ORIGINAL RESEARCH published: 23 December 2020 doi: 10.3389/fpsyg.2020.580730 Psychological Factors Influencing Pro-environmental Behavior in Developing Countries: Evidence From Colombian and Nicaraguan Students Manuel Francisco Díaz 1, Andrés Charry 2, Stefania Sellitti 3, Matteo Ruzzante 4, Karen Enciso 1 and Stefan Burkart 1* 1 Tropical Forages Program, International Center for Tropical Agriculture (CIAT), Cali, Colombia, 2 Food Environment and Consumer Behavior Research Lever, International Center for Tropical Agriculture (CIAT), Cali, Colombia, 3 NOVA School of Business and Economics, Lisbon, Portugal, 4 Development Impact Evaluation, World Bank, Washington, DC, United States Identifying the determinants of human behavior is useful to adjust interventions and lead the civil society toward a stronger commitment to climate change (CC) mitigation and adaptation objectives, achieving greater support for successfully implementing environmental policies. Existing research has largely focused on case studies of pro-environmental Edited by: Jose Q. Pinheiro, behaviors (PEBs) in developed economies but there is very little evidence for developing Federal University of Rio Grande do countries. This study provides estimations of the effect of internal factors, such as Norte, Brazil sociodemographic variables, and four psychological dimensions (CC knowledge, Reviewed by: environmental attitudes, self-efficacy, and trust in sources of environmental information) on César O. Tapia-Fonllem, University of Sonora, Mexico PEBs. Data were obtained through a survey applied with future decision makers – university Elisa Menardo, students – from Colombia (n = 4,769) and Nicaragua (n = 2,354). Indices were generated University of Verona, Italy for PEBs and the psychological dimensions using z-scores and Principal Component *Correspondence: Stefan Burkart Analysis (PCA). Partial correlations were evaluated through the Ordinary Least Squares s.burkart@cgiar.org (OLS) method. Our results suggest that, in order to reach the planned emission reduction targets, policy approaches should more strongly focus on educating and motivating citizens Specialty section: This article was submitted to and prepare them for contributing to the environmental cause, as well as provide individual Environmental Psychology, solutions to combat CC, rather than providing only information on its causes and consequences. a section of the journal Frontiers in Psychology Keywords: awareness of sustainability, education, psychological adaptation, environmental attitude, policy Received: 05 August 2020 support Accepted: 27 November 2020 Published: 23 December 2020 Citation: INTRODUCTION Díaz MF, Charry A, Sellitti S, Ruzzante M, Enciso K and As part of the commitment with the Sustainable Development Goals and the Paris Agreements, Burkart S (2020) Psychological developing countries have been increasing their responses to climate change (CC), especially Factors Influencing Pro-environmental since evidence suggests that the impacts of CC would have larger impacts in the global south, Behavior in Developing Countries: strengthening structural inequalities and leading to a vicious circle (Burke et  al., 2015; Evidence From Colombian and Nicaraguan Students. United Nations, Department of Economic and Social Affairs, 2016). This panorama has led Front. Psychol. 11:580730. to a change in the development model of nations with a deliberate direction toward sustainability doi: 10.3389/fpsyg.2020.580730 (Bárcena et al., 2018; Intergovernmental Panel on Climate Change, 2018), resulting in programs Frontiers in Psychology | www.frontiersin.org 1 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior and strategies for CC mitigation and adaptation, which demand disseminating information, and provides guidance on the type context specific approaches. However, the success of such of knowledge that should circulate in order to improve the approaches depends largely on the social norms, preferences, efficacy of both public and private communication strategies. beliefs and values of the targeted individuals (Adger et  al., In addition, policy approaches will be  more effective when 2009). Factors such as public awareness and knowledge of CC, taking into account the psychosocial context and factors that attitudes and opinions regarding environmental problems, and influence environmental actions (Steg and Vlek, 2009; Stern, knowledge about appropriate behaviors determine the public 2011), considering that students not only increase their own support or opposition of environmental or CC policies, strategies, contribution to CC mitigation, but also their empowerment and initiatives (Arcury, 1990; Leiserowitz, 2006; Lorenzoni et al., to become change agents and influencers for other segments 2007; Howe et  al., 2015). Lee et  al. (2015) argue that some of the population (González and Maldonado, 2015). countries are more advanced than others in terms of executing environmental policies resulting from differences in risk perception of the targeted populations. THEORETICAL FRAMEWORK During the last 4 decades, important advances were made in the identification of factors influencing environmental The Theory of Planned Behavior (TPB) as a perceptions and Pro-environmental Behaviors (PEBs). Guiding Principle Nevertheless, most of these studies were conducted in North Theory of Planned Behavior (TPB; Ajzen, 1985, 1991) is America, Europe, and other developed regions (Lorenzoni and considered an extension of the Theory of Reasoned Action Pidgeon, 2006; Cordano et al., 2010; Vignola et al., 2013; Salehi (TRA) (Ajzen and Fishbein, 1980), which explains behaviors et  al., 2016). While research on PEBs has been growing in under a logical framework: behavioral beliefs are supported Latin American countries recently (Padilla y Sotelo and Luna, by a favorable or unfavorable attitude about a certain behavior. 2003; Pato et  al., 2005; Bertoni and López, 2010; Calixto and Normative beliefs refer to the subjective norm and thus the Herrera, 2010; Barazarte et  al., 2014; Sánchez et  al., 2014; social pressure associated with behavior. In this sense, actions González and Maldonado, 2015; Ideam et al., 2016; Pávez-Soto are supported by individual attitudes, available information, et  al., 2016). These studies have covered different groups such and subjective norms, which are based on beliefs formed as students (e.g., Tikka et  al., 2000; Spellman et  al., 2003; through knowledge, understood as the element that allows Palavecinos et  al., 2016; Salehi et  al., 2016), consumers (e.g., evaluating the consequences of actions. TPB introduces an Tobler et  al., 2012a,b; Yadav and Pathak, 2016), citizens with additional element: the control of perceived behaviors. This diverse political and religious positions (e.g., Arbuthnot, 1977; element refers to the understanding of the factors that can Tobler et  al., 2012b), professors (e.g., Pe’er et  al., 2007) and hinder the performance of actions and the subsequent behavior communities related to recycling (e.g., Sidique et  al., 2010). derived from them (Ajzen, 1991). The theory suggests that This study aims at enhancing the knowledge base for the people are much more likely to adopt a certain behavior when Latin American context by evaluating perceptions and behavior they feel able to perform it successfully, a dimension also toward CC with a large sample of university students in affected by self-efficacy (Bandura, 1977). This concept refers Colombia and Nicaragua. Knowledge and attitudes about CC, to “people’s beliefs about their capabilities to exercise control self-efficacy, and trust in different information sources are over their own level of functioning and over events that affect measured, and relationships within PEBs, knowledge, attitudes, their lives” (Bandura, 1991, p.  257). and socioeconomic characteristics of the selected population While TRA and TPB models do not include sociodemographic are explored. variables, the authors do not deny their importance. Rather The applied approach leads to two questions: (1) Why choose the opposite: they argue that any external variable can influence this segment of the population as study group? and (2) Why the intention – and indirectly, the actual behavior – if it identifying relationships among the variables? According to influences the attitudinal and/or the normative component. Bradley et  al. (1999), university students – future scientists, Although some studies have not taken into account legislators, consumers, and voters – will be  responsible for sociodemographic variables to relate PEBs with the theories generating solutions to environmental problems, and thus should presented above (e.g., Bang et  al., 2000; Mishra et  al., 2014; be persuaded to adopt and pay the costs of future environmental Paul et al., 2016; Beckage et al., 2018), others did (e.g., Goldenhar policies. Likewise, students will have to make complex political and Connell, 1993; Kim et al., 2013; Paço and Lavrador, 2017). decisions about CC mitigation and should do so from an The latter shows that the relationship between knowledge, informed perspective. Consequently, current and future educators attitudes, behaviors, and external variables differ among contexts require a better understanding of the dimensions affecting the change over time and are perceived differently from one culture students’ perceptions in order to develop teaching programs to another. This highlights the importance of combining that contribute in a more effective way to the fight against sociodemographic and cognitive factors to study PEBs in CC (Wachholz et  al., 2014). Next, identifying the relationships context-specific cases. between PEBs and the variables that affect them provides a Various studies have suggested that the general framework clearer landscape to define strategies and prioritize efforts for of TPB could be  enriched and broadened by adding new increasing the level of environmental awareness. Accordingly, constructs or altering the pattern of variables contemplated in this research allows identifying the most reliable agents in the TRA and TPB (e.g., Perugini and Bagozzi, 2001; Moons Frontiers in Psychology | www.frontiersin.org 2 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior and De Pelsmacker, 2015; Yadav and Pathak, 2016; Wan et  al., actions (Bord et  al., 1998; O’Connor et  al., 1999; Gifford and 2017). This is common practice: for example, in a meta-analysis Comeau, 2011; Perez et al., 2015; Ideam et al., 2016; Palavecinos on the application of the TPB to examine environmental et al., 2016; Paço and Lavrador, 2017). Among younger people, behaviors, Yuriev et  al. (2020) found that 72% of the analyzed women have also shown better environmental attitudes and studies used an extended version of the TPB. By including knowledge, are more concerned about environmental problems, these constructs, these studies added factors that increase the and are more involved in CC mitigation actions (e.g., predicting power of the model and may account for observed Freudenburg and Davidson, 2007; McCright, 2010). differences between groups, as they take into account specific contextual and idiosyncratic factors that can influence behavior. Age Similar to these approaches, the present study is inspired Children and elderly experience more aggressively the effects by but extends on the TPB, to the extent that it involves of CC. Both populations present higher mortality and disease dimensions different to the ones described above, such as rates due to hurricanes, floods, and droughts (Intergovernmental sociodemographic characteristics, knowledge of CC, and self- Panel on Climate Change, 2014b). Though Otto and Kaiser efficacy. Additionally, the dimension “trust in sources of (2014) found that older people have better PEBs than younger environmental information” was included with the aim of individuals, the effect of age is ambiguous and appears to exploring both the individual and broader explanatory factors be  affected by access to information. Furlow and Knott (2014) and thus improve the predictive power of the framework and and Bedard and Tolmie (2018) argue on the importance of identify sources of variation between such stated behaviors. the internet and digital communications for younger generations, The next section presents the selected dimensions and their who tend to be  better informed and more concerned about importance for the present study. social and environmental issues. As a result, younger individuals have more tools to understand CC and consequently generate Explanatory Variables environmental actions. Similar studies have addressed topics such as energy saving (e.g., Sapci and Considine, 2014), recycling (e.g., Goldenhar Geography and Connell, 1993; Sidique et  al., 2010; Paço and Lavrador, Similar demographic and cultural characteristics tend to cluster 2017) or the willingness to pay for environmentally friendly (Leiserowitz, 2006; Motyl et  al., 2014; Howe et  al., 2015). products (e.g., Furlow and Knott, 2014; Paul et al., 2016; Bedard Likewise, perceptions of CC exhibit geographic patterns due and Tolmie, 2018), using a broad set of possible explanatory to differences in experiences with extreme weather events and variables. These range from psychological dimensions (Arbuthnot, climate variability (Akerlof et  al., 2013; Howe et  al., 2015). In 1977; Helm et al., 2018), to the orientation of messages (Gifford their analysis in 89 countries, Howe et  al. (2013) found that and Comeau, 2011), geographical variation (Howe et al., 2015), people living in places more susceptible to CC are the most or a combination of both (Zhang et  al., 2018). concerned about the phenomenon. For this study, socioeconomic characteristics and four dimensions were determined for their potential explanatory Field of Study relationship with PEBs. PEBs can be  understood in two ways: Students of certain academic fields show a better understanding first, as behavior that “harms the environment as little as possible, of CC. Several authors (e.g., Tikka et  al., 2000; Spellman et  al., or even benefits the environment” (Steg and Vlek, 2009, p.  309) 2003; Pe’er et  al., 2007; Salehi et  al., 2016) found that students and second, as behavior “that is undertaken with the intention from disciplines related to environmental and natural sciences to change (normally, to benefit) the environment” (Stern, 2000, possess a significantly higher level of environmental knowledge p. 408). In order to simultaneously identify factors that influence and attitudes than those from other programs. both future intentions and current PEBs undertaken by the population of interest, this construct includes 14 statements referring to both declared behaviors and intentions to conduct PEBs. Education Level and Academic Cycle Educational achievements are the strongest predictor for environmental knowledge and understanding of CC (Polonsky Sociodemographic Variables et  al., 2011; Lee et  al., 2015). According to Meyer (2015), Gender education can lead people to care more about general social Women and men do not experience CC in the same way. welfare, including the external benefits of their actions. Literature shows that women, particularly in rural areas, Furthermore, the time spent at university can have a positive present greater concerns about CC since they carry out impact on individuals, since higher education institutions tend activities such as raising children, or planting and harvesting, to encourage students to incorporate principles of environmental which depend largely on both natural resources and a healthy responsibility (Kagawa, 2007; Emanuel and Adams, 2011). Spellman environment (Blocker and Eckberg, 1989; Davidson and et al. (2003), Meyer (2016), and Paço and Lavrador (2017) have Freudenberg, 1996; Food and Agriculture Organization of found significant differences between students of higher semesters the United Nations, 2015; Vicente-Molina et al., 2018). In that and those who have recently started their studies. However, sense, women are strongly affected by changes in the other authors did not find significant relationships between environment and show to be  more committed to mitigating knowledge of CC and the academic cycle (e.g., Salehi et al., 2016). Frontiers in Psychology | www.frontiersin.org 3 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior Income Level/Socioeconomic Strata but also on the level of trust in different sources (Dietz et  al., People with less resources are the most affected by CC despite 2007). Lorenzoni et  al. (2007) describe that public distrust in not being the main emitters of greenhouse gas (GHG; media constitutes an important impediment to CC adaptation. Mendelsohn et  al., 2006; Intergovernmental Panel on Climate According to them, media tactics such as exaggeration, Change, 2014a; Hallegatte and Rozenberg, 2017; International sensationalism, or partiality (in addition to contradictory frames) Monetary Fund, 2017; Bárcena et al., 2018). Low-income people end up generating confusion. Gifford and Comeau (2011) found are often located in places more vulnerable to climatic phenomena that the orientation of messages influences both the commitment and experience higher levels of worries and a greater sense toward mitigation and the intentions of the behavior. Kellstedt of insecurity. They are the ones who know more about the et al. (2008) state that trust in Non-Governmental Organizations effects of CC, but lack an adequate understanding of the causes (NGOs), media, or political institutions can facilitate or obstruct as well as strategies for coping with the consequences (Hardoy the understanding of CC. Spence and Pidgeon (2010) found and Pandiella, 2009), and therefore the relationship with PEBs that the effectiveness of the messages also varies according to may be  ambiguous. the geographic location where the information circulates, with different effects on the recipient’s behavior. Dimensions of Reference CC Knowledge Environmental Attitude A higher knowledge about CC leads to increasing concerns The inclusion of attitudes as an explanatory factor of behavior about it. Consequently, informed citizens are more likely to is the most adopted approach (Li et  al., 2019). Authors such perform actions that promote environmental protection and as Arcury (1990) and Kaiser et  al. (1999) affirm that there support related policies (Ramsey and Rickson, 1976; O’Connor exists a link between knowledge and PEBs, and that they are et  al., 1999; Bord et  al., 2000; Kellstedt et  al., 2008; Shi et  al., generally connected by attitudes. To measure this dimension, 2016). However, CC is a complex phenomenon that encompasses the present study adopts the New Environmental Paradigm multiple causes and a great variety of consequences. Various (NEP) scale. Dunlap and Van Liere (1978) propose NEP to authors found that, in order to properly face CC, knowledge respond, not only to different development theories, but also about the anthropogenic causes of CC might be  more relevant to a new way of understanding the relationships between human than, for example, knowledge about its physical effects (Bord beings and their environment, which translates into a radical et  al., 2000; Lee et  al., 2015; Shi et  al., 2016). Other studies change in attitudes. By covering different environmental issues (e.g., Tobler et  al., 2012a; Salehi et  al., 2016) present the need with relative standardization, the NEP has become a conventional to know and differentiate the causes and consequences of CC scale to capture this information. as well as the knowledge of concrete actions to mitigate it. However, Shi et  al. (2016) state that it is essential to focus studies on all dimensions since measuring the perceptions of MATERIALS AND METHODS the phenomenon transversally becomes necessary. Participants Higher education students from the main cities of Colombia Self-Efficacy and Nicaragua were selected as the objective population of This dimension is of special relevance for the TPB since it this study. A sampling frame was developed using a list of contributes to the determination of perceived behavioral control universities from four large cities in Nicaragua and 10  in and thus to PEBs. Expectations such as motivation, performance, Colombia, respectively. Simple random sampling was carried and feelings of frustration determine behavioral reactions. Some out to select the universities, and the survey was then applied studies demonstrate the perceived efficacy of individual actions to students belonging to the faculties where authorization was in the fight against CC, showing how deeply they influence granted. Total number of valid surveys obtained was 7,123: PEBs and environmental knowledge (Heath and Gifford, 2006; 2354  in Nicaragua and 4,769  in Colombia. Respondents of Kellstedt et  al., 2008). Not only that, high self-efficacy can the city of Managua concentrated 59.2% of the Nicaraguan influence the transition from easy-to-perform PEBs to those sample, whereas in Colombia, the cities with higher participation with a greater degree of difficulty (Lauren et  al., 2016). were Cali (26.1%), Medellín (18%) and Bogotá (15.4%; Figure 1). Table  1 shows the most relevant educational and socio- Trust in Sources of Environmental Information demographic characteristics of the sample. Discourses projected by the media contribute to the evaluation Both the survey and methodology were approved by the and social interpretation of a problem (like CC). By this, media International Center for Tropical Agriculture’s ethics committee. help transferring certain elements of science into common Before the administration of the survey, the students were culture (Pinheiro and Farias, 2015). Although neither TRA provided with information on the objectives of the study, data nor TPB consider this dimension within their models, it is privacy and management, and their rights as participants, after necessary to bear in mind both the representations and the which an informed consent was confirmed. This consent social interactions through which common culture is constructed highlighted three aspects: (1) the use of data for academic and shared (Meira-Cartea et al., 2018). Not only that, decision- research purposes only and protection of the participant’s making is dependent not only on the availability of information identity, (2) the voluntary character of participation and Frontiers in Psychology | www.frontiersin.org 4 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior FIGURE 1 | Geographical distribution of the study population for Nicaragua (left) and Colombia (right). opportunity for withdrawal at any time, and (3) the possibility household is located), which normally is used to relate to the for participants to request a copy of the results after analysis. level of income and socioeconomic conditions of an individual. In both countries, the survey was mainly conducted at university Instruments classrooms through self-administered questionnaires. In In addition to sociodemographic information, data to measure Colombia, some students participated through an online survey, different dimensions were obtained using a five-point Likert- which was sent to their institutional emails by the respective type survey. The instrument included five modules with a series university authorities. of questions and statements related to the previously described dimensions, presenting 48 statements. To verify the internal Data Analysis consistency of the instrument, a pilot test was conducted with Exploration of Data 100 students from the city of Cali (Colombia) and a Cronbach’s An index was calculated for each of the modules that made alpha coefficient was calculated for each variable (Table  2). up the questionnaire. These indices were obtained by granting For data collection, Information on income was captured five points for answers that were (a) in accordance with the differently in both countries: in Nicaragua, the level of income dimension CC knowledge or (b) consistent with the dimensions was measured in monetary terms. In Colombia, the measurement environmental attitude (NEP), self-efficacy, trust in sources of was in accordance with the national socioeconomic stratification environmental information and PEBs. This score decreases as categories,1 ranging from one to six (the higher the number, the selection moves away from the desired response. However, the better the living conditions in the place where the respondent’s there was no discount if the answer was wrong. In this way, maximum and minimum scores were established, and intervals 1S ocioeconomic stratification in Colombia categorizes housing units in a scale were created with the aim of classifying all the variables within of one to six according to their physical characteristics, immediate surroundings the established ranges (Figures  2, 3). In the following section, and rural or urban context. It was implemented for assigning differential public the mean values and the observed deviations for each variable utilities rates to different strata, enabling higher strata (five and six) to pay are presented using as reference the values presented in Table 2. higher rates and subsidize the costs for lower strata (one, two and three) (Departamento Administrativo Nacional de Estadística, Colombia). Retrieved The effects of sociodemographic variables on the reference from: https://www.dane.gov.co/index.php/servicios-al-ciudadano/servicios- dimensions were analyzed through tests of mean differences. informacion/estratificacion-socioeconomica As a first measure, the normality of the dimensions was analyzed Frontiers in Psychology | www.frontiersin.org 5 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior through the Shapiro-Wilk test. Next, a Student’s t-test, ANOVA Table  1 also shows statistical differences between both and a Kruskal-Wallis one-way ANOVA were conducted. For countries for the sociodemographic variables. Section the latter, post-hoc tests were also carried out, such as Tukey “Exploratory analysis Colombia” highlights significant differences and post-hoc Kruskal-Wallis Dunn, in order to identify which among the two countries regarding the analyzed dimensions. group showed the largest differences. Regression Model: Ordinary Least Squares (OLS) TABLE 1 | Educational and sociodemographic characteristics of the study and Principal Component Analysis (PCA) population in Nicaragua and Colombia. Pro-environmental Behaviors (PEBs) were evaluated through the OLS method. As a first measure, the variables were Nicaragua (%) Colombia (%) standardized and five z-score indices were created (Eq.  1), (n = 2,354) (n = 4,769) which were defined as the weighted average of the z-score of their variables, following the methodology proposed by Kling Age (average in years)*** 19.6 (σ = 2.62) 21.3 (σ = 4) and Liebman (2004) and Kling et al. (2007). Next, the following Gendera regression was made: Female 44.3 49.5 Male 54.4 50.1 yic =αic +β×Knowledge ′ic + Xic + εic (1) Income level (Nic – US$)/socioeconomic strata (Col)a ≤ US 250/1 31.5 13.9 Where yic is the result of interest (Pro-environmental 251–500/2 35.6 31 Behaviors), i and c are identifiers for the individuals and the 501–750/3 7.6 33.9 ¢ 751–1,000/4 3.7 10.7 country. Xic is the vector of sociodemographic and educational 1,001–1,250/5 6.7 4.9 characteristics, and eic is the standard error. β is the coefficient ≥1,251/6 11.7 2.1 that measures the effect of knowledge about CC on the PEBs Education level*** of students. In the absence of identifiable exogenous shocks Candidate undergraduate or other means to establish causality, estimates should degree (BSc) 99.6 97.3 Candidate postgraduate be  interpreted as (partial) correlations. degree (MSc, PhD) 0.4 2.7 As a way to mitigate omitted variable biases, the variation Academic cyclea*** of the β coefficient was observed when adding the covariates First year 35.9 31.7 to the initial regression. Five different specifications were used Second year 20.7 19.3 Third year 17 16.2 in two phases. First, a regression of the result of interest on Fourth year 14.5 18.4 knowledge. Second, the inclusion of the sociodemographic Fifth year 11.9 13.8 characteristics of the individuals. Third, in addition to the previous Field of study models, the educational characteristics of the students were Agricultural sciences 44.4 32.6 included. Fourth, in the second phase, a regression of student Engineering 30.5 24.7 Natural sciences 1.1 11.1 behavior was performed in all z-score indices. Fifth, all the indices Health sciences 10 2.3 and sociodemographic characteristics of the students as well as Administrative sciences 7.8 7.8 the educational variables were included in the model (Eq.  2): Othersb 6.2 21.5 yic =α+β1×Knowledge+β2×Efficacy+ aNot all percentages add up to 100%. Some participants did not provide all the β3×Trust+β4×NEP+ X ′icγ+ ε (2) requested information. bOthers: Humanities, Laws, Basic sciences, Arts, pedagogy and comunications. Here, α is the constant parameter and β1, β2, β3, and β4 ***p < 0.001. are the coefficients for each of the z-scores. g is the vector of coefficients for the matrix of sociodemographic and educational variables, and ε is the standard error of the model. The PCA TABLE 2 | Dimensions of the survey. method was used as an additional robustness test. This statistical procedure uses an orthogonal transformation to convert a set Dimension Selected references Selected Total Cronbach’s of observations of possibly correlated variables into a set of items values alpha (α) values of variables that are not linearly correlated. Thus, the pilot test same indices used in the OLS method were recreated with CC knowledge Spellman et al. (2003) 16 80 0.64 the PCA and variables were regressed on the PCA indices. Self-efficacy Kellstedt et al. (2008) 4 20 0.62 Trust in sources of environmental Kellstedt et al. (2008) 6 30 0.65 information RESULTS Pro- Markle (2013), Paço and environmental 14 70 0.73 Lavrador (2017) Exploratory Analysis Nicaragua behaviors New The majority of the population (71.6%) possess moderate Dunlap and Van Liere Environmental 8 40 0.71 knowledge about CC (  =  51.5; σ  =  9.1) while only a small (1978) Paradigm (NEP) portion of the sample possess high CC knowledge (16.5%). Frontiers in Psychology | www.frontiersin.org 6 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior FIGURE 2 | Knowledge about climate change (left) and PEBs (right) by cities in Nicaragua. FIGURE 3 | Knowledge about climate change (left) and pro-environmental behaviors (right) by cities in Colombia. As shown in Figure  2, the values obtained are slightly higher Regarding CC knowledge, data reveal that the vast majority for the country’s capital, Managua, which, in turn, also has the of students are aware of the responsibility of humans in lowest proportion of the population unaware of the causes and global warming, and consider that individual actions can effects of CC. Although the mean differences in knowledge scores have an influence on global warming. Despite this awareness, were low, they were significant between Managua and León important gaps were found. The majority of students consider (p < 0.01) and Managua and Matagalpa (p < 0.01). Men possess that nuclear energy contributes to CC (false), that global significantly higher CC knowledge (p < 0.01), and no significant warming does not affect agricultural activities such as differences were found for both income and education level. agriculture and fishing (false), and that the industry sector Regarding the field of study, engineering students show produces the largest amount of GHG emissions (false). About better knowledge than those studying agricultural sciences, law, half of the sample is unaware of the importance of economics, pedagogy, agricultural sciences, and basic sciences clouds and water vapor in the atmosphere – in fact, the (p  <  0.05). For their part, agricultural science students present majority of the sample affirmed that without clouds, the better knowledge than students from administrative sciences, earth would not be  in danger (false). Despite this, a correct basic sciences and humanities (p  <  0.001). understanding was observed of the problems that ultraviolet Frontiers in Psychology | www.frontiersin.org 7 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior radiation can cause to people’s health and of the consequences Figure  3 shows that the majority of the student population of ozone depletion (Appendix A). (78.3%) is located within the intermediate knowledge and to Regarding self-efficacy, students are, on average, akin to a lesser extent within the high knowledge levels (14.3%). 76.3% of the items (  =  15.4; σ  =  3.4), corresponding to a Regarding gender, the Colombian students show similar results high level of affiliation with a perceived importance of individual as their Nicaraguan peers. However, for Colombia the differences actions in the environment. No significant differences on self- are significant: men possess higher CC knowledge (p  <  0.01), efficacy were found for any of the sociodemographic variables. while women show better environmental behavior (p  <  0.01), Responses of younger people show that they are less prone greater self-efficacy (p < 0.01), and better environmental attitude to carry out active and participatory processes in mitigation (p < 0.01). Regarding CC knowledge, there were no significant and adaptation strategies to CC. In contrast, 88.8% of the differences found between Colombian and Nicaraguan students. surveyed population expressed high levels of awareness about However, at the country level, significant differences were the impact of human actions on CC (Appendix B). observed: engineering students present a higher knowledge In contrast, the results of trust in the institutions that provide level than those studying humanities, economics, basic sciences, environmental information were ambivalent. Although the total administrative sciences, communication, humanities, and trust level was above 50%, this variable was not concentrated agricultural sciences (p  <  0.05). in any extreme ( = 19.5; σ = 4.66). With a shared perspective Bogotá (the capital city) has the most informed student between both genders, the national government presents the population ( = 54.2; σ =7.5), followed closely by Tunja ( = 53; lowest trust levels. Students have more trust in NGOs, educational σ =8), an intermediate city close to Bogotá. Armenia, on the institutions, and the scientific community. The latter exhibits other hand, has the least informed student population ( = 50.2; the highest trust levels (Appendix C). σ =7.5). The latter and Bogotá stand out for presenting more As shown in Figure  2, the students’ scores on PEBs are significant differences compared to the other cities of this mostly favorable (  =  47.7; σ  =  9.8). Engineering students sample. Though students shows a high understanding of the show better PEBs than those studying humanities, administrative impact of global warming on agriculture and fishing, they sciences, education, and pedagogy (p  <  0.05). Students from present a wrong understanding of the contribution of nuclear humanities on the other hand, have lower PEBs levels than energy and the industrial sector to CC. Students also ignore health and agricultural sciences students (p  <  0.05). Several the relationship between ozone and ultraviolet radiation, but discrepancies were identified regarding the students’ real behavior: are aware of the effects that the latter has on people’s health Household waste separation receives a low qualification across (Appendix A). There were no significant mean differences the whole sample. Likewise, about half of the sample prefers observed for the effects of the academic cycle and socioeconomic to use private instead of public transport. It is worth highlighting strata on CC knowledge. that most people tend to pay attention to savings in water Compared to the results for Nicaragua, self-efficacy was and electricity consumption and decide to adopt practices that significantly higher in Colombia (  =  16.3; σ  =  2.9; p  <  0.01), contribute to preventing the wastage of these resources especially in terms of individual actions to reduce global (Appendix D). warming and CC. Although in Colombia both genders show The surveyed population showed affinity with ecological a high degree of awareness about the impact of human actions premises of the NEP scale. On average, 78.2% of the population on the environment, in all items women presented a greater respond in accordance with the overall statements (  =  31.1; degree of affinity toward the statements (Appendix B). σ  =  6). No significant differences were observed when testing With regard to trust in institutions that provide environmental for gender. By field of study, engineering students show better information, the Nicaraguan students show significantly higher behavior for this variable than students from humanities, levels than their Colombian peers (p < 0.001). Although positive administrative sciences, pedagogy, and agricultural sciences responses were observed for Colombia, the means are mainly (p < 0.05). The latter, for their part, perform better than students concentrated within intermediate ranges (  =  19.2; σ  =  4.2). from administrative sciences and pedagogy, but worse than those Perceptions about the institutions were widely divergent. On from health sciences (p < 0.05). When analyzing this dimension average, students rather trust the scientific community, by cities, Managua showed the best results, with significant educational institutions, and NGOs, but are reluctant to the differences to Matagalpa (p  <  0.01) and León (p  <  0.05). It information offered by the government (Appendix C). was evidenced that students assign importance to building a The results for the PEB dimension show that Colombian better balance between humans and nature. However, affinity students have a significantly better behavior than their with the statement “The ultimate goal of plants and animals Nicaraguan peers (p  <  0.001). This dimension is located in must be  to serve the needs of the population”, presented an the upper part of the mean range (  =  50.6; σ  =  9). An important variability in the distribution of the degree of affinity, interesting aspect is that economics students present the lowest and it is observed that the response rate decreases drastically levels of environmental behaviors compared to the other fields compared to the previous questions (Appendix E). of study (p  <  0.05), while the behavior of the means of the other programs was much more stable. A slight relation among Exploratory Analysis Colombia PEBs per city and CC knowledge can be observed (Figure 3). In general, it was observed that students from cities possess Tunja stands out because it ranks second in the proportion middle to high knowledge regarding CC (  =  51.6; σ =7.8). of students with more CC knowledge and first in PEBs. In Frontiers in Psychology | www.frontiersin.org 8 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior fact, it is the only city with significant differences. No significant i.e., self-efficacy, trust in the institutions that provide differences were found for this dimension when analyzing environmental information and NEP. socioeconomic strata. Similarities were observed with the The results from the OLS models in both countries show Nicaraguan sample: students pay special attention to avoid that the dimensions of self-efficacy, trust in information sources wastage of electricity and water, but the proportion drops and NEP account for nearly 20% of the variation in the drastically when it comes to household waste separation explanatory power of the model as displayed by its adjusted (Appendix D). Environmental attitude presents the highest R-squared, with a slight reduction of the percentage of variation score among the Colombian students, who particularly display explained from these dimensions in the models employing a high level of affinity with the importance of building a PCA. On the other hand, the sociodemographic and educational better balance between humans and nature (  =  33; σ  =  5.2). controls account for 2% or less of the variation in the model When compared with Nicaragua, this dimension was for Nicaragua and 5% for Colombia, respectively. significantly higher among Colombian students (p  <  0.001). Adequate CC knowledge is associated with better PEBs, For this variable, humanities students display better behavior both in Colombia and in Nicaragua, although with greater than those studying engineering, economics, administrative magnitude in the latter. The coefficients are statistically significant sciences, communication, pedagogy, basic sciences, and at the 1% level in all specifications. In particular, the inclusion agricultural sciences (p < 0.05). There is also a better behavior of sociodemographic and educational control variables does among law students over communication and pedagogy not determine relevant changes in the correlation between the (p  <  0.05), economics students over pedagogy (p  <  0.05), variables. While the coefficient decreases when adding other and basic sciences over communication (p < 0.05). On average, indexes of the survey to the regression, it is suggested that the attitude’s score increases as the academic cycle progresses. the correlation between PEBs and attitudes is explained mostly When analyzing this variable by socioeconomic strata, the by self-efficacy, trust in the institutions that provide environmental means of the attitudes show a uniform behavior. Significant information, and NEP. differences among strata can be observed, but these differences In both countries, both NEP and self-efficacy are more do not have a definite pattern, or in other words, it is not correlated with PEBs than CC knowledge itself. In Nicaragua, possible to state whether as the strata increases there are the NEP coefficient indicates that for each increment of a better attitudes or vice versa. Cali stands out as the city with standard variation, the behavior changes 0.25 standard the highest mean behavior for this dimension (  =  34.3; deviations in the same direction; an increase in affinity with σ  =  5.2) and presents significant differences with all other the NEP indicator represents a greater increase in PEBs cities of this sample (p  <  0.01). Unlike in Nicaragua, the compared to the other dimensions analyzed. The same happens Colombian students’ responses on the usefulness of plants in Colombia, but with self-efficacy. For each variation of and animals for the satisfaction of human needs show greater one standard deviation, PEBs increases by 0.21 homogeneity in their distribution, mostly rejecting the statement. standard deviations. Likewise, it is observed that students are aware of the human- The consistency of the results was evaluated through a induced impact over natural resources (Appendix E). PCA (Tables 5 and 6). When all the variables are included, the explanatory factor measured by the R-squared indicator Regression Model: OLS and PCA is higher, which indicates a clear correlation between the According to the Brush Pagan test, the assumption of behavior and the variables analyzed. Even at a disaggregated homoscedasticity in the residuals is rejected. Therefore, robust level, results were similar, meaning that significant results are standard errors were included in the models. Multicollinearity not due to the aggregation of the variables in each index. problems are ruled out in the variables according to the In fact, the knowledge index for the survey questions has correlation matrix. Since there were no variables that could been decomposed and the correlation between each z-score be  affected by endogeneity, no control tests were performed. has been estimated (Annex 6 and 7), showing that most of Assumption of normality in the residuals was verified. the coefficients are positive and statistically significant at 1%, Ordinary Least Squares (OLS) regressions were estimated with the exception of the outliers mentioned in the for the integrated environmental behavior index. Tables 3 previous subsection. and 4 show the results of the OLS model with the z-score For both countries, statements with negative coefficients indicators for Nicaragua and Colombia, while Tables 5 and were identified. In Colombia and Nicaragua, the statement 6 show the results of the PCA. For both cases, the first “the industry is the sector that produces the highest level of column shows the result of the first regression, which measures GHG emissions” presented this effect. The same was found in the relationships between CC knowledge and PEBs without Colombia with the statement “the high amounts of ozone in including control variables. The second column shows the the atmosphere increase the ultraviolet radiation on the surface relationships between PEBs and CC knowledge, adding of the earth” and in Nicaragua with the statements “Nicaragua sociodemographic variables, while the third column shows is one of the main producers of GHG” and “the use of renewable the relationship between the dependent variable and CC energy can increase global warming”. knowledge, controlling both the socioeconomic and educational Although the results indicate patterns, they might be biased variables. Finally, the fourth and fifth columns include all due to omitted variables. Given the cross-sectional nature of the indicators for the different dimensions of the regression, the data used, it is not possible to identify the causal effect Frontiers in Psychology | www.frontiersin.org 9 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior TABLE 3 | Results of the OLS model for Nicaragua. TABLE 6 | Results PCA Colombia. (1) (2) (3) (4) (5) Dimension (1) (2) (3) (4) (5) CC Knowledge 0.538*** 0.561*** 0.563*** 0.113*** 0.119*** CC Knowledge 0.434*** 0.447*** 0.466*** 0.120*** 0.140*** (0.035) (0.041) (0.043) (0.032) (0.039) (0.037) (0.039) (0.040) (0.035) (0.036) Self-efficacy 0.160*** 0.162*** Self-efficacy 0.950*** 0.901*** (0.020) (0.023) (0.099) (0.095) Trust 0.140*** 0.126*** Trust 0.306*** 0.329*** (0.016) (0.019) (0.080) (0.079) NEP 0.235*** 0.249*** NEP 0.527*** 0.578*** (0.023) (0.026) (0.106) (0.104) Observations 2,312 1872 1713 2,253 1,677 Observations 1,216 1,174 1,148 1,213 1,145 R^2 adjusted 0.152 0.161 0.172 0.353 0.370 R^2 adjusted 0.183 0.218 0.237 0.332 0.378 Socio-economic Socio-economic controls No Yes Yes No Yes controls No Yes Yes No Yes Education controls No No Yes No Yes Education Controls No No Yes No Yes ***p < 0.01. ***p < 0.01. TABLE 4 | Results of the OLS model for Colombia. DISCUSSION Dimension (1) (2) (3) (4) (5) Our results can contribute to policy formulation and indicate CC Knowledge 0.383*** 0.404*** 0.408*** 0.084*** 0.097*** the direction of future research in various ways. First, results (0.030) (0.031) (0.031) (0.022) (0.023) Self-efficacy 0.229*** 0.211*** show that the R-squared is very similar between countries (0.013) (0.012) with considerable similarities but also important cultural and Trust 0.065*** 0.069*** social differences (38%). This reveals that the measured (0.011) (0.011) psychological and socioeconomic factors have similar effects NEP 0.180*** 0.200*** on PEBs across two cultures, but also that there is still a need (0.014) (0.014) Observations 4,764 4,563 4,460 4,744 4,442 to identify other dimensions or factors that explain the remaining R^2 ajusted 0.070 0.111 0.128 0.271 0.320 variability of the PEBs and if these too are consistent. Socio-economic Against our initial expectations, the results show that while controls No Yes Yes No Yes the sociodemographic and educational variables are significant Education controls No No Yes No Yes factors for explaining PEBs, their effect was relatively small, ***p < 0.01. compared to the included dimensions. This highlights the strength of idiosyncratic and cognitive factors for explaining PEBs. A possible hypothesis suggests that a “common culture” TABLE 5 | Results PCA Nicaragua. prevails in students not related to university education, but rather explained by informal communication processes that Dimension (1) (2) (3) (4) (5) occur in the social environment (Meira-Cartea et  al., 2018). CC Knowledge 0.455*** *** * * Thus, for example, widely disseminated knowledge, such as 0.494*** 0.475 0.087 0.103 (0.045) (0.050) (0.053) (0.048) (0.058) the effects of carbon dioxide in the atmosphere or the importance Self-efficacy 0.614*** 0.621*** of the ozone layer, seem to be  widely recognized, but the (0.145) (0.162) characterization of knowledge becomes more diffuse when Trust 0.526*** 0.458*** trying to clarify the role of chlorofluorocarbons in the mitigation (0.115) (0.137) NEP 1.038*** 1.181*** of climate change. Furthermore, this generalized knowledge (0.176) (0.202) denotes certain mistakes, such as when trying to clarify the Observations 736 625 583 730 579 contributions of nuclear energy (Appendix A). In accordance R^2 adjusted 0.219 0.236 0.236 0.378 0.383 with Meira-Cartea et  al. (2018), the prevalence of this general Socio-economic culture questions the relevance of the training received in its controls No Yes Yes No Yes Education controls No No Yes No Yes ability to influence the scientific and social representations of students, in this case, from both countries. *p < 0.10; ***p < 0.01. Beyond this, it is important to emphasize the need to conduct further research to identify additional sources of variation. of CC knowledge, environmental attitudes, or trust in the Some studies suggest that a greater degree of explanation can institutions that provide environmental information on PEBs. be achieved by taking into account additional internal dimensions, However, the data allow controlling for the sociodemographic external forces, and contextual factors. These include e.g., characteristics and educational level, and confirm the meaning physical infrastructure, technical facilities, availability of products and orientation of the proposed estimates. and their characteristics (Steg and Vlek, 2009), environmental Frontiers in Psychology | www.frontiersin.org 10 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior policies, financial strategies (Bertoldi, 2017), social norms and his or her own assessment of the personal effort or cost involved. the influence of the social nuclei of an individual, and the Truelove and Gillis (2018) found that for so-called laypeople duration of and adaptation to a technology (Truelove and Gillis, (people who are not experts in PEBs), the monetary costs of 2018; Li et  al., 2019). Recently, Truelove and Gillis (2018) carrying out a certain behavior outweighs the frequency a behavior revealed a new explanatory dimension: CC impacts on health needs to be  carried out. However, several other studies (e.g., and safety, which is of major relevance in developing countries Delmas et  al., 2013; Schwartz et  al., 2015; Steinhorst et  al., 2015) since they are the most vulnerable to CC. Other dimensions show that monetary incentives are not a good standalone solution with explanatory potential are those that combine internal and and might even be  counterproductive if they form the sole basis external forces (psychosocial constructs), understood as the of environmental behavior campaigns. It is much more useful interactions or socialization of a person with the physical world, to take into account the psychological, social, and civic stimuli and can stimulate the occurrence of behavior or sustainable of conservation, especially when it comes to a population group actions. Within this construct stands out not only the fact such as university students. While monetary framing might that socio-environmental actions depend on environmental be  efficient for one particular environmental action/behavior, identity, altruism or the sense of equity, but also the analysis environmental framing goes beyond that and can motivate of emotions (e.g., the appreciation of nature or feelings of additional non-targeted actions or behaviors (Steinhorst et al., 2015). resentment for ecological deterioration) (Pérez Ibarra et al., 2020). Thus, monetary incentives should rather be  considered a Next, based on the psychological factors included in this complementary element in pro-environmental campaigns. This is research, our results allow providing recommendations for the also consistent with findings from Gifford and Comeau (2011) who design of policies and other instruments intended to modify observed that young people participate to a lesser extent in mitigation student behavior toward a more sustainable direction. Although actions and thus should be  a target of motivational strategies. CC knowledge and trust in institutions showed significant Finally, the fact that in many cases no significant differences correlations with PEBs, attitudes and self-efficacy exhibit a were found could be  a result of the measurement scale, which much stronger relationship, which is why it could be  fruitful was used to provide continuity. Knowledge about CC has multiple to focus strategies on these dimensions. Following Spence and evaluation dimensions and not all of them have the same effect Pidgeon (2010), emphasizing on the potential negative impacts on attitudes and/or behaviors toward mitigation. Given that the of not carrying out actions, abandoning alarmist paradigms, knowledge dimension featured here was not only focused on the and generating more binding strategies (combining personal causes but also on the manifestations of CC, important information and collective solutions) – arise as directive vectors, which might have been lost in the aim of establishing a stronger have been proven useful techniques for students (Parant et  al., relationship. It is important to note that this situation does not 2016). Nevertheless, transmission of motivational communication imply that the used scale has measured knowledge incorrectly. should be  carried out carefully, as people may not respond On the contrary, Shi et al. (2016) recognize that CC is a complex adequately. Students in Colombia and Nicaragua (contrary to phenomenon, which results from a function of multiple causes the results presented by Vignola et  al. (2013) for the case of and presents various characteristics and consequences. Therefore, Costa  Rica) do not have sufficient trust in the institutions it is important to measure the perceptions of CC transversely. that generate such communications, particularly in governmental sources. Thus, actions addressed at increasing trust of public sources while better leveraging on the credibility of NGOs STUDY LIMITATIONS and the scientific community in the communication strategies for adaptation and mitigation, may provide better results. The methodology used for this study assumes that students The present study did not consider the impact of economic accurately reflect and report on their behaviors. However, there incentives and disincentives on PEBs, which could be an additional is no certainty on their actual PEBs. In order to examine the explanatory factor. However, it was observed that students weigh factors explaining actual behavior of a certain population, the effort or the costs related to certain PEBs. This allows different approaches would have been necessary. Given the establishing a clear differentiation between low-cost behaviors large sample size, time, and financial limitations, it was not and mobility behaviors, as described by Tobler et  al. (2012b). possible to adopt such approaches in the methodology. This In both countries, commitment with certain environmental actions indicates that there may be  a systematic difference between varied according to their rigidity. While closing the water tap what students say they do and what they actually do, which and turning off the light or the fan when not in use are commonly is not captured by the selected variables. The literature is applied actions among the studied population, other, more ambiguous regarding this issue. Corral-Verdugo and Figueredo determinant actions, like traveling less, recycling, or using the (1999) and Kormos and Gifford (2014) suggest that results bicycle instead of a car are less common. These results are in based on stated information may have high levels of validity line with several other studies, such as Tobler et  al. (2012b), in terms of predicting real behavior. Other researchers, such Steinhorst et  al. (2015) or Truelove and Gillis (2018), and can as Fuj et  al. (1985) and Sheeran (2002), argue that these types be  explained not only from the psychological dimension but of measurements may have inherent distortions in the process also from an overlapping economic dimension. Although the of explaining such behavior. social benefit of a given environmental action is greater than In addition, by adding different dimensions, not including the individual cost, the agent may prefer not to apply it given explicitly subjective norms and by combining both declared Frontiers in Psychology | www.frontiersin.org 11 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior behaviors and intentions in the PEB dimension, this study DATA AVAILABILITY STATEMENT extends and deviates from the initial framework provided by the TPB. This may hinder comparison with other models The raw data supporting the conclusions of this article will following this framework more strictly. Nevertheless, both the be  made available by the authors, without undue reservation. results from the Cronbach’s alpha and PCA show that the constructs used for measuring PEBs are internally consistent, and that the added elements provide new information to further ETHICS STATEMENT complement the original framework. The studies involving human participants were reviewed and approved by Internal Review Board; International Center for CONCLUSION Tropical Agriculture; E.Sweitzer@cgiar.org. Written informed consent to participate in this study was provided by the This study examined the relationships of four psychological participants’ legal guardian/next of kin. Written informed consent dimensions on PEBs of higher education students from two was obtained from the individual(s), and minor(s)’ legal guardian/ developing countries, Colombia and Nicaragua. This investigation next of kin, for the publication of any potentially identifiable responds to the tendency toward decentralization of global images or data included in this article. governance in environmental issues, in which more cost-efficient strategies and policies are needed. Therefore, this work is a contribution to creating empirical evidence for multiple authorities and decision-makers and helps in assessing the AUTHOR CONTRIBUTIONS capacity of the civil society to contribute to the fight against SB and MD: Conceptualization MD, AC, KE, MR, SS, and SB: CC, and thus p rovides valuable inputs for the design of more Methodology. MD, SS, MR, KE, and AC: Formal analysis. MD, effective and efficient environmental initiatives. SB, and AC: Writing the original draft and review and editing. By comparing countries with different cultural contexts and MD, SB, AC, and KE: Resources. SB: Supervision and funding political systems, this study provides strong evidence that CC acquisition. MD and SB: Project administration. All authors knowledge, trust in sources of environmental information, self- contributed to the article and approved the submitted version. efficacy, and environmental attitudes are important predictors of PEBs in a developing country’s population. Our literature review revealed an increasing preponderance of the topic, with studies being conducted around the globe. FUNDING In Latin America, a constant rise of such studies was observed, appearing largely in South American countries, while in the This work was funded by the CGIAR Research Program on Central American region they were conducted rather sporadically. Livestock. The funders had no role in the design of the study; Yet, findings remain obscure. Therefore, our study contributes in the collection, analyses, or interpretation of data; in the to expanding the body of knowledge for the region and provides writing of the manuscript, or in the decision to publish the results. a point of comparison for further research within the region and across other cultures. Our results highlight the importance of the scientific ACKNOWLEDGMENTS community and activists of generating and communicating information on CC for guiding public concerns toward This work was carried out as part of the CGIAR Research appropriate environmental behavior, and suggest that efforts Program on Livestock. We  thank all donors who globally should focus on teaching and communicating CC, emphasizing support our work through their contributions to the CGIAR on the potential impact of private and collective action, System. The views expressed in this document may not be taken establishing trust in institutions, and reducing an anthropocentric as the official views of these organizations. The authors wish vision of the world. This, in turn, can help people and to thank Jhon Jairo Muñoz, Rein van der Hoek, Martín Mena, policymakers to better address the risks and consequences of Elvis Chavarría, and Orlando Téllez for their support in data CC, and thus gain support in the construction and obtainment and logistics in Colombia and Nicaragua. Finally, implementation of effective adaptation and mitigation policies we would like to thank the professors and university authorities and plans. from Nicaragua and Colombia, who kindly supported our It should be noted that the relationships found in this study survey at their facilities. are dynamic and can vary as personal values, educational processes, or exposure to CC alter. What is not in doubt is that these changes are strongly dependent on the SUPPLEMENTARY MATERIAL effectiveness of public policy and its congruency with the specific realities of a target population. Thus, promoting a The Supplementary Material for this article can be found online continuous measurement of the analyzed variables will help at: https://www.frontiersin.org/articles/10.3389/fpsyg.2020.580730/ to improve policy design and communication strategies over time. full#supplementary-material Frontiers in Psychology | www.frontiersin.org 12 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior REFERENCES Cordano, M., Welcomer, S., Scherer, R., Pradenas, L., and Parada, V. (2010). Understanding cultural differences in the antecedents of pro– Adger, W. N., Dessai, S., Goulden, M., Hulme, M., Lorenzoni, I., Nelson, D. R., environmental behavior: a comparative analysis of business students et al. (2009). Are there social limits to adaptation to climate change? Clim. in the United  States and Chile. J. Environ. Educ. 41, 224–238. doi: Chang. 93, 335–354. doi: 10.1007/s10584-008-9520-z 10.1080/00958960903439997 Ajzen, I. (1985). “From intentions to actions: a theory of planned behavior” Corral-Verdugo, V., and Figueredo, A. J. (1999). Convergent and divergent in Action control: From cognition to behavior. eds. J. Kulh and J. Beckmann validity of three measures of conservation behavior: the multitrait-multimethod (Berlin, Heidelberg: Springer-Verlag), 11–39. approach. Environ. Behav. 31, 805–820. doi: 10.1177/00139169921972353 Ajzen, I. (1991). The theory of planned behavior. Organ. Behav. Hum. Decis. Davidson, D. J., and Freudenburg, W. R. (1996). Gender and environmental Process. 50, 179–211. doi: 10.1016/0749-5978(91)90020-T risk concerns: A review and analysis of available research. Environ. Behav. Ajzen, I., and Fishbein, M. (1980). Understanding attitudes and predicting social 28, 302–339. doi: 10.1177/0013916596283003 behavior. Englewood Cliffs: Prentice-Hall. Delmas, M. A., Fischlein, M., and Asensio, O. I. (2013). Information strategies Akerlof, K., Maibach, E. W., Fitzgerald, D., Cedeno, A. Y., and Neuman, A. and energy conservation behavior: A meta-analysis of experimental studies (2013). Do people “personally experience” global warming, and if so how, from 1975 to 2012. Energy Policy 61, 729–739. doi: 10.1016/j.enpol.2013.05.109 and does it matter? Glob. Environ. Chang. 23, 81–91. doi: 10.1016/j. Dietz, T., Dan, A., and Shwom, R. (2007). Support for climate change policy: gloenvcha.2012.07.006 social psychological and social structural influences. Rural. Sociol. 72, 185–214. Arbuthnot, J. (1977). The roles of attitudinal and personality variables in the doi: 10.1526/003601107781170026 prediction of environmental behavior and knowledge. Environ. Behav. 9, Dunlap, R. E., and Van Liere, K. D. (1978). The “new environmental paradigm.” 217–232. doi: 10.1177/001391657792004 J. Environ. Educ. 40, 19–28. doi: 10.3200/JOEE.40.1.19-28 Arcury, T. (1990). Environmental attitude and environmental knowledge. Hum. Emanuel, R., and Adams, J. N. (2011). College students’ perceptions of campus Organ. 49, 300–304. doi: 10.17730/humo.49.4.y6135676n433r880 sustainability. Int. J. Sustain. High. Educ. 12, 79–92. doi: 10.1108/ Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. 14676371111098320 Psychol. Rev. 84, 191–215. doi: 10.1037/0033-295x.84.2.191 Food and Agriculture Organization of the United Nations (2015). The state Bandura, A. (1991). Social cognitive theory of self-regulation. Organ. Behav. of food and agriculture. Social protection and agriculture: breaking the Hum. Decis. Process. 50, 248–287. doi: 10.1016/0749-5978(91)90022-L cycle of rural poverty 2010–2011. Available at: http://www.fao.org/publications/ Bang, H. K., Ellinger, A. E., Hadjimarcou, J., and Traichal, P. A. (2000). Consumer sofa/2015/en/ (Accessed February 5, 2020). concern, knowledge, belief, and attitude toward renewable energy: an application Freudenburg, W. R., and Davidson, D. J. (2007). Nuclear families and nuclear of the reasoned action theory. Psychol. Mark. 17, 449–468. doi: 10.1002/(SI risks: the effects of gender, geography, and progeny on attitudes toward a CI)1520-6793(200006)17:6<449::AID-MAR2>3.0.CO;2-8 nuclear waste facility. Rural. Sociol. 72, 215–243. doi: 10.1526/ Barazarte, R., Neaman, A., Vallejo, F., and García, P. (2014). El conocimiento 003601107781170017 ambiental y el comportamiento pro-ambiental de los estudiantes de la Fuj, E. T., Hennessy, M., and Mak, J. (1985). An evaluation of the validity enseñanza media. Rev. Educ. 364, 66–92. doi: 10.4438/1988-592X- and reliability of survey response data on household electricity conservation. RE-2014-364-255 Eval. Rev. 9, 93–104. doi: 10.1177/0193841X8500900106 Bárcena, A., Samaniego, J., Galindo, L. M., Ferrer, J., Alatorre, J. E., Stockins, P., Furlow, N. E., and Knott, C. (2014). Who’s Reading the label? Millenials’ use et al. (2018). La economía del cambio climático en América Latina y el of environmental product labels. JABE 10, 1–12. Caribe. Comisión Económica para América Latina y el Caribe. 1–62. Gifford, R., and Comeau, L. A. (2011). Message framing influences perceived Available at: https://www.cepal.org/es/publicaciones/42228-la-economia-cambio- climate change competence, engagement, and behavioral intentions. Glob. climatico-america-latina-caribe-vision-grafica (Accessed March 15, 2019). Environ. Chang. 21, 1301–1307. doi: 10.1016/j.gloenvcha.2011.06.004 Beckage, B., Gross, L. J., Lacasse, K., Carr, E., Metcalf, S. S., Winter, J. M., Goldenhar, L. M., and Connell, C. M. (1993). Understanding and predicting et al. (2018). Linking models of human behaviour and climate alters projected recycling behavior: an application of the theory of reasoned action. J. Environ. climate change. Nat. Clim. Chang. 8, 79–84. doi: 10.1038/s41558-017-0031-7 Syst. 22, 91–103. doi: 10.2190/92KU-NXLT-XC32-RHD6 Bedard, S. A. N., and Tolmie, C. R. (2018). Millennials’ green consumption González, E. G., and Maldonado, A. L. (2015). La percepción social del cambio behaviour: exploring the role of social media. Corp. Soc. Responsib. Environ. climático. Capítulo 5. Representaciones sociales y cambio climático el caso Manag. 25, 1–9. doi: 10.1002/csr.1654 de Veracruz. Universidad Iberoamericana Puebla. 82–106. http://repositorio. Bertoldi, P. (2017). Are current policies promoting a change in behavior, iberopuebla.mx/bitstream/handle/20.500.11777/247/Lapersepcionsocialdelcam conservation and sufficiency? An analysis of existing policies and bioclimatico-Gonzalez?sequence=1 (Accessed March 20, 2019). recommendations for new and effective policies. European Commission. Hallegatte, S., and Rozenberg, J. (2017). Climate change through a poverty 201–2011. Available at: https://ec.europa.eu/jrc/en/publication/are-current- Lens. Nat. Clim. Chang. 7, 250–256. doi: 10.1038/nclimate3253 policies-promoting-change-behaviour-conservation-and-sufficiency-analysis- Hardoy, J., and Pandiella, G. (2009). Urban poverty and vulnerability to climate existing change in Latin America. Environ. Urban. 21, 203–224. doi: Bertoni, M., and López, M. J. (2010). Percepciones sociales ambientales: Valores 10.1177/0956247809103019 y actitudes hacia la conservación de la reserva de biósfera “Parque Atlántico Truelove, H. B., and Gillis, A. J. (2018). Perception of pro-environmental behavior. Mar Chiquita”Argentina. Estudios y perspectivas en turismo 19, 835–849. Glob. Environ. Chang. 49, 175–185. doi: 10.1016/j.gloenvcha.2018.02.009 Blocker, T., and Eckberg, D. L. (1989). Environmental issues as women’s issues: Heath, Y., and Gifford, R. (2006). Free–market ideology and environmental general concerns and local hazards. Soc. Sci. Q. 70, 586–593. degradation: the case of belief in global climate change. Environ. Behav. 38, Bord, R. J., Fisher, A., and O’Connor, R. E. (1998). Public perceptions of global 47–71. doi: 10.1177/0013916505277998 warming: United  States and international perspectives. Clim. Res. 11, 75–84. Helm, S. V., Pollitt, A., Barnett, M. A., Curran, M. A., and Craig, Z. R. (2018). Bord, R. J., O’Connor, R. E., and Fisher, A. (2000). In what sense does the Differentiating environmental concern in the context of psychological adaptation public need to understand global climate change? Public Underst. Sci. 9, to climate change. Glob. Environ. Chang. 48, 158–167. doi: 10.1016/j. 205–218. doi: 10.1088/0963-6625/9/3/301 gloenvcha.2017.11.012 Bradley, J. C., Waliczek, T. M., and Zajicek, J. M. (1999). Relationship between Howe, P. D., Mildenber, M., Marlon, J. R., and Leiserowitz, A. (2015). Geographic environmental knowledge and environmental attitude of high school students. variation in opinions on climate change at scale and local scales in the J. Environ. Educ. 30, 17–21. doi: 10.1080/00958969909601873 USA. Nat. Clim. Chang. 5, 596–603. doi: 10.1038/nclimate2583 Burke, M., Hsiang, S. M., and Miguel, E. (2015). Global non-linear effect of Howe, P. D., Markowitz, E. M., Lee, T. M., KO, C. Y., and Leiserowitz, A. temperature on economic production. Nature 527, 235–239. doi: 10.1038/ (2013). Global perceptions of local temperatura change. Nat. Clim. Chang. nature15725 3, 352–356. doi: 10.1038/nclimate1768 Calixto, R. F., and Herrera, L. F. (2010). Estudio sobre las percepciones y la Ideam, PNUD, MADS, DNP, Cancillería, Observatorio de Ciencia y Tecnología educación ambiental. Tiempo de educar 11, 227–249. (2016). ¿Qué piensan los colombianos del Cambio Climático? Frontiers in Psychology | www.frontiersin.org 13 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior Primera encuesta nacional de percepción pública del cambio climático en implications. Glob. Environ. Chang. 17, 445–459. doi: 10.1016/j. Colombia. Tercera Comunicación Nacional de Cambio Climático. Bogotá, gloenvcha.2007.01.004 Colombia. Available at: http://www.minambiente.gov.co/images/ Markle, G. L. (2013). Pro-environmental behavior: does it matter how it’s AsuntosMarinosCosterosyRecursosAcuatico/que_piensan_los_colombianos_ measured? Development and validation of the pro-environmental behavior sobre_el_cambio_clim%C3%A1tico.pdf (Accessed January 20, 2019). scale (PEBS). Hum. Ecol. 41, 905–914. doi: 10.1007/s10745-013-9614-8 Intergovernmental Panel on Climate Change (2014a). Climate Change 2014: Impacts, McCright, A. M. (2010). The effects of gender on climate change knowledge Adaptation, and Vulnerability. Summaries, Frequently Asked Questions, and and concern in the American public. J. Popul. Environ. 32, 66–87. doi: Cross-Chapter Boxes. A Contribution of Working Group II to the Fifth 10.1007/s11111-010-0113-1 Assessment Report of the Intergovernmental Panel on Climate Change. eds. Meira-Cartea, P. A., Gutiérrez-Pérez, J., Arto-Blanco, M., and Escoz-Roldán, A. C. B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, M. D. Mastrandrea, T. E. Bilir et al (2018). Influence of academic education vs. common culture on the climate (World Meteorological Organization, Geneva, Switzerland). 190. literacy of university students/Formación académica frente a cultura común Intergovernmental Panel on Climate Change (2014b). Climate Change 2014: Impacts, en la alfabetización climática de estudiantes universitarios. Psyecology 9, Adaptation, and Vulnerability. Chapter11: Human Health: Impacts, adaptation, 301–340. doi: 10.1080/21711976.2018.1483569 and Co–benefits. eds. C. B. Field, V. R. Barros, D. J., Dokken, K. J. Mach, M. D. Mendelsohn, R., Dinar, A., and Williams, L. (2006). The distributional impact Mastrandrea, T. E. Bilir et al (United Kingdom and New York, NY, USA: of climate change on rich and poor countries. Environ. Dev. Econ. 11, Cambridge University Press), pp.  709–754. 159–178. doi: 10.1017/S1355770X05002755 Intergovernmental Panel on Climate Change (2018). Global warming of 1.5°C: Meyer, A. (2015). Does education increase pro-environmental behavior? Evidence an IPCC special report on the impacts of global warming of 1.5°C above from Europe. Ecol. Econ. 116, 108–121. doi: 10.1016/j.ecolecon.2015.04.018 pre-industrial levels and related global greenhouse gas emission pathways, Meyer, A. (2016). Heterogeneity in the preferences and pro-environmental behavior in the context of strengthening the global response to the threat of climate of college students: the effects of years on campus, demographics, and external change, sustainable development, and efforts to eradicate poverty. Chapter factors. J. Clean. Prod. 112, 3451–3463. doi: 10.1016/j.jclepro.2015.10.133 4: Strengthening and implementing the global response. Available at: http:// Mishra, D., Akman, I., and Mishra, A. (2014). Theory of reasoned action www.ipcc.ch/report/sr15/ (Accessed April 3, 2019). application for green information technology acceptance. Comput. Hum. International Monetary Fund (2017). World economic outlook: seeking sustainable Behav. 36, 29–40. doi: 10.1016/j.chb.2014.03.030 growth: short- term recovery, long-term challenges. Chapter 3: The effects Moons, I., and De Pelsmacker, P. (2015). An extended decomposed theory of of weather shocks on economic activity: how can low-income countries cope? planned behaviour to predict the usage intention of the electric car: A Available at: https://www.imf.org/en/Publications/WEO/Issues/2017/09/19/world- multi-group comparison. Sustainability 7, 6212–6245. doi: 10.3390/su7056212 economic-outlook-october-2017 (Accessed November 3, 2018). Motyl, M., Iyer, R., Oishi, S., Trawalter, S., and Nosek, B. A. (2014). How Kagawa, F. (2007). Dissonance in students’ perceptions of sustainable development ideological migration geographically segregates groups. J. Exp. Soc. Psychol. and sustainability: implications for curriculum change. Int. J. Sustain. High. 51, 1–14. doi: 10.1016/j.jesp.2013.10.010 Educ. 8, 317–338. doi: 10.1108/14676370710817174 O’Connor, R. E., Bard, R. J., and Fisher, A. (1999). Risk perceptions, general Kaiser, F. G., Wolfing, S., and Fuhrer, U. (1999). Environmental attitude and environmental beliefs, and willingness to address climate change. Risk Anal. ecological behavior. J. Environ. Psychol. 19, 1–19. doi: 10.1006/jevp.1998.0107 19, 461–471. doi: 10.1111/j.1539-6924.1999.tb00421.x Kellstedt, P. M., Zahran, S., and Vedlitz, A. (2008). Personal efficacy, the Otto, S., and Kaiser, F. G. (2014). Ecological behavior across the lifespan: why information environment, and attitudes toward global warming and climate environmentalism increases as people grow older. J. Environ. Psychol. 40, change in the United  States. Risk Anal. 28, 113–126. doi: 331–338. doi: 10.1016/j.jenvp.2014.08.004 10.1111/j.1539-6924.2008.01010.x Padilla y Sotelo, L. S., and Luna, A. M. (2003). Percepción y conocimiento Kim, S., Jeong, S., and Hwang, Y. (2013). Predictors of pro-environmental ambiental en la costa de Quintana Roo: una caracterización a través de behaviors of American and Korean students: the application of the theory encuestas. Investigaciones geográficas 52, 99–116. doi: 10.14350/rig.30335 of reasoned action and protection motivation theory. Sci. Commun. 35, Palavecinos, M., Amérigo, M., Ulloa, J. B., and Muñoz, J. (2016). Preocupación 168–188. doi: 10.1177/1075547012441692 y conducta ecológica responsable en estudiantes universitarios: estudio Kling, J., and Liebman, J. (2004). Experimental Analysis of Neighborhood Effects comparativo entre estudiantes chilenos y españoles. Psychosoc. Interv. 25, on Youth. John F. Kennedy School of Government, Harvard University. 143–148. doi: 10.1016/j.psi.2016.01.001 1–31. Parant, A., Pascual, A., Jugel, M., Kerroume, M., Felonneau, M., and Guéguen, N. Kling, J. R., Liebman, J. B., and Katz, L. F. (2007). Experimental analysis of (2016). Raising students awareness to climate change: an illustration with Neighborhood effects. Econometrica 75, 83–119. doi: binding communication. Environ. Behav. 49, 339–353. doi: 10.1177/ 10.1111/j.1468-0262.2007.00733.x 0013916516629191 Kormos, C., and Gifford, R. (2014). The validity of self–report measures of Pato, C., Ros, M., and Tamayo, A. (2005). Creencias y comportamiento ecológico: proenvironmental behavior: a meta–analytic review. J. Environ. Psychol. 40, un estudio empírico con estudiantes brasileños. Medio Ambiente y 359–371. doi: 10.1016/j.jenvp.2014.09.003 Comportamiento Humano 6, 5–22. Lauren, N., Fielding, K. S., Smith, L., and Louis, W. R. (2016). You did, so Pávez-Soto, I., León-Valdebenito, C., and Triadú-Figueras, V. (2016). Jóvenes you  can and you  will; self-efficacy as a mediator of spillover from easy to universitarios y medio ambiente en Chile: percepciones y comportamientos. more difficult pro-environmental behaviour. J. Environ. Psychol. 48, 191–199. Revista Latinoamericana de Ciencias Sociales Niñez y Juventud 14, 1435–1449. doi: 10.1016/j.jenvp.2016.10.004 doi: 10.11600/1692715x.14237041215 Lee, T. M., Markowitz, E. M., Howe, P. D., Ko, Ch., and Leiserowitz, A. A. Paço, A., and Lavrador, T. (2017). Environmental knowledge and attitudes and (2015). Predictors of public climate change awareness and risk perception behaviors towards energy consumption. J. Environ. Manag. 197, 384–392. around the world. Nat. Clim. Chang. 5, 1014–1020. doi: 10.1038/ doi: 10.1016/j.jenvman.2017.03.100 nclimate2728 Paul, J., Modi, A., and Patel, J. (2016). Predicting green product consumption Leiserowitz, A. (2006). Climate change risk perception and policy preferences: using theory of planned behavior and reasoned action. J. Retail. Consum. the role of affect, imagery, and values. Clim. Chang. 77, 45–72. doi: 10.1007/ Serv. 29, 123–134. doi: 10.1016/j.jretconser.2015.11.006 s10584-006-9059-9 Pe’er, S., Goldman, D., and Yavetz, B. (2007). Environmental literacy in teacher Li, D., Zhao, L., Ma, S., and Zhang, L. (2019). What influences an individual’s training: attitudes, knowledge, environmental behavior of beginning students. pro-environmental behavior? A literature review. Resour. Conserv. Recycl. J. Environ. Educ. 39, 45–59. doi: 10.3200/JOEE.39.1.45-59 146, 28–34. doi: 10.1016/j.resconrec.2019.03.024 Pérez Ibarra, R. E., Tapia-Fonllem, C. O., Fraijo-Sing, B. S., Nieblas Soto, N., Lorenzoni, I., and Pidgeon, N. F. (2006). Public views on climate change: and Poggio, L. (2020). Psychosocial predispositions towards sustainability European and USA perspectives. Clim. Chang. 77, 73–95. doi: 10.1007/ and their relationship with environmental identity. Sustainability 12:7195. s10584-006-9072-z doi: 10.3390/su12177195 Lorenzoni, I., Nicholson-Cole, S., and Whitmarsh, L. (2007). Barriers perceived Perez, C., Jones, E. M., Kristjanson, P., Cramer, L., Thornton, P. K., Förch, W., to engaging with climate change among the UK public and their policy et al. (2015). How resilient are farming households and communities to a Frontiers in Psychology | www.frontiersin.org 14 December 2020 | Volume 11 | Article 580730 Díaz et al. Psychological Factors Influencing Pro-environmental Behavior changing climate in Africa? A gender-based perspective. Glob. Environ. Stern, P. C. (2011). Design principles for global commons: natural resources Change 34, 95–107. doi: 10.1016/j.gloenvcha.2015.06.003 and emerging technologies. Int. J. Commons 5, 213–232. doi: 10.18352/ Perugini, M., and Bagozzi, R. P. (2001). The role of desires and anticipated ijc.305 emotions in goal-directed behaviours: broadening and deepening the theory Tikka, P. M., Kuitunen, M. T., and Tynys, S. M. (2000). Effects of educational of planned behaviour. Br. J. Soc. Psychol. 40, 79–98. doi: background on students’ attitudes, activity levels, and knowledge concerning 10.1348/014466601164704 the environment. J. Environ. Educ. 31, 12–19. doi: 10.1080/00958960009598640 Pinheiro, J. Q., and Farias, A. C. (2015). In search of a positive framework Tobler, C., Visschers, V. H. M., and Siegrist, M. (2012a). Consumers’ knowledge for communications about Global Climate Change. Positive communication about climate change. Clim. Chang. 114, 189–209. doi: 10.1007/ about Global Climate Change/En busca de un encuadre positivo en la s10584-011-0393-1 comunicación sobre el cambio climático global. Comunicaciones positivas Tobler, C., Visschers, V. H. M., and Siegrist, M. (2012b). Adressing climate sobre el cambio climático global. Psyecology 6, 229–251. doi: change: determinants of consumers’ willingness to act and to support policy 10.1080/21711976.2015.1026084 measures. J. Environ. Psychol. 32, 137–207. doi: 10.1016/j.jenvp.2012.02.001 Polonsky, M. J., Garma, R., and Landreth Grau, S. (2011). Western consumers’ United Nations, Department of Economic and Social Affairs (2016). World understanding of carbon offsets and its relationship to behavior. Asia Pac. Economic and Social Survey 2016: Climate change Resilience, an opportunity J. Mark. Logist. 23, 583–603. doi: 10.1108/13555851111183048 for reducing inequalities. Economic and Social Affairs, 2–21. Ramsey, C. E., and Rickson, R. E. (1976). Environmental knowledge and attitudes. Vicente-Molina, M., Fernández-Sáinz, A., and Izagirre-Olaizola, J. (2018). Does J. Environ. Educ. 8, 10–18. doi: 10.1080/00958964.1976.9941552 gender make a difference in pro-environmental behavior? The case of the Salehi, S., Pazuki, Z., Mahmoudi, H., and Burkart, S. (2016). Knowledge of Basque Country University students. J. Clean. Prod. 176, 89–98. doi: 10.1016/j. global climate change: view of Iranian university students. Int. Res. Geogr. jclepro.2017.12.079 Environ. Educ. 25, 226–243. doi: 10.1080/10382046.2016.1155322 Vignola, R., Klinsky, S., Tam, J., and Mcdaniels, T. (2013). Public perception, Sánchez, M., de la Garza-González, A., and Rodríguez, M. (2014). Análisis de knowledge and policy support for mitigation and adaptation to climate la percepción y conducta ambiental mediante una red bayesiana. Revista change in Costa  Rica: comparisons with north American and European de Psicología 23, 56–70. doi: 10.5354/0719-0581.2014.36148 studies. Mitig. Adapt. Strateg. Glob. Chang. 18, 303–323. doi: 10.1007/ Sapci, O., and Considine, T. (2014). The link between environmental attitudes s11027-012-9364-8 and energy consumption behavior. J. Behav. Exp. Econ. 52, 29–34. doi: Wachholz, S., Artz, N., and Chene, D. (2014). Warming to the idea: university 10.1016/j.socec.2014.06.001 students’ knowledge and attitudes about climate change. Int. J. Sustain. High. Schwartz, D., Bruine de Bruin, W., Fischhoff, B., and Lave, L. (2015). Advertising Educ. 15, 128–141. doi: 10.1108/IJSHE-03-2012-0025 energy saving programs: the potential environmental cost of emphasizing Wan, C., Shen, G. Q., and Choi, S. (2017). Experiential and instrumental monetary savings. J. Exp. Psychol. Appl. 21, 158–166. doi: 10.1037/xap0000042 attitudes: interaction effect of attitude and subjective norm on recycling Sheeran, P. (2002). Intention – behavior relations: a conceptual and empirical intention. J. Environ. Psychol. 50, 69–79. doi: 10.1016/j.jenvp.2017.02.006 review. Eur. Rev. Soc. Psychol. 12, 1–36. doi: 10.1080/14792772143000003 Yadav, R., and Pathak, G. S. (2016). Young consumers’ intention towards buying Shi, J., Visschers, V. H. M., Siegrist, M., and Arvai, J. (2016). Knowledge as green products in a developing nation: extending the theory of planned a driver of public perceptions about climate change reassessed. Nat. Clim. behavior. J. Clean. Prod. 135, 732–739. doi: 10.1016/j.jclepro.2016.06.120 Chang. 6, 759–762. doi: 10.1038/nclimate2997 Yuriev, A., Dahmen, M., Paillé, P., Boiral, O., and Guillaumie, L. (2020). Sidique, S. F., Lupi, F., and Joshi, S. V. (2010). The effects of behavior and Pro-environmental behaviors through the lens of the theory of planned attitudes on drop-off recycling activities. Resour. Conserv. Recycl. 56, 163–170. behavior: a scoping review. Resour. Conserv. Recycl. 155:104660. doi: 10.1016/j. doi: 10.1016/j.resconrec.2009.07.012 resconrec.2019.104660 Spellman, G., Field, K., and Sinclair, J. (2003). Assessing UK higher education Zhang, B., van der Linden, S., Mildenberger, M., Marlon, J. R., Howe, P. D., students’ awareness of global climatic change. Weather 58, 212–219. doi: and Leiserowitz, A. (2018). Experimental effects of climate messages vary 10.1256/wea.62.02 geographically. Nat. Clim. Chang. 8, 370–374. doi: 10.1038/s41558-018-0122-0 Spence, A., and Pidgeon, N. (2010). Framing communicating climate change: the effects of distance and outcome frame manipulations. Glob. Environ. Conflict of Interest: The authors declare that the research was conducted in Chang. 20, 656–667. doi: 10.1016/j.gloenvcha.2010.07.002 the absence of any commercial or financial relationships that could be construed Steinhorst, J., Klöckner, C., and Matthies, E. (2015). Saving electricity – for as a potential conflict of interest. the money or the environment? Risks of limiting pro-environmental spillover when using monetary framing. J. Environ. Psychol. 43, 125–135. doi: 10.1016/j. Copyright © 2020 Díaz, Charry, Sellitti, Ruzzante, Enciso and Burkart. This is an jenvp.2015.05.012 open-access article distributed under the terms of the Creative Commons Attribution Steg, L., and Vlek, C. (2009). Encouraging pro-environmental behaviour: an License (CC BY). The use, distribution or reproduction in other forums is permitted, integrative review and research agenda. J. Environ. Psychol. 29, 309–317. provided the original author(s) and the copyright owner(s) are credited and that doi: 10.1016/j.jenvp.2008.10.004 the original publication in this journal is cited, in accordance with accepted academic Stern, P. C. (2000). Toward a coherent theory of environmentally significant practice. No use, distribution or reproduction is permitted which does not comply behavior. J. Soc. Issues 56, 407–424. doi: 10.1111/0022-4537.00175 with these terms. Frontiers in Psychology | www.frontiersin.org 15 December 2020 | Volume 11 | Article 580730