Socially Responsible Consumption: An Application in Colombia

This study constructs a measurement scale for Socially Responsible Consumption in the particular context of Colombia. It uses a mixed qualitative and quantitative methodological approach, beginning with four focus groups and ending with a quantitative validation exercise employing Exploratory Factor Analysis. The result is a Socially Responsible Consumption measurement scale consisting of four dimensions that reflect paradigms found in the existing literature. These are, however, expressed differently in Colombia. In particular, Socially Responsible Consumption involves consumer behavior that favors corporate social responsibility practices and the rational consumption of resources and products while recognizing the importance of maintaining a healthy lifestyle through healthcare.


Introduction
Our planet is facing a global state of emergency in which environmental capacity and the recovery thresholds of ecosystems, societies and weak institutions are being pushed to their limits. We are entering the sixth mass extinction (Ceballos et al. 2015), and this is only one of many problems humanity is facing, such as climate change, overpopulation, deforestation, and the lack of political commitment and collaboration (Ehrlich & Ehrlich 2013). The consumption of goods and services lies at the heart of these problems, and in some regions of the planet, average consumption is simply beyond the earth's capacity to sustain it (Behrens et al. 2007;Mancini et al. 2016).
The rise of Socially Responsible Consumption includes a complex reappraisal of consumer demands, characterized by a heightened awareness of the impacts our behavior has on society and on the equilibria of ecosystems ( € Ozça glar-Toulouse 2005; Francois-Lecompte 2010; Perdomo-Ortiz & Escobar 2011; Dueñas Ocampo et al. 2014). Socially Responsible Consumption introduces social and environmental responsibility into the criteria used when making purchasing and consumption decisions, extending to the evaluation of every step of the value chain. 1 Responsible consumption means 'doing more and better with less, increasing net welfare gains from economic activities by reducing resource use, degradation and pollution along the whole lifecycle, while increasing quality of life' (United Nations 2015). Understanding specific features of Socially Responsible Consumption worldwide and in relation to particular contexts is a relevant research stream in need of further exploration.
This study examines a construct for measuring Socially Responsible Consumption in a specific country (Colombia), contributing to the understanding of how this type of consumption is contingent upon the specific context in which it is being measured. Our research question derives from a gradual change in consumer values, currently guided by environmental concerns and social awareness (Inglehart 2015). This has led to the emergence of evolving Socially Responsible Consumption behavior among consumers that is contingent on their cultural, socioeconomic and political environment (Maignan 2001;Francois-Lecompte & Roberts 2006). From a scholarly perspective, this behavior has been studied using different types of metrics. Hence, the validation of Socially Responsible Consumption must be tested in multiple contexts, as noted in numerous studies (Francois-Lecompte & Roberts 2006;Singh 2009;Lee & Shin 2010;Yan & She 2011;Wesley et al. 2012;P erez-Barea et al. 2015). Joining this line of research, this paper will contribute to the construction of a Socially Responsible Consumption measurement tool for use within one such specific context. This study aims to answer the following question: What is the measurement scale of the construct known as Socially Responsible Consumption for measuring Socially Responsible Consumption in Colombia?
This study develops an appropriate metric using a mixture of qualitative and quantitative approaches. The authors assembled four focus groups for the qualitative inquiry. Next, a set of items was drafted using focus group data and information provided through background research. This data set was subsequently analyzed using quantitative methods. Thus, we sought evidence of content validity by consulting expert judges, of construct validity through Exploratory Factor Analysis (EFA), and of criterion validity through regression models using Socially Responsible Consumption tool scores as a dependent variable.
This article is divided into six sections. The first is this introduction, followed by a review of the conceptual framework of Socially Responsible Consumption. The third part details our methodology. The fourth and fifth sections present our results in detail before discussing them. The article ends with the study's conclusions, its limitations, and suggestions for further research.

Conceptual framework
The Socially Responsible Consumption concept Our research on Socially Responsible Consumption investigates the methods by which consumers make use of their rationality in making purchasing and consumption decisions (Moosmayer 2012). Additionally, this study analyzes the socially responsible consumer considering different disciplines and assuming a multidisciplinary perspective (Peattie 2010).
Building on the work of Berkowitz & Lutterman (1968) and Anderson & Cunningham (1972), the first definition of Socially Responsible Consumption that appears in the literature comes from Webster (1975: 188), who states that Socially Responsible Consumption describes 'a consumer who takes into account the public consequences of his or her private consumption or who attempts to use his or her purchasing power to bring about social change'. In this definition, the emphasis is not merely upon the simple act of purchasing goods and services. The new consumer understands that their purchasing decisions exercise power to the extent that these actions can transform the consumer's environment. Thus, we can affirm that this new rationality, in addition to being moral, is also political: thoughtful purchasing can drive social change.
Webster's definition informed subsequent research on Socially Responsible Consumption (Brooker 1976), and following a review of the literature, we can see that the concept evolved through Roberts' work (1995). Roberts states that Socially Responsible Consumption is practiced by the consumer who 'purchases products and services which he or she perceives to have a positive (or less negative) impact on the environment or uses his/her purchasing power to express current social concerns'. Roberts (1995) elaborates on previous research by Antil (1984). This emerging conception of Socially Responsible Consumption includes a new element of environmental responsibility through consumerism. As Francois-Lecompte and Roberts (2006: 52) suggest, Socially Responsible Consumption includes purchasing products and services which are perceived to have a positive or less negative impact on the physical environment and/or the use of purchasing power to express social concerns. This definition is consistent with the concept of corporate social responsibility: the socially responsible consumer is a person who thinks about the well-being of stakeholders that may be affected by his or her purchasing.
This definition shows that environmental and social factors play a key role for the consumer. Additionally, Corporate Social Responsibility (CSR) is an important component of Socially Responsible Consumption (Pivato et al. 2008). Therefore, this approach to purchasing is not only oriented towards the effects that producing goods and services has on the environment, but also towards the responsibilities assumed by the producer regarding labor, environmental protection and transparency. This perspective assumes that a socially responsible consumer has adequate information about a firm's actions. Mohr et al. (2001) previously addressed this understanding of Socially Responsible Consumption, and their work supports the conceptual development of Socially Responsible Consumption offered by Webb et al. (2008), in which it is described as 'a person basing his or her acquisition, usage, and disposition of products on a desire to minimize or eliminate any harmful effects and maximize the long-run beneficial impact on society'. They explicitly include CSR as an aspect of Socially Responsible Consumption in their measurement scale.
Meanwhile, Yan & She (2011) note that socially responsible consumers 'include the recognition of long-term interests for society and the country as a whole in each stage of consumption' (p. 260). In the practice of Socially Responsible Consumption, one can observe behaviors in every step of a consumer's consumption process influenced by their concerns over social, moral and environmental issues. Thus we define SRCB (Socially Responsible Consumer Behavior) as taking the long-term interests of society and the nation as a whole into consideration at every stage of consumption (p. 260).
One unique aspect of the definition given by Yan & She (2011) was that it expanded on a concept previously formulated by Francois-Lecompte & Roberts (2006) regarding geographic limits to Socially Responsible Consumption (at national borders, for instance) and consumer loyalty towards national brand consumption. This means that socially responsible consumers act on their principles and national values and therefore show brand loyalty to national products and businesses as a method of benefiting local producers rather than foreign-based ones (Maignan 2001).
The second column in Table 1 shows that the concept of Socially Responsible Consumption has evolved in response to an individualistic rationality in which the act of consumption is intended to affect the environment as little as possible. The concept of Socially Responsible Consumption widens that sphere of rationality, placing a special emphasis on good business practices and the protection of domestic industry. Thus, the evolution of the concept reveals an informed consumer who recognizes not only the impacts that consumption may have on the individual, but also those social and environmental impacts generated by corporate activity.

Domain of measurement scales
The concept of Socially Responsible Consumption is dynamic and complex. Since the 1970s, different metrics have been developed to facilitate a deeper discussion of the research on the concept and its practical application in the business world. This study, therefore, identifies critical aspects of Socially Responsible Consumption measurement scales, which are set forth in Table 1.
Additionally, a subjective reading was made of the contents of the measurement scales referred to in Table 1, considering the item-level measures. Table 2 shows the detailed exercise, and certain common characteristics are identified in the domain of measurement scales. Emerging issues in the measurement of Socially Responsible Consumption are identified. These are as follows: 1. Environment: From Webster (1975) to Yan & She (2011), environmental concerns are decisive for the socially responsible consumer when purchasing or consuming goods and services. 2 shows that protection of the environment is related to care for natural resources, recycling, and the effects of pollution on quality of life. SRCB is related to an ecological awareness that seeks not only to preserve natural resources but also to avoid using products that generate negative impacts on the environment in their development or usage. 2. CSR: Socially Responsible Consumption measurement scales from Antil (1984) to Yan & She (2011) reveal the relationship between the behavior of a socially responsible consumer and responsible practices at the corporate level. In other words, when making purchasing decisions, the socially responsible consumer considers the enterprise's duties regarding (i) philanthropy, (ii) respect for labor rights, (iii) using misleading advertising, and (iv) protecting the environment. An enterprise's observation of responsible practices becomes a key factor for the socially responsible consumer, and this due diligence is incorporated within the rationale for purchasing (Tsai et al. 2015 In sum, concern for the environment and CSR are latent themes in the metrics employed in Socially Responsible Consumption measurement scales for the different countries that have been studied. Nevertheless, the literature also shows that Socially Responsible Consumption varies in accordance with the values and context of each country (Maignan, 2001;Francois-Lecompte & Roberts, 2006;Yan & She 2011). Some themes emerge that are contingent upon the measurement context.

The contingency of Socially Responsible Consumption
It is important to emphasize the contingent nature of Socially Responsible Consumption and to clarify its meaning. As we have seen above, the bulk of the literature regarding the concept of Socially Responsible Consumption and its measurement has focused on the universality of the concept itself (  A growing body of research reinforces the idea of Socially Responsible Consumption as being contingent in nature, and it is necessary to develop this idea, especially in contexts with differentiated value structures. The work of P erez- Barea et al. (2015) has demonstrated the relevance of this gap in the literature. These researchers use the measurement scale developed by Francois-Lecompte & Roberts (2006) to perform a validation exercise in Spain, showing how the same relevant characteristics of communitarianism and patriotism that differentiated the US from France are equally present in the Spanish context. Moreover, they establish how certain factors related to the acquisition of products and services (e.g. consumption volume or firm behavior) are more important to Spanish people than they are to the French (P erez- Barea et al. 2015: 54).
Similarly, other countries display context-specific trends when Socially Responsible Consumption is measured. For example, Asian countries such as China emphasize nationalistic characteristics and highlight the entrepreneurial spirit of the individual as an engine of national development (Yan & She 2011). In Hong Kong, studies have observed the importance of Socially Responsible Consumption for the development of identity in young people. There, adolescents consume certain goods and services with their social and/or environmental impacts in mind and engage in responsible behavior as a way of satisfying their need to belong to specific groups or to be accepted (Lee & Shin 2010).
In the particular context of Latin America, there are features that anticipate differentiated and contingent consumption behavior. It is important to stress that Latin American countries display a homogeneous set of values, albeit with a complex and diverse array of responses to CSR strategies (Trapero et al. 2010;Carvalho et al. 2010;Marquina 2010;Feldman & Reficco 2015). Maignan (2001) advanced the Socially Responsible Consumption literature by showing that contextspecific cultural differences in consumption can appear in various ways in Latin America's diverse settings. The World Values Survey (2010)(2011)(2012)(2013)(2014) shows that Latin America forms a homogeneous group of countries sharing a tradition rooted in the values of self-expression and increased deliberative citizen participation, which stem from traditional beliefs. These traditional values are in the process of transforming into secular-rational values, resulting in a higher degree of demand for environmental protections and increasing calls for popular participation in economic and political decision-making (Institute for Future Studies 2015). 2 These trends could explain the differences and similarities in particular forms of consumption arising in specific contexts. As such, these countries offer suitable cases for the comparison and further validation of Socially Responsible Consumption metrics. The content of Socially Responsible Consumption in Colombia is still developing. This process involves contingent cultural features particular to Latin American countries, and for this reason, Colombia offers a suitable case study for the contingency debate on contextspecific Socially Responsible Consumption.
In sum, this study assumes the existence of Socially Responsible Consumption that is contingent upon the idiosyncratic values, principles and socioeconomic conditions arising from a particular country's development. Taking this into account, the authors sought evidence to use in building a scale for measuring Socially Responsible Consumption in Colombia.

Methodology
This study employed a mixed methods approach. Our design used qualitative and quantitative data sequentially. The results and all decisions in the quantitative investigation stem from the conclusions reached in the qualitative analysis (Tashakkori & Teddlie 2009;Feldman & Reficco 2015). Our qualitative research sought to establish the limits of the Socially Responsible Consumption construct, while the quantitative side allowed us to explore the instrument's validity and reliability. Figure 1 illustrates the methodological process.

Qualitative inquiry
The study involved four focus groups with different cohorts. This technique is commonly used in the development of measurement instruments in business ethics scholarship (Campbell & Cowton 2015;Cowton & Downs 2015). Theoretical sampling was used, as it allows for including cases that are relevant to understanding the studied phenomena (Denzin & Lincoln 2011).
The literature shows that Socially Responsible Consumption is more common in populations with higher levels of education, as these have more information and better judgment criteria for evaluating that information (Berkowitz & Lutterman 1968;Mitchell 1983;Roberts 1996;Li et al. 2012;Lee 2014;Pedrini & Ferri 2014). Accordingly, the theoretical criteria identified in the literature were used to form four focus groups, following the sizes recommended by Krueger & Casey (2000): (i) a group of five undergraduates aged 18-23 (two women and three men), (ii) a group of seven young professionals (four women and three men) aged 25 to 30, (iii) a group of five post graduate students (two women and three men) aged 31 to 34 and (iv) a group of seven university professors (two women and five men) aged 35 to 40. We performed a focus group session for each cohort. The students and professors were drawn from the Faculty of Business Administration at a private university in Bogota.
Each group session began by identifying the discussion subject as Socially Responsible Consumption. This was followed by an individual written exercise on (i) what they understood by Socially Responsible Consumption, (ii) key elements or characteristics of Socially Responsible Consumption, (iii) decision-making criteria used at the moment of purchasing mass-produced items in a supermarket, and (iv) a discussion about the hierarchy of purchasing criteria employed. With the information collected from the qualitative exploration and our background research, a proposal was delineated of construct items for the measuring instrument.

Quantitative inquiry
The quantitative inquiry had three elements. First, three experts, two in marketing and one in CSR, judged the pertinence, relevance, semantics and syntax of each item in search of evidence of content validity. Second, we conducted a construct validity exercise, in which the instrument was applied to a sample of 323 post graduate students from Business Administration programmes. Third, validity evidence for the time criterion was sought to test the  instrument's ability to predict ratings in criteria identified in previous research. In this case, the variables used were gender, age and social stratum (Singh 2009;Straughan & Roberts 1999). The cohorts used here are ideal for the construct validity exercise because identically constructed cohorts have been used in the development of previous Socially Responsible Consumption metrics (Straughan & Roberts 1999;Webb et al. 2008;Yan & She 2011). Similarly, in the qualitative inquiry, the group of post graduate students showed that they possessed information relevant to the topic, and they reported using criteria associated with Socially Responsible Consumption when purchasing. In total, 58.5% of the participants were women. The age distribution was as follows: under 25 (17%), 26-30 (39.6%), 31-35 (20.1%), 36-40 (7.1%) and over 40 (14.2%), with data missing for 2%. The socioeconomic levels are characterized as follows: low (4.3%); middle (81.1%), high (13.3%), with data missing for 1.7%. 3 These data were analyzed using EFA with SPSS 21 software. The possible presence of social desirability bias was explored using the Common Latent Factor (CLF) method.

Qualitative analysis
The four questions that guided the focus groups were used as categories of analysis, with the results shown in Table 3.
Undergraduate students associate Socially Responsible Consumption with the adequate use of a purchase, which means buying environmentally friendly products and calls for the consumer to have a broad understanding of the production cycle. When buying products, they emphasize the importance of reducing environmental impact and spending only what is necessary. However, their purchasing criteria focus on price, quality and functionality, relegating Socially Responsible Consumption criteria to a secondary level.
Young professionals recognize certain Socially Responsible Consumption elements, such as the importance of the proper use of products and an awareness of the impact of consumption on society. They indicate that it is important to rationalize consumption and to avoid harmful environmental impacts. For this group, consumption is informed by a single question 'Can I afford it?' Both undergraduates and young professionals claim that cost reduction is not an element of Socially Responsible Consumption. Although consumer rationality is underpinned by price, Socially Responsible Consumption implies that the purchaser should consider the impact that products have on other agents. It is not merely a matter of price; it is also a matter of awareness in the act of purchasing.
Post graduate students show a clear understanding of the concept of Socially Responsible Consumption. For them, Socially Responsible Consumption means conscious consumption that takes into account economic, social and environmental impact. They consider 'Socially Responsible Consumption' to refer to a broad concept, rather than to the mere act of purchasing. Socially Responsible Consumption considers the limited resources available for a product's development, and it shies away from instrumental consumption in the pursuit of 'joy'. Additionally, responsible healthcare is emerging as a feature of Socially Responsible Consumption. Post graduate students include characteristics of Socially Responsible Consumption in both their buying criteria and their criteria hierarchy.
Finally, the group of university professors used technical and academic definitions to approach the concept. They define Socially Responsible Consumption as 'the third link of CSR'; the first link is philanthropy, the second is concern for stakeholders, and the third is a focus on consumer responsibility. Again, this group makes mention of responsible healthcare. However, responses from this group do not reflect the broad conceptual scope revealed by the post graduate students' responses. Further, the professors do not include elements of Socially Responsible Consumption in their purchasing criteria.
This analysis concludes that Socially Responsible Consumption is an emerging phenomenon in Colombia, and it is understood imprecisely. In particular, undergraduates and young professionals reveal conceptual 'fuzziness', but post graduate students define the concept more clearly. This latter group recognizes elements associated with Socially Responsible Consumption and situates them within the hierarchy of consumption criteria. Consequently, the group of post graduate students emerges as the most suitable one to use for quantitative validation.
To summarize, a review of the literature reveals a theoretical framework that enables us to define Socially Responsible Consumption. By comparing that definition with the elements of Socially Responsible Consumption named in the focus groups, conceptual coherence can be found. Socially Responsible Consumption is understood to be a type of consumption that takes into account the public consequences of private consumption, whereby it seeks to minimize damage and maximize long-term benefits to society. This type of consumption encompasses practices regarding environmental care, the regulated consumption of energy, the use of environmentally friendly products, and fair trade. However, an important difference emerges. For the focus group participants, Socially Responsible Consumption includes healthcare, wherein consumption should take into account any impacts on one's own health and that of others. This new element, which emerges in this study's qualitative analysis, can be understood as a broadening of the concept's limits relative to those found in the existing literature.

Quantitative analysis
After a group discussion, the study's team members decided to take the items identified in previous metrics as a starting point. Considering those, in addition to the focus group results, a 45-item Likert scale of seven response categories was proposed, where 1 represents 'I never do it' and 7 represents 'I always do it'. This set of items was used as the point of departure for the clean-up and validation processes.

Content validity
Three experts assessed the initial scale. The agreement among the judges was 0.78, which is a strong or substantial endorsement (Shrout & Fleiss 1979;Hancock & Mueller 2010). The judges indicated that 19 items had issues involving pertinence and relevance, and that three had issues with semantics and syntax. For each of the items presenting problems, the judges provided written reasons for their objections. For example, 'some items are unclear'; 'you are using confusing terms'; 'there was no strict correspondence with the definition of Socially Responsible Consumption'; and 'there were items with repetitive contents that should be eliminated'. Each of these items was considered individually, and 17 of them were eliminated. An EFA was performed with the remaining 28 items.

Construct validity
Before performing the factor analysis, we examined the normality of the data. For our analysis of the normality assumptions, the skewness and kurtosis indexes were evaluated. Based on our normality analysis, the data's internal structure was identified through an EFA. Generalized least squares factor analysis was used as a method of extraction because some of the items did not fulfil the criteria of normality (Hancock & Mueller 2010).
The sample's adequacy indicators guaranteed the continuity of the EFA (Hair et al. 2010). Bartlett's test was significant at the 0.01 level, and the KMO (Kaiser-Meyer-Olkin) measure of sample adequacy was 0.9. The determinant was more than zero, and the analysis converged with ten iterations. For the analysis, an oblique rotation method -promax -was applied due to the existing association between the factors, which arose from the EFA.
To determine the number of factors, a parallel analysis used the SPSS syntax developed by O' Connor (2000). Parallel analysis is the most efficient and precise technique for determining the number of factors that should be withheld in an EFA (Timmerman & Lorenzo-Seva 2011). As the criterion for item retention, a minimum-weight factor of 0.3 was considered, as recommended by Hair et al. (2010), in response to the sample's reference size.
Four factors or dimensions of the construct were extracted from the EFA, explaining 57.2% of the variance. The process for the elimination of items was systematic. Six items were eliminated because they recorded a factorial weight below 0.3, and one item was eliminated because of a cross-loading greater than 0.35. At the same time, it was verified that no dimension retained fewer than three items, as recommended by McDonald (1985). The final set held 21 items: eight items in factor 1, five in factor 2, five in factor 3, and three in factor 4. The instrument's reliability was evaluated through the omega coefficient calculation, considering that the scale used is an ordered categorical response type (McDonald 1999). The first three dimensions record omega values greater than 0.7, registering adequate levels of reliability within the coefficient's standards. The dimension of Healthcare has an omega value of 0.65, which is on the lower limit of reliability (Gadermann et al. 2012;McDonald 1999). The corrected item-total correlation was evaluated at the same time as the reliability exercise, verifying each item's association with its corresponding dimension. All of the results from evaluating the Socially Responsible Consumption measuring model are presented in Table 4.
The factorial structure of the Socially Responsible Consumption construct supports evidence for four factors or dimensions. The dimensions are labeled External-CSR, Internal-CSR, Consumption Rationalization and Healthcare.
After evaluating the metrics model, a new reliability exercise (apparent reliability) was conducted using the content (Borsboom et al. 2004;Newton & Shaw 2013). This exercise sought to corroborate the assignment of items to each dimension and to eliminate possible sample bias. Three post graduate students from the study sample were identified and given randomly ordered items. Additionally, a conceptual definition of each item in the dimensions was provided in the EFA, which was supported by the literature review. The post graduate students' task was to classify each of the items, taking into account each dimension's conceptual definition.
At the same time, an inter-judge agreement on content reliability was calculated using the kappa index, and a value of 0.87 was recorded. This means there was agreement in accordance with the guidelines set forth by Shrout & Fleiss (1979) and Hancock & Mueller (2010). Four items fit into different dimensions; nevertheless, after reviewing their content, contrasting it with the dimension's conceptual definition, and observing each item's factorial weight, it was deemed justified to maintain the factorial structure. A dimensional relationship arising in the analysis is expected when using an oblique rotation. This means that there will be shared information within the items, making the dimensions nonexclusive.
Yan & She (2011) rekindled the debate on social desirability when they measured constructs based on impacts on morally correct behavior. In this instrument, there is the potential for problems stemming from social desirability. Podsakoff et al. (2003) noted that this aspect of the instrument could be a source of common method variance and performed a CLF test to control for this. According to Podsakoff et al. (2003: 891), 'items [in the CLF test] are allowed to load on their theoretical constructs, as well as on a latent common methods variance factor, and the significance of the structural parameters is examined both with and without the latent common methods variance factor in the model'. The adjustment indexes were calculated as v 2 /df, as well as CFI and RMSEA values, thereby indicating that the common factor variance does not significantly improve when adjusting the model (see Table 5) (Harrington 2009;Hancock & Mueller 2010).
Accordingly, there are no major indicators of a problem with common factor variance in the measurement of Socially Responsible Consumption that could affect the instrument's factorial structure. Therefore, common factor variance, considering possible external factors such as social desirability or any other factor that could distort the analysis, was reasonably controlled.

Criterion validity
Finally, the researchers conducted a criterion validity exercise examining the demographic variables used in the process and recognized in the literature. A regression model was implemented.
An individual is more likely to be a responsible consumer as they grow older; additionally, women are recorded as displaying higher levels of responsible consumption (for the analysis, the gender variables were labeled as male 5 0 and female 5 1). No strong evidence was found for the relationship between Socially Responsible Consumption and socioeconomic level. As previously mentioned, socioeconomic strata were considered as proxy variables for income levels. These strata were labeled as follows: low income 5 0, mid-level income 5 1, and high income 5 2 (see Table 6).
It is interesting to consider the relationship between each dimension of Socially Responsible Consumption   Table 7). Thus, age is a relevant variable in the four dimensions of Socially Responsible Consumption. An ANOVA test was used to evaluate each dimension of the instrument as a function of demographic variables. Significant overall differences were found between the group of participants aged 26 to 30 when compared with the over-40 age group. In all dimensions, the 26-30 age group has the lowest average Socially Responsible Consumption, while the over-40 group has the highest. There are statistically significant differences between age groups in the overall Socially Responsible Consumption dimension, and the post-hoc tests only reveal a difference between the 26-30 and over-40 age groups. Systematically, the older the group, the higher the scores in the Socially Responsible Consumption instrument's dimensions. Indeed, the influence of Socially Responsible Consumption is greater in women because they recognize the external evidence of CSR more clearly than men do. In Healthcare, the gender variable is significant in the regression model when introduced together with the other control variables, but it is not significant when comparing the measurements for men and women separately. Finally, income stratum only affects Socially Responsible Consumption in the dimension of Consumption Rationalization, but in a negative direction. Upon comparing the strata, significant differences were found in the Consumption Rationalization dimension. Once again, after completing post-hoc comparisons, the middle stratum recorded a higher figure than the high stratum. Table 7 summarizes the results of the significance tests.

Discussion
This study developed a measurement scale for Socially Responsible Consumption considering that it expresses or emerges in differentiated forms contingent upon the context in which it is measured Finally, and in contrast to previous studies, a new dimension appeared considered to be indicative of responsible consumption (Healthcare). The proposed Socially Responsible Consumption measurement scale shows that, in terms of purchasing behavior, a specific kind of Colombian consumer (post graduate students) recognizes and values the business practices associated with CSR. Accordingly, the emphasis on content, introduced in the late 1990s by Brown & Dacin (1997) and Webb & Mohr (1998)   of the measurement dimensions differs from the expressions in previous studies. In Colombia, it was found that the recent debate in the business world regarding the existence of Internal-CSR and External-CSR (as a function of company stakeholders) is recognized by the consumer in a differentiated form as purchasing criteria (Fuentes-Garc ıa et al. 2008;Kansal & Singh 2012;Jamali et al. 2015;Taghian et al. 2015;Frynas & Yamahaki 2016).
In the External-CSR dimension, the responsible consumer recognizes that the company is accountable for the effects of its activities on the environment, clients, or specific communities of national interest (such as people displaced by violence, those reincorporated into civil society, or local producers). In the Internal-CSR dimension, the responsible consumer recognizes the importance of a company acting responsibly towards its workers. In fact, through their purchasing choices, consumers reward companies that foster good labor practices, pay fair wages, hire disabled workers, and avoid discriminatory practices. Generally, the External-CSR and Internal-CSR dimensions encourage proactive consumer behavior in encouraging good business practices.
By contrast, in the cases of the antecedent scales measuring green consumption behavior, health-conscious behavior orientation or health consumption, the sense of responsibility in consumption is governed by individual values, not by egoistic-altruistic social values (Stern & Dietz 1994;Kim et al. 2012). In other words, the sense of responsibility for health-conscious behavior, green consumption behavior, or health orientation is personal rather than being primarily oriented towards 'others' or toward society as a whole (Vitell 2015).
In Colombia, as in the rest of the world, debates about sustainable health systems are ongoing, and public policy has promoted preventative health as a socially responsible approach. A typical example of this type of policy is the regulation of smoking and its prohibition in indoor public areas. The same logic can be applied to other harmful products, the consumption of which can be regulated by consumers themselves, and which, from the perspective of moral decision-making, can be considered in terms of showing solidarity with 'others'.
Indeed, considering the topics and subtopics occupying the conceptual domain of Socially Responsible Consumption (see Table 2), the measurement results in Colombia reflect a general pattern. First, measurement criteria associated with Socially Responsible Consumption include consumer environmental awareness, including specific items relating to rationalizing consumption through the reduced use of resources and decreasing the consumption of polluting products. Socially Responsible Consumption metrics do not forcefully express responsible consumption designed to protect against environmental pollution. Second, Socially Responsible Consumption metrics in Colombia include consumption criteria that recognize and reward business practices within the framework of CSR. Socially Responsible Consumption measurement criteria are associated with companies engaging in good environmental practices, good labor practices, and respect for human rights. Third, the extent of Socially Responsible Consumption reflects some emerging issues. A specific dimension is not generated for the subtopics of nationalism, fair trade, support for SMEs or consumer rights. However, some items on the scale provide evidence that these elements are emerging in the Colombian context: for example, concerns about supporting local producers through consumption; the political commitments of companies to societies in conflict, such as Colombia; and the recognition of human rights through companies' adoption of fair trade practices.
Finally, the validation of Socially Responsible Consumption criteria considering the demographic variables of age, gender and income provides evidence for the abundant empirical literature on Socially Responsible Consumption and consumer demographics (Roberts 1996). This study found that women are more prone to practice Socially Responsible Consumption, as evidenced in previous studies (e.g. Webster 1975 andTrapero et al. 2011), and that age has a positive relationship with Socially Responsible Consumption. In this case, the previously published evidence for other variables is much less robust than that found for the effect of gender and does not achieve consensus. For example, Webster (1975) finds no significant relationship between Socially Responsible Consumption and age, while Pedrini & Ferri (2014) find it to be significant and positive. No significance was found with the proxy variable for income.
In sum, the exercise of creating an instrument for measuring Socially Responsible Consumption within the context of a developing country fulfilled the aims of this study. It has contributed to the debate on the content of the construct's scope by validating the trend towards recognizing business practices associated with CSR (Mohr et al. 2001;Francois-Lecompte & Roberts 2006;Webb et al. 2008). Additionally, it supports the identification of other social elements that comprise the construct, including the desire to reduce or moderate consumption and the promotion of healthcare that considers the well-being of the 'other'. This exercise considered Socially Responsible Consumption to be contingent on the geographic context within which it is measured. The metrics of Socially Responsible Consumption used here include the dimensions generally found in previous studies; however, they are expressed differently in this study, which focused specifically on the context of Colombia.

Conclusions
This Socially Responsible Consumption metrics exercise has theoretical/methodological and practical implications. This study provides evidence supporting the premise that the scope of the Socially Responsible Consumption construct and its measurement are contingent on the context of the study (Francois-Lecompte & Roberts 2006;Yan & She 2011).
This premise implies that the structure of the metrics model for Socially Responsible Consumption must be expressed differently based on the context. Thus, evidence is provided demonstrating that the Socially Responsible Consumption model depends on the substrate values shared by the study population (Maignan 2001;Inglehart 2015).
From a practical perspective, measuring Socially Responsible Consumption is useful for building consumer profiles, and it could be used to identify differentiating factors that contribute to effective market penetration by companies (Straughan & Roberts 1999;Singh 2009). In the case of Colombia, for example, the evidence showed that marketing strategies should promote the introduction of business practices associated with CSR. Similarly, companies need to consider that consumers have embraced a sense of restraint, and that they consume according to need. Additionally, in the specific context of this study, it is apparent that responsible consumers associate healthcare with an awareness of the 'other'.
This study has various limitations regarding its methodological design. The attitudinal scale, using self-report methods, is an indirect measure of real consumer behavior. The factorial structure of the scale was not affected by external factors such as social desirability. This means that the common variance problem was reasonably controlled. Additionally, there should be stronger empirical evidence for the measurement scales. It is evident that, due to its level of reliability and number of items, the Healthcare dimension requires more empirical evidence. Further, the factorial locations of some of the items on the External-CSR and Consumption Rationalization dimensions are debatable; hence, there is a need for additional empirical verification. It is worth noting that the study sample is not representative of the population, and for this reason, the study of Socially Responsible Consumption in Colombia will require the scale to be validated using other samples in which the Socially Responsible Consumption construct is expected to be present.
Finally, the research possibilities for the future are promising. First, this study has confirmed that the Socially Responsible Consumption construct is contingent on the measurement context and therefore suggests a series of questions or new models for empirical testing. Second, being able to measure Socially Responsible Consumption allows us to approach other research questions that use Socially Responsible Consumption as an explanatory or explained variable. Alternatively, Socially Responsible Consumption may be considered to be an explained variation; therefore, other research pathways should lead towards identifying demographic or psychographic determinants (Roberts 1996;Francois-Lecompte & Roberts 2006). Finally, Socially Responsible Consumption is a complex and dynamic construct that not only needs to be explored through quantitative methodologies but also, in this specific case, requires more qualitative research ( € Ozça glar-Toulouse 2009; Campbell & Cowton 2015). Notes 1. We adopt the concept of 'morality' as defined by Habermas (1998) in The inclusion of the other, in which the author states that each person has a social character and, based on that, understands that what is moral should come before specific ethical commitments. Thus, morality has characteristics of rationality. This means that the capacity to reason about moral decisions tends towards solidarity, which implies the possibility of living together in a fairer society.