Pro-Circular Consumer Profile: An Approach to Their Identification and Characterization Based on the Components of the Value-Belief-Norm Theory

Abstract

Circular economies have focused on managing organizations and changes in production and consumption models that lead to the better use of resources, generating the least waste. These changes toward new circular models will only be possible if consumers become involved through their behavior. In this sense, the first step is to understand who pro-circular consumers are and what characterizes them. Thus, this exploratory study aims to profile pro-circular consumers based on their behaviors and to characterize them based on the components of the value-belief-norm (VBN) theory. Using survey data of a representative sample of 417 participants in the city of Medellín (Colombia), as well as cluster and multiple correspondence analyses, this study identified some pro-circular consumer profiles, mainly characterized by factors like moral norms and perceived consumer effectiveness. Our results suggest that even when consumers without habitual behaviors toward circularity exist, those who engage in them do so because they consider it the right thing to do and because they believe that their pro-circular action is effective for solving environmental problems. On the contrary, green consumption values, beliefs about awareness and responsibility toward the environment, and sociodemographic factors do not seem to be associated with and characterize this type of consumer.

1. Introduction

The main socio-environmental challenges faced by society are the result of human activity; for example, the progressive scarcity of conventional energy; environmental damage stimulated by the indiscriminate use of land, water, forest and fishery resources; and climate tension [1]. Among the various alternatives that arise to face these problems, circular economies are one of the most recent proposals to address some of the environmental sustainability challenges [2]. Circular economies have been defined as economic systems that replace the end of product life, aiming to close the cycle of linear economic models through different recovery types and levels [3,4]. The basic reuse principles used (e.g., reduce, reuse, recycle, and recover) seek to transform materials into useful goods and services through resource efficiency and use products for as long as possible, thus eliminating waste. Consequently, the principal purpose of a circular economy is to return products to the product life cycle in an economically and ecologically reasonable way, improving the efficiency of the use of materials and reducing the consumption of natural resources [5].

Until recently, authors have focused on circular economies by studying circular objectives, methodologies, and models [6,7]. Another group of authors has examined different strategies, criteria, and indicators for circularity [8,9,10,11]. Finally, another group has addressed the barriers to implementing circular economy principles [12]. All these studies have had as a common element the analysis of companies, mainly because circular economy directly intervenes in production, marketing, and supply processes to generate material use. However, while circular economy efforts mostly fall on industry management processes, there is a key player to achieve the success of these efforts and strategies: the consumer. Hence, it is fundamental to analyze and understand the individual consumer as an essential piece to promoting and implementing circular-economy models [3].

Indeed, previous literature has pointed out that a change in the behavior not only of producers but also consumers is required for both actors to apply circular-economy principles [13,14,15]. As the objective is to circulate products at their highest level of value, to reduce the harmful effects of consumption on the environment, customer behavior can become a fundamental part of the system [16], since the participation of consumers and users allows for changes in purchasing patterns and product use [17]. In addition, the success of a circular economy depends, for the most part, on consumer behavior and perceptions of products designed and manufactured according to circularity criteria [18,19].

Despite the emphasis that previous research has placed on linking the consumer to the analysis of circular economy, few studies have paid attention to the topic so far (e.g., [20,21]. This leaves a gap in the literature, but it simultaneously opens the opportunity to propose analyses that involve the individual and their role in the transition toward and development of more circular business models [22].

Different authors have referred to the importance of identifying aspects that allow for improving the circular behavior of individuals [17] as well as the need to understand the challenges of the attributes of consumer behavior to encourage their participation in the adoption of more circular business models [15]. Enabling the demand for circular products requires a deep understanding of the consumer and the processes by which consumption patterns are transformed [22]. Hence, the relevance of carrying out an initial exercise to explore whether there exists a pro-circular consumer based on their behaviors. To this end, this research is based on the concept of sustainable consumption and on the few circular-economy studies that have considered consumer behavior, addressing behaviors with circularity criteria at the three stages of consumption (i.e., acquisition, use, and final disposal). Additionally, this research examines pro-circular behaviors through the lens of three product categories: means of private transportation (car or motorcycle), furniture, and clothing.

Diverse behavioral theories have approached the analysis of the factors that explain behaviors that seek to impact the environment (i.e., pro-environmental behaviors). Among them, the theory of planned behavior (TPB) [23] and the value-belief-norm (VBN) theory [24] have been among the most widely used models [25,26,27,28]. Based on the circular economy approach, TPB has been used to explain intentions, willingness, and some pro-circular behaviors, mainly regarding electronic products and devices (e.g., [29,30,31,32,33,34]). On the contrary, the VBN theory has been empirically studied and verified for the analysis of a variety of pro-environmental behaviors [35]. Nevertheless, examples of this type of research on pro-circular behaviors are limited so far [36,37], without using statistical and conclusive analyses on the significance of moral constructs it addresses (i.e., values, beliefs, and norms) in behaviors of this type.

Following a moral approach, the VBN theory addresses intrinsic motivators associated with the personality, moral norms, and beliefs of the individual about the relationship between human beings and nature [15], all of them leading to a long-term behavioral change, which is highly desirable in the context of a circular economy [38]. Given the importance of the VBN theory in analyzing pro-environmental behaviors, this research uses it as its basis to characterize pro-circular consumers according to the theory’s components. To our knowledge, no study in the field of circular economy has approached the consumer from this perspective. Therefore, this research proposes a first step to fill this gap and contribute to previous literature on the subject, seeking to identify a pro-circular consumer profile based on their behaviors and characterizing them based on the values, beliefs, and pro-environmental norms they claim to have.

2. Conceptual Framework

2.1. Pro-Circular Behaviors and the Stages of Consumption

Pro-circular behaviors can be defined as those consumer behaviors necessary in a circular economy [22]. Hence, these behaviors are related to strategies or opportunities proposed to move from linear business models to others with a circular perspective, which are framed in different Rs (e.g., reuse, recycle, reduce, recover, remanufacture, and redesign) [3]. Like pro-environmental behaviors, pro-circular behaviors are expected to follow the stages of sustainable consumption and are linked both to purchasing decisions and product use and final disposal decisions [18,22]. Through these essential stages (i.e., acquisition, use, and disposal), consumers are considered to be directly involved in the life cycle of products [15]. For example, renting and investing in more durable products and committing to support circular business models through acquisition are essential in the initial stage, whereas opting for repairing, preserving, and reusing functional products are associated with pro-circular decisions at the product-use stage. Finally, resale and return or devolution that contribute to the subsequent recovery of resources through recycling are considered consumer behaviors at the final disposal stage, at the end of the useful life of products [15,22].

This research is based on this proposal to define and address pro-circular behaviors at the three stages of consumption, described in detail below:

First, at the stage of product acquisition, renting refers to a method in which the consumer buys the practical purpose of a good from the owner and not its ownership [3,22]. On the other hand, repair consists of putting back into operation a product that has ceased to serve its original purpose [16,39]. At the stage of product use, reuse consists of putting back into use a discarded product that is still in good condition and fulfills its original function [16,39]. Finally, at the stage of product disposal, return for recycling consists of the consumer relinquishing ownership of a product after use and handing it over, in good condition, to manufacturers, marketers, or other agents, through the available collection mechanisms [22,40]. Reselling is also part of the stage of final disposal since it extends the product’s life [22,39,41]. This behavior refers to having a product on the market, which may or may not need an adaptation to function like new, for a second consumer to purchase it at a price that is usually lower than a new product.

Table 1. Stages of consumption and pro-circular behaviors.

Stages of Consumption	Behaviors
Acquisition	Rent
Use	Repair Reuse
Final disposal	Return Recycle Resell

illustrates the stages of consumption and the previously defined pro-circular behaviors, classified in each of them.

2.2. Values, Beliefs, and Norms to Characterize Pro-Circular Consumers

The values-belief-norm (VBN) theory postulates that pro-environmental action or ecological behavior results from a combination of values, pro-environmental views, awareness of consequences, ascription of responsibility, and moral norms of each individual, which leads them to act in a certain way, according to the expectations they have of themselves [24]. Since its inception, the VBN theory has been widely used to analyze pro-environmental intentions and behaviors. For example, Steg et al. [42] examined the acceptance factors of energy policies according to the VBN theory, while Whitley et al. [43] implemented this theory to study sustainable behaviors in university students. These examples show the application of the VBN theory as a model to explain pro-environmental intentions and behaviors. Nevertheless, as this research aims to define a pro-circular consumer profile, it uses the components of the VBN theory as essential factors that can contribute to characterizing such a consumer rather than the model to explain behaviors.

The VBN theory encompasses three dimensions of values: self-interest, altruism toward other human beings, and altruism toward other species and nature [24]. This research, however, seeks to address values focused on sustainable consumption, considering them closer to its object of study, which is consumer behavior that resonates with the principles of a circular economy. To this end, it employs the approach proposed by Haws et al. [44], which introduces the construct of green consumption values and defines it as “the tendency to express the value of environmental protection through one’s purchases and consumption behaviors” (p. 2).

There are beliefs that individuals develop about the environment and how they relate to nature [24]. The VBN theory considers three types of beliefs: those involved in the New Ecological Paradigm (NEP) [45], beliefs about awareness, and beliefs about the responsibility of the individual toward environmental issues [24]. NEP beliefs measure the adverse consequences of environmental conditions and the interaction between people and the natural (ecological) environment. With regard to this interaction, there are beliefs about the power of man over nature (anthropocentric view) and beliefs in which people are only participants in the environment (ecocentric view) [46]. Similarly, beliefs about awareness are associated with recognizing potential threats that environmental conditions may pose on what an individual values, while beliefs about responsibility refer to the actions an individual could take in favor of alleviating or reducing such threats [42,47].

This research extends the belief component of the VBN theory by linking it to beliefs about perceived consumer effectiveness (PCE) as a crucial component to characterize pro-circular consumers. PCE has been conceptualized as the degree to which consumers believe that their individual actions can make a difference [48,49]; in this context, this means achieving an environmental objective or solving an environmental problem [50].

Finally, personal norms refer to the commitment to act correctly, representing an obligation felt by the individual to behave in a consistent manner in favor of the environment [24].

2.3. Sociodemographic Factors to Characterize Pro-Circular Consumers

Given that the explanatory power of sociodemographic factors in pro-environmental behaviors has been widely discussed in previous literature, these factors have been linked to the analysis, characterization, and profiling of green consumers [51,52]. This is because different authors have referred to sociodemographic variables as objective consumer characteristics that are identifiable, measurable, and useful for the segmentation of individuals [53]. Additionally, some studies have pointed out the possible relationship between these factors and environmentally sustainable attitudes and behaviors (e.g., [54,55,56,57,58,59].

2.4. Research Goals and Expectations

This research aims to identify a pro-circular consumer profile based on their behaviors and characterizing them based on the values, beliefs, and pro-environmental norms they claim to have. As an exploratory study, this research does not aim to test dependency relationships between the variables and behavior; however, it does use previous literature as a starting point to attempt to characterize consumers:

Consumer values might be one of the most compelling explanations and influences on consumer behavior [60]. Previous studies have evidenced a close relationship between values and pro-environmental intentions and behaviors [46,61]. Thus, when environmental conditions are believed to threaten people (awareness of consequences) and the ability of individuals to avoid these consequences and damage is recognized (attribution of responsibility), pro-environmental actions are expected to occur [24]. Different authors have demonstrated a link between awareness of the consequences of human activities in the environment and pro-environmental behaviors, as well as the explanatory power of the attribution of responsibility on environmentally sustainable behaviors [62,63].

With regard to beliefs about NEP, those associated with an ecocentric view may also explain the adoption of pro-environmental behaviors [24]. On the contrary, previous literature has pointed out that pro-environmental behaviors can rarely be predicted when an anthropocentric view emerges from NEP measurements [64].

As for beliefs about the effectiveness of consumer actions on the environment, previous literature has shown the possible predictive power of this variable on pro-environmental attitudes, intentions, and behaviors [49,65,66,67,68]. Finally, it has been pointed out that personal obligation precedes any behavior [46]; therefore, pro-environmental actions are the result of moral norms that each individual has regarding particular actions, which are evidenced in behaviors [24].

Consequently, given this prior literature on the relationship between psychographic variables and pro-environmental behaviors, we expect that green consumption values, beliefs about awareness and responsibility, and a view focused on the power of nature (i.e., ecocentric view) will characterize consumers with pro-circular behaviors. It is also expected that these consumers can be described in terms of moral norms and the perceived effectiveness of individual actions in the face of environmental issues.

In addition, this research builds on previous literature that has used sociodemographic variables in the analysis and characterization of pro-environmental consumers. It proposes that some of these factors may also contribute to characterizing pro-circular consumers, including age, gender, income level, educational level, marital status, occupation, type of housing, and household size.

3. Materials and Methods

3.1. Pilot Study to Choose Products to Measure Pro-Circular Behaviors

Based on previous literature on circular economy, five products were initially chosen that have been analyzed for their significant impact on the environment: household appliances and mobile devices, whose waste is estimated to increase between 3 and 5% annually, being the ones that grow the fastest compared to the other products [69,70,71,72]; private means of transportation, given that the automotive industry plays an outsized role in generating environmental damage [70]; furniture, since the consequences derived from deforestation (e.g., the extinction of species) are considered one of the most critical aspects of ecological deterioration on a global scale [70]; and, finally, clothing, as the excessive consumption and rotation of clothes, a phenomenon known as “fast-fashion,” has had adverse effects on the environment, including the impact on natural resources, the capacity to absorb greenhouse gas emissions, the discharge of hazardous chemicals into water sources, the increase in water use, and billions of tons of fashion waste entering landfills [20,70].

A measurement instrument was built to prioritize these products, considering the relation of consumers to each product category. For this purpose, a survey was applied to identify the three products with the highest importance and perceived value for individuals. The importance level of the products was measured using the scale and methodology proposed by Schneider et al. [73], while the scale proposed by Saura and Vivó [74] was used to measure perceived value by consumers. In the survey design, the means of transportation category was divided into car/motorcycle and bicycle, resulting in a total of six products to be evaluated in terms of importance and perceived value for consumers. There were 133 questions after combining the items of the abovementioned scales with the six product categories (see Table S1 in Supplementary Materials).

Based on the information received from 217 respondents, the mean was calculated for each product regarding the importance and perceived value variables. After adding the two values obtained for each product, the data were ordered, which led to a ranking that allowed for choosing three products to carry out the research (see

Table 2. Ranking of product prioritization according to importance and perceived value.

Product Category	Mean Importance	Mean Perceived Value	Total
Furniture	3.850	3.664	7.514 *
Car/Motorcycle	3.920	3.590	7.510 *
Clothing	3.627	3.745	7.372 *
Mobile devices	3.610	3.658	7.268
Bicycle	3.672	3.340	7.013
Household appliances	3.502	2.783	6.285

Note: (*) Corresponds to the selected products, based on the sum of the mean of both variables.

).

3.2. Population and Sample

For the development of this research, the city of Medellín in Colombia was chosen as a base, characterized by an innovative culture toward environmental sustainability, which has made it a protagonist and reference point at the national and regional levels. Medellín has been described as an ecocity, with efficient waste management systems, green spaces, and intelligent urban equipment models, identified even as the Latin American Capital of Electric Mobility [75,76]. Due to all this, it was considered the ideal place to carry out a circular economy analysis.

In Medellín, the study focused on the population of neighborhoods (Comunas) with a medium- to high-level quality of life (Comuna in Spanish refers to a subdivision of the urban area of the city of Medellín [77]). To determine this, neighborhoods with a Quality-of-Life Index (QLI) greater than 50 were used as selection criteria. (The QLI Index is a statistical compendium that measures the following variables: predominant material of the walls, predominant material of the floors, place from where the house receives water, waste management, sanitary service used, total number of household appliances, number of vehicles, educational level of the head of the household, educational level of the spouse of the head of the household, the proportion of children under 6 years, the proportion of children between 6 and 12 years of age who do not study, the proportion of children between 13 and 18 years of age who do not study, the proportion of illiterate persons, overcrowding, economic burden, social security of the head of the household, and the proportion of people in the household with social security [78]). These neighborhoods correspond to socioeconomic levels 3, 4, 5, and 6 in the city because when evaluating behaviors associated with reusing, repairing, renting, reselling, and recycling, there might be a distortion regarding behaviors carried out in neighborhoods with low quality of life. This might be a means of subsistence or the result of recursiveness, but not due to criteria associated with environmental sustainability, which is the purpose of a circular economy that frames this study.

Additionally, the population in these neighborhoods was delimited by age (between 25 and 60 years old).

Table 3. Percentage of population by neighborhood.

Neighborhood	Percentage
Poblado	16%
Laureles	15%
América	12%
Belén	21%
Candelaria	10%
Guayabal	11%
Buenos Aires	15%

shows the selected neighborhoods, weighting the population of each one of them over the sum of the selected ones (percentage presented next to their names). A sample of this population was selected to apply the measurement instrument, using the criteria and statistical developments determined by the literature [79].

Thus, data were collected through a survey programmed and distributed by Netquest, a market research company, to a sample of 417 people, according to a stratified sampling determined proportionally according to the percentages defined in Table 3. A selection criterion was established that stipulated that the panel of participants should have purchasing and consumption power over the products on which their behaviors would be evaluated (car/motorcycle, furniture, and clothing).

Of the sample (n = 417 respondents), 59% were female, and 41% were male, between 25 and 60 years of age. Most respondents were single (37%), followed by married (34%). Likewise, the majority were employed (57%), followed by self-employed (30%), and 41% had a professional educational level. Regarding their monthly income, 75% earned between COP $1,000,000 and $5,000,000; additionally, 53% lived in an apartment, and 41% in a house. Finally, as for household size, 57% live with 2 or 3 people and 35% with 4 to 6 people. Appendix A presents the description of the sample.

3.3. Description of Variables and Measuring Instrument

3.3.1. Target Variable: Pro-Circular Behaviors

As previously explained in the theoretical framework, this study contemplates six behaviors distributed at each stage of product consumption (i.e., acquisition, use, and final disposal). The behavior variable was measured using the scale proposed by Diddi and Yan [20], which measures the frequency level with 5 points (1 never; 5 always). Appendix B shows the details of the items used to operationalize each behavior according to their respective scale.

3.3.2. Association Variables to Characterize Behavior

Based on the components of the VBN theory as variables to characterize pro-circular consumers, the following variables were defined and measured to establish their association with the target variable. Green consumption values were measured using the scale proposed by Haws et al. [44], involving six items and a Likert scale with four levels of agreement (1. Strongly disagree, 2. Disagree, 3. Agree, 4. Strongly agree).

Beliefs about awareness (BA) were measured using the scale proposed by Choi et al. [47], involving five items and a Likert scale with four points measuring level of agreement (1. Strongly disagree, 2. Disagree, 3. Agree, 4. Strongly agree). Beliefs about responsibility (BR) were measured with the items of the scale proposed by Choi et al. and Ghazali et al. [47,80], involving three items and a Likert scale with four points measuring level of agreement (1. Strongly disagree, 2. Disagree, 3. Agree, 4. Strongly agree). On the other hand, the NEP scale brings together 15 items measured on a Likert scale with four points measuring level of agreement (1. Strongly disagree, 2. Disagree, 3. Agree, 4. Strongly agree) [45]. As previously mentioned in the theoretical framework, this research extends the belief construct of the VBN theory, involving the belief of perceived consumer effectiveness (PCE) as a component. This variable was measured by adapting the scale proposed by Ellen et al. [48]. With this scale, both the general (i.e., individual action in general) and specific PCE (i.e., each of the behaviors evaluated) were analyzed. Appendix B shows the details of the items used to operationalize each previously defined belief.

The variable of moral norms (MN) was measured using the scale proposed by Vining and Ebreo [81], which measures the level of agreement (1. Strongly disagree, 2. Disagree, 3. Agree, 4. Strongly agree) against three items (“I feel a strong obligation to…”; “I would feel guilty if…”; “I am willing to go the extra mile for…”). As shown in Appendix B, these items were adapted to the behaviors evaluated.

Finally, this research considered the sociodemographic factors that have been previously used in studies on pro-environmental attitudes, intentions, and behaviors, such as gender, age, marital status, occupation, educational level, income level, type of housing, and household size. As for previous variables, the operationalization of these factors is shown in Appendix B, presenting the variables, items, and scales used, which were adapted to the products and behaviors of interest defined in this research. Similarly, the

Table A2. Operationalization of Variables.

Variable	Authors	Items	Scale	$\mathbf{Cronbach} \mathit{\alpha }$
Target variable: Behavior				0.80
Rent	Scale adapted from Diddi & Yan (2019) [20]	How often do you: - Rent a car/motorcycle instead of buying a new one. - Rent a car/motorcycle service instead of using your own. - Rent clothes instead of buying new ones.	Never Rarely Occasionally Often Always	0.66
Repair	Scale adapted from Diddi & Yan (2019) [20]	How often do you: - Repair your car/motorcycle instead of thinking about buying a new one. - Repair your furniture (e.g., chairs, tables, desks) instead of buying new ones. - Repair your clothes instead of buying new ones.	Never Rarely Occasionally Often Always	0.71
Reuse	Scale adapted from Diddi & Yan (2019) [20]	How often do you: - Reuse your clothing. - Reuse your furniture (e.g., chairs, tables, desks).	Never Rarely Occasionally Often Always	0.76
Return for Recycling	Scale adapted from Diddi & Yan (2019) [20]	How often do you: - Return your clothes when you no longer use them, so that they are recycled.	Never Rarely Occasionally Often Always	N/A
Resell	Scale adapted from Diddi & Yan (2019) [20]	How often do you: - Resell your car/motorcycle. - Resell your furniture (e.g., chairs, tables, desks)	Never Rarely Occasionally Often Always	0.48
Characterization variables
Values	Haws et al. (2014) [44]	Indicate your level of agreement with the following statements: - It is important to me that the products I use do not harm the environment. - I consider the potential environmental impact of my actions when making many of my decisions. - My buying habits are affected by my concern for our environment. - I am concerned about the waste of resources on our planet. - I would describe myself as environmentally responsible. - I am willing to make myself uncomfortable to take actions that are more environmentally friendly.	I strongly disagree I disagree I agree I strongly agree	0.86
Beliefs of conscience	Choi et al. (2015) [47]	Indicate your level of agreement with the following statements: - It is true that global warming is a real problem. - Global warming is a problem for society. - Saving energy can reduce global warming. - The depletion of natural resources (oil, coal, natural gas) is a problem. - The depletion of energy sources is a problem.	I strongly disagree I disagree I agree I strongly agree	0.84
Beliefs of responsibility	Choi et al. (2015); Ghazali et al. (2019) [47,80]	Indicate your level of agreement with the following statements: - I feel jointly responsible for the energy problems. - I feel jointly responsible for the depletion of energy sources. - I feel jointly responsible for global warming.	I strongly disagree I disagree I agree I strongly agree	0.91
Beliefs of the new ecological paradigm	Liere and Dunlap (1980) [45]	Indicate your level of agreement with the following statements: - When humans interfere with nature, the consequences can be disastrous. - Plants and animals have as much right to live as human beings. - Humans are seriously abusing the environment. - The balance of nature is very delicate and easily affected. - If things continue their current course, we will soon be facing a major environmental catastrophe. - We are approaching the maximum number of people the Earth can support. - The Earth is like a ship with very limited space and resources. - Despite our special abilities, human beings are still subject to the laws of nature. - The Earth has an abundance of resources if only we would learn how to develop them. - Human beings have the right to modify the natural environment to suit their needs. - Human ingenuity will ensure that we do NOT make the Earth uninhabitable. - The balance of nature is strong enough to cope with the impacts of modern industrial nations. - The so-called “ecological crisis” that humanity is facing has been greatly exaggerated. - Human beings claimed to rule over the rest of nature. - Humans will eventually learn enough of how nature works to be able to control it.	I strongly disagree I disagree I agree I strongly agree	0.76
Perceived consumer effectiveness General	Ellen et al. (1991) [48]	Indicate your level of agreement with the following statements: - As an individual, I feel that I can contribute to solving the loss of biodiversity (flora and fauna) through my actions. - As an individual, I feel that I can contribute to preventing pollution and the indiscriminate use of water through my actions. - As an individual, I feel that I can contribute to preventing global warming through my actions. - As an individual, I feel that I can contribute to solving the existing waste-related problems through my actions. - As an individual, I feel that I can contribute to solving environmental problems through my actions.	I strongly disagree I disagree I agree I strongly agree	0.91
Personal Norms	Scale adapted from Vining & Ebreo (1992) [81]	Indicate your level of agreement with the following statements - I feel a strong personal obligation to buy a used car/motorcycle instead of buying a new one. - I feel a strong personal obligation to repair my furniture instead of buying new ones. - I feel a strong personal obligation to repurpose my clothes instead of discarding them. - I feel a strong personal obligation to rent a car/motorcycle service instead of using my car/motorcycle. - I feel a strong personal obligation to return my clothes to a point of sale or collection when I no longer wear them, so that they are recycled. - I am willing to go the extra mile to buy a used car/motorcycle instead of buying a new one. - I am willing to go the extra mile to repurpose my furniture (e.g., tables, chairs, desks) before discarding them. - I am willing to go the extra mile to repair my clothes instead of buying new ones. - I am willing to go the extra mile to rent a transportation service instead of using my car/motorcycle. - I am willing to go the extra mile to resell my furniture (e.g., tables, chairs, desks). - I feel a strong personal obligation to repair my car/motorcycle instead of buying a new one. - I feel a strong personal obligation to repurpose my furniture (e.g., tables, chairs, desks) before discarding them. - I feel a strong personal obligation to resell my furniture (e.g., tables, chairs, desks). - I feel a strong personal obligation to rent my clothes instead of buying new ones. - I feel a strong personal obligation to repair my clothes instead of buying new ones. - I would feel guilty if instead of buying a used car/bike I bought a new one. - I would feel guilty if I did not repurpose my clothes before discarding them. - I would feel guilty if instead of repairing my furniture I bought new ones. - I would feel guilty if instead of renting a transportation service I used my car/motorcycle. - I would feel guilty if instead of repairing my clothes I bought new ones. - I would feel guilty if I did not resell my furniture (e.g., tables, chairs, desks). - I would feel guilty if instead of renting clothes I bought new ones. - I am willing to go the extra mile to repair my car/motorcycle instead of buying a new one. - I am willing to go the extra mile to repair my furniture instead of buying new ones. - I am willing to go the extra mile to rent my clothes instead of buying new ones. - I am willing to go the extra mile to repurpose my clothes instead of discarding them. - I am willing to go the extra mile to return my clothes to a point of sale or collection when I no longer use them, so that they are recycled. - I would feel guilty if instead of repairing my car/motorcycle I bought a new one. - I would feel guilty if I did not repurpose my furniture (e.g., tables, chairs, desks) before discarding them. - I would feel guilty if I did not return my clothes to a point of sale or collection when I no longer use them, so that they are recycled.	I strongly disagree I disagree I agree I strongly agree	0.94
Specific perceived effectiveness	Scale adapted from Ellen et al. (1991) [48]	Indicate your level of agreement with the following statements: - By renting a transportation service, instead of buying a new car or motorcycle, I can contribute to solving environmental problems. - By repairing my car/motorcycle, instead of buying a new one, I can contribute to solving environmental problems. - By reusing my furniture (e.g., tables, chairs, desks), I can contribute to solving environmental problems. - By returning my clothes to a point of sale or collection when I no longer use them, so that they are recycled, I can contribute to solving environmental problems. - By reselling my car/motorcycle, I can contribute to solving environmental problems.	I strongly disagree I disagree I agree I strongly agree	0.90
Sociodemographic Variables		Gender Age Marital status Occupation Level of studies Level of income Type of housing Home size

Source: Own elaboration.

in Appendix B specifies the Cronbach’s alpha for each variable, which is the most commonly used criterion to evaluate the internal consistency of the items of a construct [82]. As specified in Appendix B, all Cronbach’s alphas were higher than 0.7, demonstrating the reliability of the scales used.

3.4. Analysis Procedure

3.4.1. Selection of Components and Elaboration of Descriptive Statistics

To reduce the dimensionality of several items of a construct into a few components, the principal component analysis (PCA) methodology was used. The PCA is a multivariate statistical technique that allows dividing the information of a set of variables into a reduced number of new unobservable independent variables, explaining a specific part of the information and contributing to the variability of the original variables. This methodology was used to reduce the number of variables (items) and build a single construct without affecting the properties of the original variables, to then establish future associations between the characterization variables (i.e., values, beliefs, norms, and sociodemographic factors) and the variable of interest (i.e., behaviors).

This methodology was used to verify that each independent variable, operationalized through several items, could be taken as a single construct to establish future relationships between the explanatory variables and the dependent variable of behaviors. For the latter, the purpose was to reduce the 11 behavioral items to a smaller number of categories that could be associated with the stages of consumption described in the theoretical framework. However, given that this is an exploratory research study, it is possible to determine, based on the data, other categories that will be described later in the section of analysis to classify pro-circular consumers.

3.4.2. Sampling Adequacy Statistics

As a previous step to the PCA, intercorrelation among the items had to be verified. To do this, the Bartlett sphericity test was performed, with the null hypothesis that the variables are not intercorrelated (the correlation matrix is an identity matrix). Based on this, the conclusion was that all items of each construct are significant at 1%; that is, the items within each construct serve to reduce their dimensions in components (see

Table 4. Adequacy measures and number of components.

Construct	Bartlett Sphericity Test (Chi-Square)	KMO Sampling Adequacy Measure	Number of Items	Number of Components Retained
Behaviors	860.23 ***	0.84	11	2
Stage of acquisition and purchase	378.25 ***	0.7	6	2
Stage of use	25.65 ***	0.57	3	1
Stage of final disposal	204.23 ***	0.5	2	1
Values	1115.41 ***	0.87	6	1
NEP beliefs	2641.08 ***	0.88	15	3
Beliefs about awareness	1217.84 ***	0.73	5	1
Beliefs about responsibility	935.90 ***	0.72	3	1
General PCE	1536.11 ***	0.85	5	1
Specific PCE	1635.63 ***	0.89	11	3
Moral norms	5188.13 ***	0.9	30	7
Perceived value	6399.31 ***	0.81	36	9

Note: *** Significant at 1%.

).

As a complement to the Bartlett sphericity test, the Kaiser-Meyer-Olkin (KMO) adequacy test was performed for the PCA, which allows measuring correlation among the items. A high value (above 0.8) is considered appropriate for the PCA. A value below 0.5 would not be adequate because the correlations between pairs of variables could not be explained by other variables. This analysis prompted the conclusion that PCA should be used. According to Table 4, the lowest value was 0.5 for behaviors at the final disposal stage, and the highest value was 0.9 for moral norms.

To calculate the descriptive statistics of each construct, in cases where the number of components was higher than 1 (see Table 4, last column), these were first normalized with the expression:

$$\frac{X-X_{Minimum}}{X_{Maximum}-X_{Minimum}}$$

$X$ being the Z-value of each observation within each component. Subsequently, they were averaged to obtain a single representative value for each construct.

Table 5. Descriptive statistics for each construct.

Construct	Mean	Standard Deviation	Median	Asymmetry
Behaviors	0.46	0.18	0.45	0.28
Stage of acquisition and purchase	0.41	0.18	0.40	0.39
Stage of use	0.55	0.21	0.56	0.01
Stage of final disposal	0.63	0.23	0.63	−0.46
Values	0.73	0.17	0.73	−0.74
NEP beliefs	0.70	0.13	0.69	−0.08
Beliefs about awareness	0.87	0.16	0.94	−2.27
Beliefs about responsibility	0.70	0.24	0.67	−0.50
General PCE	0.80	0.18	0.80	−0.77
Specific PCE	0.64	0.17	0.65	−0.31
Moral norms	0.58	0.16	0.58	−0.22

presents the results of the previous procedure, where mean, median, standard deviation, and bias or asymmetry coefficient values are shown for each construct.

3.4.3. Statistical Analyses

To simultaneously analyze a set of variables, different multivariate analysis techniques were reviewed, choosing the one known as multiple correspondence analysis (MCA). This technique allows identifying the association between qualitative variables—which are the variables that in their entirety correspond to those analyzed in this research—mainly due to the ordinal character represented by their measurement scales.

The MCA is a descriptive technique that seeks to summarize a large amount of information in a reduced number of dimensions while losing the least amount of data. It is used to analyze from a graphical point of view the relationships of dependence and independence in a set of categorical variables based on the data of a contingency table.

In this technique, the proximity between points representing modalities or categories is associated with the high probability of simultaneous mentions by observation units. Low-frequency modalities are found at the ends and are used to characterize the axes, while proximity to the center means high percentages of mentions; in other words, it records the most common responses given by the information units [83,84]. Therefore, the categories farther away from the origin will be more differentiated, based on the data, and with a discriminatory power, and those closer to the origin will not have information differentiating them. Additionally, the farther away the categories are from the origin, the higher the level of association with nearby points [83,84]. In this order of ideas, the categories analyzed in this study are those far from the origin and, therefore, can be highly differentiated and with a higher level of association with nearby items.

For this research, the number of categories or constructs was reduced through the MCA with the following principal constructs: Behaviors, Values, NEP beliefs, Beliefs about awareness, Beliefs about responsibility, General PCE, and Specific PCE. Additionally, supplementary variables are included, such as stages of consumption and sociodemographic factors (e.g., age, gender, educational level, income level, and household size).

Table 6. MCA reliability measures.

Dimension	Cronbach’s Alpha	Total (Eigenvalue)	Inertia	Percentage of Variance	Cumulative Variance (%)
Dim 1	0.806	3.235	0.462	46.21	46.21
Dim 2	0.761	2.877	0.411	41.10	87.31

below shows the different reliability indicators of each dimension, such as adequacy based on Cronbach’s alpha, inertia (relative measure of the importance of the dimensions), and the percentage of variance. A sedimentation graph was used to determine the appropriate number of dimensions to summarize the total number of categories (see Figure 1); according to this, the selected ones are the first two dimensions that, as shown in Table 6, explain 87.31% of data variance, with a Cronbach’s alpha of 0.81 and 0.76, respectively.

After obtaining the two dimensions standardized by MCA, the K-means algorithm was used to obtain the profiles of the pro-circular consumer according to their psychographic variables and behaviors. In this approach, we follow [85]. As Figure 2 shows, the cluster number assigned to the algorithm was obtained from the elbow method.

4. Results

4.1. Descriptive Statistics

As shown in

Table 7. Descriptive results.

	Mean (SD)	Behaviors	ACQ	USE	DISP	Values	NEP	BA	BR	G PCE	S PCE	MN
Behaviors	0.46 (0.18)	1
ACQ	0.41 (0.18)	0.9464 *	1
USE	0.55 (0.21)	0.6883 *	0.4998 *	1
DISP	0.63 (0.23)	0.7155 *	0.6072 *	0.4328 *	1
Values	0.73 (0.17)	0.3486 *	0.3743 *	0.1593 *	0.2504 *	1
NEP	0.70 (0.13)	0.2864 *	0.3151 *	0.1445 *	0.2440 *	0.3107 *	1
BA	0.87 (0.16)	0.3273 *	0.3279 *	0.1710 *	0.2527 *	0.3711 *	0.3285 *	1
BR	0.70 (0.24)	0.2869 *	0.3055 *	0.1057	0.2189 *	0.4087 *	0.2721 *	0.4735 *	1
G PCE	0.80 (0.18)	0.2537 *	0.2372 *	0.1575 *	0.2646 *	0.5068 *	0.2817 *	0.4048 *	0.4167 *	1
S PCE	0.64 (0.17)	0.5639 *	0.5794 *	0.3840 *	0.3775 *	0.4438 *	0.4170 *	0.3449 *	0.4955 *	0.5131 *	1
MN	0.58 (0.16)	0.6752 *	0.6844 *	0.4461 *	0.4957 *	0.4457 *	0.3990 *	0.2753 *	0.3754 *	0.4031 *	0.7923 *	1

Note: (*) corresponds to a percentage of significance greater than or equal to 5%.

, the results suggest that the population of the seven neighborhoods under study perform pro-circular behaviors occasionally: Table 7 (M = 0.46, SD = 0.18) according to the scale used: 0–0.20 = never, 0.20–0.40 = rarely, 0.40–0.60 = occasionally, 0.60–0.80 = often, 0.80–100 = always), which coincides with the three stages of consumption analyzed. In particular, the results show that, on average, respondents rarely perform pro-circular behaviors at the stage of acquisition ($M_{acq}$ = 0.41, SD = 0.18). With regard to the stages of product use and final disposal, they occasionally perform this type of behavior ($M_{use}$ = 0.54, SD = 0.21; and $M_{disp}$ = 0.63, SD = 0.23).

On the other hand, according to the scale 0–0.25 = Strongly disagree; 0.25–0.50 = Disagree; 0.50–0.75 = Agree; 0.75–100 = Strongly agree, the population in these neighborhoods exhibited values that resonate with green consumption ($M_{val}$ = 0.73, SD = 0.17), beliefs about responsibility ($M_{BR}$ = 0.7, SD = 0.24), and beliefs about the relationship between man and nature, as measured by NEP ($M_{NEP}$ = 0.7, SD = 0.13). Thus, the population under study agreed to actions in favor of environmental care, avoiding actions that are adverse to the environment. Likewise, the neighborhoods analyzed, on average, strongly agreed that environmental decline is a real problem with potentially negative consequences for society, showing a high awareness of what is happening with the environment ($M_{CC}$ = 0.87, SD = 0.16) (see Table 7).

With regard to general PCE, the population in the analyzed neighborhoods of Medellín, on average, stated that they strongly agree ($M_{EPCg}$ = 0.80; SD = 0.18) that their actions can contribute to solving and avoiding environmental problems. As for the perceived effectiveness of specific actions associated with pro-circular behaviors, however, the conviction is not the same. Although there is still a positive belief about what an individual’s actions can achieve in the environment, it is no longer strong ($M_{EPCe}$ = 0.64, SD= 0.17).

With regard to moral norms, as shown in Table 7, consumers consider that they somewhat agree to feeling a personal obligation to behave coherently in favor of the environment ($M_{MN}$ = 0.58, SD = 0.16).

Table 7 also shows the correlations between the variables. Although high correlations are found between pro-circular behaviors and all stages of consumption ($M_{acq}$ = 0.94; $M_{use}$ = 0.68; $M_{disp}$ = 0.71), the results suggest that, at the acquisition or purchase stage (ACQ), behaviors would be more related to the construct of pro-circular behaviors. The behaviors have significant correlations with all attitudinal variables from the components of the VBN theory and its extension, suggesting that these variables could characterize pro-circular consumers. The high correlation of pro-circular behaviors with perceived consumer-specific effectiveness (S PCE) and moral norms (MN) stands out (0.56 and 0.67, respectively) (see Table 7). In both cases, the items used to measure these variables were adapted by linking pro-circular behaviors, so there might be a higher association between the behavior performed, feelings of personal obligation, and the perception of effectiveness of this behavior in positively impacting the environment.

When discriminating by stages of consumption, it is worth noting the correlation of perceived effectiveness in specific terms and moral norms with the product acquisition and purchase stage (0.58 and 0.68, respectively). This finding could suggest that people shift personal obligations and the perception that their individual actions for the environment are relevant to pro-circular behaviors of renting and repairing products to a greater extent.

Regarding correlations between attitudinal variables, significant correlations were found between values and beliefs about responsibility (0.41), general (0.50) and specific PCE (0.44), as well as between values and moral norms (0.44) (see Table 7). Other correlations speak of the association between beliefs about the relationship between man and nature (i.e., NEP beliefs) and specific PCE (0.42). It is possible that the belief about the relevance of individual actions through specific behaviors (specific PCE) to avoid negative impacts or solve environmental problems is largely associated with beliefs about adverse effects generated in the environment, mainly those that refer to the consequences of human actions.

As shown in Table 7, beliefs about awareness (BA) correlate with beliefs about responsibility (BR) and general PCE (0.47 and 0.40, respectively). This result suggests that, by being aware of the adverse consequences that environmental decline might have on society (BA), individuals assume their role, believing that they are responsible (BR) for the mitigation of such problems and, in addition, that their individual actions are relevant (PCE) for solving or avoiding such consequences. This is also evidenced in the correlation between beliefs about responsibility (BR) and PCE, both actions in general ($PC{E}_G$ = 0.41) and specific actions ($PC{E}_S$ = 0.49) (see Table 5).

The correlation between general and specific PCE (0.51) explains the belief in the relevance of individual pro-environmental actions and how that belief becomes more tangible through concrete behaviors. Finally, Table 7 shows a strong association between the effectiveness consumers perceive regarding their actions or behaviors (specific PCE) and moral norms (0.79). This indicates that the obligation to behave in the right way, through pro-circular behaviors, makes the individual believe that they can make a difference in environmental care.

4.2. Cluster Analysis

Through cluster analysis, we could group consumers into four profiles based on the frequency of their behaviors (see Figure 3 and

Table 8. Clusters characteristics: pro-circular behaviors.

			Cluster 1 (n = 115)		Cluster 2 (n = 148)		Cluster 3 (n = 125)		Cluster 4 (n = 29)
Behavior		Scale	n	% Col	n	% Col	n	% Col	n	% Col
Rent	Car/motorcycle instead of buying a new one	Never	57	74.0	37	43.0	46	55.4	16	80.0
		Rarely	12	15.6	27	31.4	12	14.5	3	15.0
		Occasionally	6	7.8	18	20.9	10	12.0	1	5.0
		Often	2	2.6	2	2.3	11	13.3	0	0.0
		Always	0	0.0	2	2.3	4	4.8	0	0.0
		Total	77	100.0	86	100.0	83	100.0	20	100.0
	Clothes instead of buying new ones	Never	68	59.1	36	24.3	26	20.8	27	93.1
		Rarely	35	30.4	55	37.2	34	27.2	2	6.9
		Occasionally	10	8.7	37	25.0	33	26.4	0	0.0
		Often	2	1.7	19	12.8	16	12.8	0	0.0
		Always	0	0.0	1	0.7	16	12.8	0	0.0
		Total	115	100.0	148	100.0	125	100.0	29	100.0
	Car/motorcycle service instead of using your own	Never	37	48.1	14	16.3	16	19.3	17	85.0
		Rarely	22	28.6	37	43.0	18	21.7	0	0.0
		Occasionally	16	20.8	31	36.0	26	31.3	3	15.0
		Often	2	2.6	4	4.7	19	22.9	0	0.0
		Always	0	0.0	0	0.0	4	4.8	0	0.0
		Total	77	100.0	86	100.0	83	100.0	20	100.0
Return	Clothes when you no longer use them, so that they are recycled	Never	43	37.4	19	12.8	23	18.4	16	55.2
		Rarely	23	20.0	20	13.5	10	8.0	4	13.8
		Occasionally	21	18.3	37	25.0	16	12.8	1	3.4
		Often	21	18.3	50	33.8	36	28.8	2	6.9
		Always	7	6.1	22	14.9	40	32.0	6	20.7
		Total	115	100.0	148	100.0	125	100.0	29	100.0
Repair	Car/motorcycle instead of thinking about buying a new one	Never	2	2.6	0	0.0	0	0.0	0	0.0
		Rarely	5	6.5	3	3.5	0	0.0	3	15.0
		Occasionally	28	36.4	13	15.1	10	12.0	5	25.0
		Often	32	41.6	54	62.8	23	27.7	4	20.0
		Always	10	13.0	16	18.6	50	60.2	8	40.0
		Total	77	100.0	86	100.0	83	100.0	20	100.0
	Furniture instead of buying new ones	Never	1	0.9	0	0.0	1	0.8	16	55.2
		Rarely	39	33.9	2	1.4	6	4.8	4	13.8
		Occasionally	64	55.7	78	52.7	19	15.2	3	10.3
		Often	11	9.6	67	45.3	56	44.8	1	3.4
		Always	0	0.0	1	0.7	43	34.4	5	17.2
		Total	115	100.0	148	100.0	125	100.0	29	100.0
	Clothes instead of buying new ones	Never	17	14.8	0	0.0	2	1.6	14	48.3
		Rarely	37	32.2	12	8.1	9	7.2	5	17.2
		Occasionally	51	44.3	62	41.9	20	16.0	5	17.2
		Often	10	8.7	71	48.0	42	33.6	2	6.9
		Always	0	0.0	3	2.0	52	41.6	3	10.3
		Total	115	100.0	148	100.0	125	100.0	29	100.0
Resell	Furniture	Never	19	16.5	5	3.4	9	7.2	11	37.9
		Rarely	40	34.8	27	18.2	14	11.2	7	24.1
		Occasionally	36	31.3	51	34.5	30	24.0	6	20.7
		Often	19	16.5	63	42.6	26	20.8	2	6.9
		Always	1	0.9	2	1.4	46	36.8	3	10.3
		Total	115	100.0	148	100.0	125	100.0	29	100.0
	Car/motorcycle	Never	5	6.5	2	2.3	3	3.6	5	25.0
		Rarely	15	19.5	11	12.8	8	9.6	5	25.0
		Occasionally	29	37.7	35	40.7	23	27.7	4	20.0
		Often	13	16.9	23	26.7	19	22.9	5	25.0
		Always	15	19.5	15	17.4	30	36.1	1	5.0
		Total	77	100.0	86	100.0	83	100.0	20	100.0
Reuse	Furniture	Never	3	2.6	0	0.0	0	0.0	10	34.5
		Rarely	29	25.2	1	0.7	3	2.4	7	24.1
		Occasionally	61	53.0	58	39.2	12	9.6	4	13.8
		Often	19	16.5	87	58.8	54	43.2	1	3.4
		Always	3	2.6	2	1.4	56	44.8	7	24.1
		Total	115	100.0	148	100.0	125	100.0	29	100.0
	Clothing	Never	12	10.4	0	0.0	2	1.6	18	62.1
		Rarely	37	32.2	6	4.1	4	3.2	0	0.0
		Occasionally	49	42.6	49	33.1	16	12.8	3	10.3
		Often	14	12.2	82	55.4	46	36.8	4	13.8
		Always	3	2.6	11	7.4	57	45.6	4	13.8
		Total	115	100.0	148	100.0	125	100.0	29	100.0

). To describe these profiles, we used psychographic variables (see Figure 4 and

Table 9. Clusters characteristics: values, beliefs, and personal norms.

		Cluster 1 (n = 115)		Cluster 2 (n = 148)		Cluster 3 (n = 125)		Cluster 4 (n = 29)
Variable	Scale	n	% Col	n	% Col	n	% Col	n	% Col
Values	Strongly disagree	0	0.0	0	0.0	1	0.8	5	17.2
	Disagree	24	20.9	7	4.7	5	4.0	3	10.3
	Agree	75	65.2	69	46.6	26	20.8	12	41.4
	Strongly agree	16	13.9	72	48.6	93	74.4	9	31.0
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Beliefs of the new ecological paradigm	Strongly disagree	0	0.0	0	0.0	0	0.0	1	3.4
	Disagree	1	0.9	2	1.4	0	0.0	1	3.4
	Agree	91	79.1	89	60.1	56	44.8	23	79.3
	Strongly agree	23	20.0	57	38.5	69	55.2	4	13.8
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Beliefs of Awareness	Strongly disagree	0	0.0	0	0.0	0	0.0	5	17.2
	Disagree	5	4.3	0	0.0	0	0.0	1	3.4
	Agree	39	33.9	22	14.9	11	8.8	13	44.8
	Strongly agree	71	61.7	126	85.1	114	91.2	10	34.5
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Beliefs of responsibility	Strongly disagree	2	1.7	1	0.7	7	5.6	8	27.6
	Disagree	32	27.8	10	6.8	10	8.0	8	27.6
	Agree	71	61.7	73	49.3	27	21.6	12	41.4
	Strongly agree	10	8.7	64	43.2	81	64.8	1	3.4
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Perceived consumer effectiveness General	Strongly disagree	0	0.0	0	0.0	0	0.0	3	10.3
	Disagree	6	5.2	4	2.7	1	0.8	1	3.4
	Agree	80	69.6	76	51.4	16	12.8	17	58.6
	Strongly agree	29	25.2	68	45.9	108	86.4	8	27.6
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Specific perceived effectiveness	Strongly disagree	3	2.6	0	0.0	0	0.0	4	13.8
	Disagree	37	32.2	2	1.4	2	1.6	18	62.1
	Agree	71	61.7	116	78.4	26	20.8	7	24.1
	Strongly agree	4	3.5	30	20.3	97	77.6	0	0.0
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Personal norms	Strongly disagree	3	2.6	0	0.0	0	0.0	5	17.2
	Disagree	62	53.9	11	7.4	3	2.4	20	69.0
	Agree	50	43.5	129	87.2	64	51.2	4	13.8
	Strongly agree	0	0.0	8	5.4	58	46.4	0	0.0
	Total	115	100.0	148	100.0	125	100.0	29	100.0

) and sociodemographic variables (see Figure 5 and

Table 10. Clusters characteristics: sociodemographic variables.

		Cluster 1 (n = 115)		Cluster 2 (n = 148)		Cluster 3 (n = 125)		Cluster 4 (n = 29)
Variable	Categories	n	% Col	n	% Col	n	% Col	n	% Col
Gender	Male	47	40.9	65	43.9	50	40.0	12	41.4
	Female	68	59.1	83	56.1	75	60.0	17	58.6
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Age	0–24	0	0.0	0	0.0	0	0.0	0	0.0
	25–29	17	14.8	28	18.9	21	16.8	4	13.8
	30–34	21	18.3	23	15.5	20	16.0	6	20.7
	35–39	19	16.5	21	14.2	22	17.6	3	10.3
	40–44	20	17.4	18	12.2	19	15.2	4	13.8
	45–49	9	7.8	15	10.1	13	10.4	7	24.1
	50–54	16	13.9	22	14.9	14	11.2	1	3.4
	55–60	13	11.3	21	14.2	16	12.8	4	13.8
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Level of studies	Elementary school	0	0.0	1	0.7	0	0.0	0	0.0
	High school	4	3.5	4	2.7	8	6.4	2	6.9
	Associate’s degree	13	11.3	17	11.5	11	8.8	3	10.3
	Associate’s degree	15	13.0	28	18.9	18	14.4	4	13.8
	Bachelor’s degree	50	43.5	62	41.9	49	39.2	9	31.0
	Specialization	24	20.9	25	16.9	33	26.4	10	34.5
	Master’s degree	8	7.0	8	5.4	5	4.0	1	3.4
	Doctorate degree	1	0.9	3	2.0	1	0.8	0	0.0
	Total	115	100.0	148	100.0	125	100.0	29	100.0
Level of income	Less than $1.000.000	9	7.8	14	9.5	9	7.2	4	13.8
	$1.000.000–$3.000.000	55	47.8	63	42.6	56	44.8	12	41.4
	$3.000.001–$5.000.000	30	26.1	50	33.8	38	30.4	9	31.0
	$5.000.001–$10.000.000	17	14.8	15	10.1	22	17.6	2	6.9
	More than $10.000.000	4	3.5	6	4.1	0	0.0	2	6.9
	Total	115	100.0	148	100.0	125	100.0	29	100.0

).

Away from pro-circularity (Cluster 4): 7% of the sample is in this cluster. In this cluster, consumers do not show pro-circular behaviors. Mainly, they never reuse or repair (furniture or clothing) instead of buying new products. This consumer is characterized by strongly disagreeing with green consumption values and beliefs about global warming, depletion of resources, and their consequences (beliefs of awareness). Simultaneously, they do not feel responsible, morally obligated, or able to contribute to addressing environmental problems. Characterizing this cluster by sociodemographic factors is challenging, given that there is no apparent commonality.

Rarely and occasionally behave pro-circular (Cluster 1): This cluster contains 28% of the sample. These consumers’ behaviors do not favor circularity; however, some are less radical than cluster 4, moving from never to rarely and occasionally acting pro-circularly. For example, although they never rent or return clothes to be recycled, they rarely and occasionally repair, resell, and reuse. Even though the consumers in this group have pro-environmental values and are aware of environmental challenges, they do not believe they have the responsibility and power to act, through pro-circular behaviors, to face those problems. Moreover, they disagree with feeling a personal norm towards pro-circular behavior. While this type of consumer is between 40 and 44 years and holds a master’s degree, they represent 7% of this profile. Therefore, there are no apparent sociodemographic factors to describe consumers in this cluster.

On the path toward pro-circular behavior (Cluster 2): 35% of the sample is in this cluster. The results suggest that they are on the path toward pro-circular behaviors because they ‘often’ act this way. They rarely rent and occasionally repair their furniture or resell their car/motorcycle. However, they often reuse, resell furniture, and repair clothes and their car/motorcycle instead of buying new ones. In addition, they often return their clothing so that it can be recycled. These consumers are aware of environmental problems and consequences, feeling a moral obligation to act. Furthermore, they believe their pro-circular behaviors (e.g., reuse, resell, recycle) are relevant for addressing environmental challenges. These consumers are in two age brackets, below 34 and above 45 years; they have bachelor’s degrees following associate degrees and an income of less than 5 million Colombian pesos ($5.000.000).

Pro-circular consumers (Cluster 3): In this cluster is 30% of the sample. These are pro-circular consumers because they often rent and always reuse, repair, resell and even return to recycle (in the categories surveyed, at least). Most of these consumers strongly agree with green consumption values and beliefs about the adverse consequences of environmental conditions and the interaction between people and the natural environment (i.e., NEP beliefs). In addition, they firmly believe they have the responsibility and the capacity to make a difference with their actions, particularly with pro-circular behaviors. Furthermore, they behave this way because they strongly agree that this is the right thing to do. They are people between 35 and 39 years old, with specialization studies and income levels between 5 million ($5.000.000) and 10 million ($10.000.000) Colombian pesos.

4.3. Multiple Correspondence Analysis

We used the MCA explained in Section 3.4.3 within the analysis procedure to deepen this last profile of pro-circular consumers. As previously mentioned in the methodology section, the variables far from the map’s point of origin are analyzed since the farther away from the origin, the greater the differentiation among data and the greater the association level. In this case, this allows better discrimination between pro-circular behaviors and characteristics of those who perform them, which favors the profiling purpose of this study.

4.3.1. Pro-Circular Consumer Profile

The first step of profiling pro-circular consumers is to classify them based on their behaviors and the degree of association between them. As Figure 6 shows, there is a strong association among the behaviors of repairing, reusing, and reselling (see blue circle). Thus, it is possible to identify those pro-circular consumers who think about extending the useful life of their products, so before buying new items, they try to reuse and repair them (e.g., clothing and furniture). Additionally, in the specific case of the furniture category, they seek the option of reselling them, giving them another opportunity for use at the final disposal stage.

A particular similarity is found between behaviors of renting in the categories of clothing and means of transportation (car/motorcycle) (see yellow circle). Regarding clothing, renting is done to avoid purchasing new clothes, while in the case of a vehicle, consumers aim to avoid using their own car/motorcycle when renting one.

Finally, other behaviors were not associated with any other and, therefore, are not grouped into a specific category of pro-circular consumer: renting or repairing a car/motorcycle instead of buying new ones, reselling these goods, and returning clothing items to be recycled.

The second step is to characterize each sub-category of pro-circular consumers with the variables of interest (i.e., values, beliefs, norms, and sociodemographic factors). As in the first step, we follow the principle of MCA, which allows us to analyze the variables far from the map’s point of origin. The farther away from the origin, the greater is the differentiation among data and the greater the association level. Thus, although within cluster 3 (pro-circular consumers) most people exhibited pro-environmental values, beliefs, and norms, it may be that not all of these variables are strongly associated with the specific pro-circular categories and behaviors analyzed in this group. Hence, we focused on those variables far from the origin to profile the pro-circular consumers identified above. As Figure 7 shows, those variables are personal norms and specific perceived consumer effectiveness (see red square). Then, the proximity visualized among the variables and categories analyzed is the criterion that allows us to identify their association or similarity.

Consumers Who Extend the Life of Products (Repairing, Reusing, and Reselling)

As Figure 8 shows, consumers who repair, reuse, and resell their furniture and those who repair and reuse clothing are characterized by strong moral norms to act (see the proximity between the blue circle and being ‘Strongly agree’ with personal norms). Thus, they are consumers who feel an obligation and willingness to go the extra mile to repair/reuse rather than buy new products. Moreover, they exhibit a moral norm, a personal commitment to do the right thing, which guides them to resell their furniture before discarding it.

An additional and specific characteristic of consumers who reuse their clothing items is that they do so because they consider that, with this concrete action, they contribute to solving environmental problems (PCE). (See the proximity between ‘reuse your clothing’ and being ‘Strongly agree’ with specific perceived effectiveness.) These consumers are mainly people with an educational level equivalent to specialization, with an average monthly income between 3 and 5 million Colombian pesos. However, they are challenging to characterize in other sociodemographic terms, since the factorial map does not show clear associations.

Consumers Who Rent

The consumers who rent clothes to avoid buying them and those who rent cars/motorcycles to prevent using their private ones could not be characterized. They are far from the variables analyzed (see the distance between these renting behaviors and beliefs of consumer effectiveness and personal norms); therefore, none of these factors show associations with such behaviors. Thus, the results suggest that, although there are consumers who exhibit rental behaviors that favor circularity, it is not yet clear whether these consumers have values, beliefs, and norms favoring environmental care. Establishing an association between these behaviors and sociodemographic factors is also challenging.

Other Consumers with Specific Pro-Circular Behaviors

As Figure 9 shows, those consumers with specific pro-circular behaviors also are associated with the intrinsic variables which already described the grouped consumers (i.e., personal norms and specific PCE). For example, consumers who repair and resell their means of transportation (car/motorcycle) believe that they can contribute to solving environmental problems (specific PCE) when performing these behaviors. Similarly, consumers who rent a car instead of buying one of their own believe in the effectiveness of their actions. Specifically, they are convinced that by renting rather than buying a car, they may help with environmental issues (specific PCE). These consumers are also characterized by agreeing to go the extra mile to perform this behavior and would feel guilty if they did not. In other words, they are consumers who evidence personal norms regarding the car. Finally, consumers who return their clothing to be recycled also claim to have perceived effectiveness regarding recycling clothes. Thus, although returning clothing to a recollection point to be later recycled may imply an additional effort, they do it because they consider this concrete action beneficial and relevant to the environment. (See the proximity between each specific behavior and being ‘Strongly agree’ with personal norms or specific perceived effectiveness.) As with other profiles, no association was found between these particular behaviors and sociodemographic variables.

5. Discussion

This research aimed to identify whether there exist consumers with pro-circular behaviors and characterize them based on the components of the Value-Belief-Norm (VBN) theory, extending the category of beliefs by including perceived consumer effectiveness (PCE) in it. Few studies in the field of the circular economy have focused on understanding the consumer; to our knowledge, none have approached this topic from this framework (i.e., VBN theory), which is widely used in the analysis of pro-environmental behaviors. We used cluster and multiple correspondence analyses to group and profile consumers based on the frequency of their behaviors and their association with the components of VBN theory.

The descriptive and associative results show that it is impossible to speak of a consumer who frequently behaves in favor of circularity at all stages of consumption. However, it was possible to identify some pro-circular consumer profiles and characterize them based on some components of the VBN theory. As one of the few studies that focus on the consumer as an essential actor in circular economy, these exploratory findings can be valuable as a starting point to understand consumers who behave in a pro-circular way and guide future research and organizational strategies toward circular economy models.

Initially, through a cluster analysis, we grouped consumers into four profiles based on the frequency of their behaviors. We identified people who never act in favor of circularity and clearly do not claim pro-environmental values, beliefs, or norms. Moreover, we found less radical consumers who rarely or occasionally behave in pro-circular ways. Although they show green consumption values and consciousness of environmental challenges, they do not feel responsible, guilty, or empowered to act through pro-circular behaviors. This finding is consistent with the literature on the ‘green behavioral gap’, which asserts that being aware of environmental issues and having an environmental attitude is not enough to act sustainably [29,86].

Conversely, people on the path towards circularity often behave accordingly, translating their environmental consciousness into reusing, repairing, reselling, and recycling behaviors. They probably often act this way as they go beyond environmental awareness to specific feelings of moral obligation and effectiveness of their action. Likewise, those established as pro-circular consumers (i.e., who always behave pro-circularly) not only have values, awareness, and beliefs about environmental issues, but they also have strong beliefs about their responsibility, duty, and capacity to act to cope with those problems.

We found it worth deepening this last group. Hence, a multiple correspondence analysis was used to establish the association between behaviors as a first step to identifying types of pro-circular consumers. Then we characterized them by establishing associations between such behaviors and the variables of interest.

Based on previous literature, this study presented a first proposal to classify pro-circular consumers by using the stages of consumption, analyzing behaviors that, by definition, fall in the product acquisition, use, and final disposal stages [15,22]. Our results contrast with this classification, showing that pro-circular behaviors of renting, reselling, reusing, repairing, and recycling do not tend to be classified under this criterion and follow different grouping parameters instead, even displaying behaviors that are unique and not associated with any other.

With regard to the grouping of behaviors, the most obvious case was the one that allowed defining consumers who repair, reuse, and resell, which coincides with previous literature that has suggested that the extension of the useful life of products is one of the crucial aspects of behavior within a circular economy [39]. Here, reusing and repairing are the preferred options [15] because these behaviors can help reduce the environmental impact at all stages of the value chain [87].

On the other hand, the findings related to behaviors that were not associated with any other—such as returning a product (clothing) for proper recycling or renting a car/motorcycle to avoid purchase—are consistent with the statement of other authors that each pro-environmental behavior is different. Therefore, consumers cannot be expected to always behave consistently in environmental terms (e.g., that consumers recycle while making sustainable purchases and reducing their water consumption) [22]. The results suggest that this characteristic also applies to behaviors that favor circular economy models and initiatives; thus, not all pro-circular behaviors should be analyzed as a whole, but rather it is essential to understand and recognize their particularities.

Contrary to expectations, neither values nor beliefs about awareness or responsibility or those associated with the power of nature in the new ecological paradigm (NEP) showed association with the behaviors studied. Therefore, none of these variables allowed for characterizing the identified pro-circular consumer profiles. This contrasts with previous literature on pro-environmental behaviors, which has pointed out the importance of values as antecedents of pro-environmental action [88]. Similarly, several authors have pointed out beliefs about awareness as an essential factor in undertaking sustainable behaviors [24,46,89]. However, consumers who engage in pro-circular behavior do not appear to do so because they believe that human activity has dire consequences on the environment. At least, this is what the results of this research suggest, coinciding with previous studies on some pro-circular (e.g., acceptance of remanufactured products) and some pro-environmental behaviors (e.g., recycling and composting) [20,90,91,92].

Most of the identified pro-circular consumer profiles were characterized based on moral norms and PCE. Although this study relied on associations between the variables to establish such characterization and not on the dependency relationships between the variables of interest, the results are close to and somewhat agree with previous research that has highlighted the role of moral norms in explaining pro-environmental behavior [24,26]. In fact, previous literature has highlighted personal norms as one of the most significant factors in adopting a pro-environmental behavior that, by definition, is associated with circularity, such as recycling [93,94]. Similarly, this research suggests that consumers who behave in a pro-circular manner are those who believe in the effectiveness of their individual actions—specifically, their pro-circular actions—to solve environmental problems. These findings are in line with different authors who have underlined the importance of an individual’s belief about the usefulness of pro-environmental actions to motivate and favor the adoption of sustainable behaviors [49,65,66,67,68,95,96,97].

This research does not offer a conclusive answer about a single pro-circular consumer profile. The fact of reaching different possible profiles with diverse behaviors toward different product categories suggests that future analyses should be oriented to groups of behaviors with some similarity or to specific actions for specific products.

In this case, we used three product categories of high environmental impact that, in addition, had high importance and perceived value for consumers (i.e., clothing, car/motorcycle, and furniture), whose only purpose was to operationalize different behaviors. Future research could consider these and other products to analyze their role in adopting pro-circular behaviors.

The results obtained through MCA showed that it was challenging to establish associations between pro-circular behaviors and sociodemographic variables. These findings are consistent with previous literature that has suggested the low predictive power of sociodemographic factors to explain and predict pro-environmental behaviors, for example, recycling [93,94,98,99]. However, considering a large number of studies that started to focus on the influence of sociodemographic factors on pro-environmental behavior decades ago [45,100,101], as well as the results of previous research that have shown how sociodemographic variables can significantly determine consumer behaviors related to circular economy [21], it is worth developing future studies, of a correlational type, using other methods of analysis, to confirm the role of sociodemographic variables in different behaviors associated with circular economy. This is because it has been suggested that the way in which these factors influence behaviors varies according to specific aspects of the behavior and the category of products under study [21].

This research has examined some representative behaviors from each consumption stage based on previous literature to identify pro-circular consumer profiles [22,69]. Nevertheless, the analysis of other product categories might lead to a range of possibilities that allow expanding the profiles determined in this research. Similarly, this research has addressed the characterization of pro-circular consumers based on the components of the VBN theory as a reference to study sustainable pro-environmental behaviors. Other factors that come from diverse behavioral theories can also be considered for future research, which would be beneficial for the profiling of pro-circular consumers.

Although this study has mainly focused on the association between variables to identify and characterize pro-circular consumers, the findings open the way for future research to address a correlational perspective in which intrinsic factors, such as moral norms and PCE, can be analyzed in relation to specific pro-circular behaviors aimed at the extension of product life (e.g., reusing, repairing, and reselling) and new circular business models (e.g., rental and remanufacturing).

6. Conclusions

6.1. Management Implications

Transitioning to a circular economy is impossible without a change in consumer behaviors that would enable the embrace of new practices, such as returning a product for recycling, accepting other types of products (i.e., remanufactured and repaired ones), and adapting to new business models (i.e., renting and reselling) [5]. The first step to motivate such behavioral change is to identify what consumers already do and understand what factors characterize such behaviors. Although exploratory, the results suggest some managerial implications to promote business models toward a circular economy. The managerial implications are detailed below.

The most evident consumer profile was the one who performs behaviors oriented to the extension of the useful life of products, which might be a first niche on which organizations can focus their actions. Thus, a first effort could be generating new consumption patterns, mainly at the product-use stage in which reuse, repairing, and reselling are the keys to promoting circularity.

This study showed that some consumers engage in rental activities, seeking to avoid buying new products (e.g., clothing and car/motorcycle) or using private means of transportation. This suggests that, in addition to efforts that promote the extension of the useful life of products, it is also possible to develop strategies and initiatives to promote alternative business models (e.g., renting) that have been highlighted as part of circular business models [15].

Clothing is a product category that has gained relevance in environmental issues, rejecting models of mass production and consumption [102] that favor constant changes in fashion, at low prices, and high turnover in purchase, which leads to rapid disposal (fast fashion) [103]. Instead, it encourages sustainable models that, in addition to production with less negative impact on the environment, extend the duration of products and promote their reuse and repairing before discarding them [104]. This research sheds light on an additional behavior that consumers can also perform regarding clothing: returning them to the point of sale for proper recycling. Currently, there are different initiatives worldwide that promote the extension of the useful life of clothes (e.g., Patagonia, Eileen Fisher, Filson, REI, and Fjällräven) [20]. However, only a few brands promote the consumer behavior of returning for recycling (e.g., H&M). Therefore, more initiatives are required with the respective logistics to facilitate this pro-circular behavior that allows initiating circular models in the textile industry.

The characterization of the profiles obtained reflected the relevance of psychographic variables centered on the perceptions and feelings of individuals about their duty and effectiveness to act in a pro-circular way. Thus, this study gives indications about the role of intrinsic motivation aspects (e.g., moral norms and perception of effectiveness of individual actions) in this type of behaviors, which could be considered when designing strategies to promote the change of behaviors and routines from linear to circular business models. In fact, the descriptive results of this research showed a high positive correlation between these variables, which suggests that working together can be an efficient way to promote circularity.

Some public and private initiatives tend to work with extrinsic variables (e.g., incentives, regulations) to promote pro-environmental actions. However, the findings of this research suggest that those who display today pro-circular behaviors are the same ones who believe that their actions count and are necessary and effective in meeting the environmental demands of the planet. Therefore, future initiatives could take advantage of these variables to design more comprehensive messages and interventions in favor of circularity.

6.2. Limitations

With an exploratory approach, this study analyzed seven neighborhoods in the city of Medellín in Colombia. Although this city was chosen due to its initiatives and management in terms of environmental sustainability, which makes it a reference point to study the profile of pro-circular consumers, the results can only be inferred for this population. It is necessary to develop further studies to expand the scope of these findings, at least at the national level.

This research has focused on identifying and characterizing a type of consumer but not on predicting pro-circular behaviors. Thus, although the study used the components of the VBN theory to develop the characterizations of the identified pro-circular consumers, the research was not based on the model proposed by the authors of this theory [24,105]. Thus, the findings do not contradict or support what has been found in previous literature that applied the VBN theory to the explanation of sustainable behaviors. Given the relevance of this theoretical approach [26], it is necessary to take a step forward from identification and characterization to propose new analyses to understand the application of the model in adopting pro-circular behaviors.

Although this study used three product categories to operationalize pro-circular behaviors, the analyses carried out did not focus on the products themselves, nor they highlighted the role of products in the studied behaviors. New studies could have a broader purpose and analyze the role of different products in pro-circular behaviors. This approach certainly deserves greater attention.