balanced and ethical lifestyle. The widespread adoption of sustainable consumption habits is essential to address global challenges and ensure a healthier and more sustainable future for future generations. The model resulting from this research enables the motivators of sustainable consumption behavior to be identified, which provides elements to encourage this behavior in people through the formulation of programs, strategies, guidelines, and any other relevant action for this purpose. This way, governments, companies, educational institutions, associations, and citizens can become managers and actors in society’s transition toward sustainability and the common good.

Limitations

The limitations of this study are mainly related to the geographic area analyzed since, by focusing this research on the Querétaro metropolitan area, the results and conclusions may be limited to similar contexts. Likewise, as a future line of research, stratified sampling is suggested to compare results based on differentiated demographic criteria through a multigroup analysis.