Establishing the value-psychological-educational dimensions for “learning to action” model for pro-environmental behaviour

Abstract

Pro-environmental behaviour is imperative to promote sustainable management and consumption of energy in the fight against climate change. The Theory of Planned Behaviour (TPB) has been widely used to explain and predict behaviour in a multitude of behavioural domains including pro-environmental behaviour. However, the TPB does not prioritise the influence of knowledge and habits on pro-environmental behaviour. Past studies also confirmed the influence of cultural factors in predicting environmental intent and behaviour, which were not covered by TPB. Considering that Malaysia is a multicultural country, on top of the environmental education factors, cognitive competencies, and affective factors, the influence of values that contribute to the pro-environmental intention behaviour on energy conservation has also to be considered. Hence, this study employed Fuzzy Delphi Method to determine the indicators explaining the effects of educational and psychological factors on pro-environmental behaviour. It was hypothesized that educational factors, cognitive competencies and affective factors influence secondary students’ pro-environmental behaviour on energy conservation. The students’ cultural viewpoint was also proposed as the regulating effect on their energy conservation behaviour. A total of 25 experts were selected to validate the questionnaire developed. The items/constructs developed were environmental factors (formal, non-formal and informal educations), cognitive factors, affective factors (attitude, subjective norm, perceived behavioural control and civic values) and environmental conservation behaviour. The value construct, which comprised of values of empathy, respect, cooperation, responsibility, justice, equality, integrity, and altruism. Majority of the items/constructs were accepted by the experts. This reflects that the items/constructs developed were relevant to the study as the experts agreed with these items/constructs.

Keywords:

• pro-environmental behaviour
• value-psychological-educational dimensions
• Theory Planned Behaviour
• Fuzzy Delphi method

PUBLIC INTEREST STATEMENT

This study was conducted in Malaysia—a multicultural country, to investigate the influence of knowledge, habits, and cultural factors on pro-environmental behaviour to promote sustainable management and consumption of energy in the fight against climate change. On top of the environmental education factors, cognitive competencies, and affective factors, the influence of values that contribute to the pro-environmental intention behaviour on energy conservation has also to be considered. Items/constructs (questionnaire) related to these factors: environmental (formal, non-formal, and informal educations), cognitive, affective (attitude, subjective norm, perceived behavioural control, and civic values), and environmental conservation behaviour, were developed and validated by 25 experts, before distributed to secondary students to assess their pro-environmental behaviour on energy conservation. The value construct consists of empathy, respect, cooperation, responsibility, justice, equality, integrity, and altruism. The items/constructs developed were relevant to the study as the majority of the items/constructs were accepted by the experts.

1. Introduction

Anthropogenic activities have been known as the main contributor to climate change and global warming. In the past few decades, extreme weather such as unusual rainfall and prolonged drought fuelled by climate change have become more common worldwide, threatening not only humanity but also causing irreversible damage on the ecosystem (Shaftel, 2022). The physical impact of climate change also contributes to psychological, emotional, and social effect to human (Noremy, 2020). As a result of these pressing issues, Sustainable Development Goals (SDGs) was established by the United Nations in 2015 as a blueprint to achieve a better and more sustainable future for all (United Nations Development Programme, 2021). Accordingly, in order to minimise the adverse consequences of climate change, suggestion was put forth for global warming to be limited to less than 1.5°C above its preindustrial level. To meet this goal, all countries need to take immediate action to reduce global carbon dioxide (CO2) emissions by about 45% by 2030 (Guldberg et al., 2018). Since a significant amount of these emissions can be traced back to the energy sector (31%), managing energy demand and energy consumption has become an important measure to curb CO2 emissions (Center for Climate and Energy Solutions, 2022). For instance, energy consumption in the residential sector currently accounts for 30% of global energy demand (Skodienė et al., 2020). It is projected that the energy demand from the residential sector will continue to rise because of economic growth, improved quality of life, and population increase, which will subsequently challenge the effort in reducing CO2 emissions (Gago et al., 2011; Santiago et al., 2014; Skodienė et al., 2020). Hence, it is imperative that environmental ethics are nurtured among the people and behaviours that are more pro-environmental are induced through various approaches such as education, provision of environmental information, introduction of environmental regulations, and the use of environmental taxes and charges (Lucas et al., 2008; Rioux, 2011).

Environmental education plays a central role in addressing the challenges of climate change and to promote pro-environmental behaviours (Alves & Azeiteiro, 2018; Chawla & Cushing, 2007; Hungerford & Volk, 1990; Mochizuki & Bryan,). The fundamental component of environmental education is the promotion of environmental knowledge, and it is a prerequisite to foster environmentally significant behaviour (Misra & Panda, 2017; Von Hauff & Kuhnke, 2017). Climate change education can contribute to capacity-building, helping to develop systemic and holistic thinking skills among the young learners and preparing them to better understand the issues as well as the consequences of their actions for solving the global climate crisis (Monroe, 2003; Wolfgang et al., 2014).

Thus, as mentioned earlier, schools play a crucial role in helping the young generation develop global competence. Educating for global competence can help form new generations who care about global issues and environmental challenges. The 2030 Agenda for Sustainable Development recognises the critical role of education in reaching the sustainability goals which include ensuring that learners acquire the knowledge and skills needed to promote sustainable development, and among the ways this can be achieved are through education for sustainable development and promotion of global citizenship and appreciation of cultural diversity and of culture’s contribution to sustainable development (PREPARING OUR YOUTH FOR AN INCLUSIVE AND SUSTAINABLE WORLD, 2018) Hence, the young generation needs a solid foundation in environmental issues in order to promote and support sustainability through pro-environmental behaviour.

Thus, this study aimed to explore and determine the environmental factors (formal and non-formal education), cognitive competencies (knowledge), and affective factors (attitude, subjective norm, control perceived behaviour and values) that contribute to the pro-environmental intention behaviour on energy conservation by using the Fuzzy Delphi Method.

2. Literature review

Environmental Citizenship is a relatively new concept of defining pro-environmental actions and their driving forces which functions as a unifying term for a holistic pattern of pro-environmental behaviours. It is a term that encompasses an array of characteristics such as the knowledge, skills, attitudes, values and beliefs that are needed to address environmental issues on climate change (Ajzen, 2020) Thus, educating young Malaysians to become environmental citizens is a crucial step that needs to be taken, with a strong basis of evidence for actions. Previous studies claim that values can potentially be formed and changed through education (Bruno et al., 2015). This is key for educating pro-environmental citizens, since, despite some innate components that are associated with personality traits, most individuals have the capacity for change in their value orientations, adapting them based on how their environmental influences change. To date, there is no single study that can lend an insight into how value orientations relate to all the components of Environmental Citizenship. However, many studies are available pertaining to how values relate to a very important component of Environmental Citizenship, i.e., pro-environmental behaviours (Myyry et al., 2013).

Pro-environmental behaviours have been studied through the lens of the TPB since 1995 as the theory states that the main driver for behaviour is the intention to perform the behaviour. The immediate antecedent of behaviour in the TPB is the stronger the intention, the more likely it is that the behaviour will follow. Research has revealed that intentions are the best predictors of behaviour (Steg et al., 2012). Similarly, (Christropher & Mark, 2010) claimed that behavioural intention is widely recognised as a key to actual behaviour.

Analysis of literature showed that some studies employed extended versions of the TPB model by introducing new variables as direct predictors of intention or behaviour. However, (Ajzen, 2020) argued that no additional constructs should be required to secure accurate prediction of intention and behaviour. Nonetheless, the TPB is, in principle, open to the inclusion of additional predictors. Among the most popular additional variables include moral norm, self-identity, past behaviour, and habit (Ajzen, 2002). Some of these additional variables such as environmental values and environmental awareness are specific to pro-environmental behaviour. Such domain-specific predictors should be prioritised in studies on pro-environmental behaviours because more studies are needed to draw convincing conclusions on the increased predictive power of extended versions of the TPB for pro-environmental behaviours. The variables (direct and indirect) of the original theory of planned behaviour do not contain multiple aspects that sometimes influence individual behaviour (Ajzen, 2002). It has also been frequently suggested for the Theory of Planned Behaviour to be integrated with another model such as Norm Activation Model (NAM) to explain pro-environmental behaviours (Heesup, 2014; Heesup et al., 2017). The TPB is generally more appropriate for explaining the self-interest aspect of pro-environmental behaviours, while the NAM is more applicable for interpreting the pro-social or pro-environmental aspect of the behaviours. Analysis of selected publications indicates that studies with additional variables always perform better and on average, increase the explained variance for intention and behaviour. Thus, (Masoud & Masoumen, 2015)found that adding personal moral norm as an additional construct to the TPB significantly increased the explanatory power of the original model. The idea of including moral norms is based on the fact that individuals’ intentions and actual behaviours do not always depend solely on cost benefit calculations (Botetzagias et al., 2015; Chan & Bishop, 2013). The finding of (Heesup et al., 2017) revealed that an extended TPB model which included moral obligation raised the proportion of the explained variance of intention to perform pro-environmental behaviours. Meanwhile, environmental values are an ordered set of beliefs about desirable end states that guide the evaluation or selection of pro-environmental behaviours (Morgan et al., 2019).

Educational approach is the key focus for country-specific strategies and adaptation measures to build resilience against climate change and to promote climate action (Anderson, 2012; Reid, 2019; SLYCAN Trust (GTE) Ltd, 2020; UNESCO (2021),National Research Council, 2012). Environment competencies to cope with climate change can be delivered to the public, especially youth, through formal, non-formal, and informal education. It is targeted to provide children and youth worldwide access to innovative and effective forms of climate change education in order to empower them with the knowledge, skills, values, and attitudes needed to act as agents of change in their community and environment (Rousell et al., 2020).

Education systems need to be adapted and reoriented to find the best way to integrate climate change education in order to provide understanding of the causes, effects, and consequences of anthropogenic climate change and to help develop competencies so that students eventually develop and perform climate-friendly behaviours (Anderson, 2012). As a response to climate change impact and issues, it is important that students acquire cognitive competencies that involve creative thinking (focuses on various thinking styles such as legislative, global, and local thinking styles) and critical thinking (includes reasoning, making inferences, self-reflection, and coordination of multiple views) [86]. Students’ socio-emotional needs regarding the environment must also be addressed appropriately especially as information about climate change’s impacts has become more salient and future projections have become increasingly overwhelming (Hermans, 2016; Ojala & Bengtsson, 2018). In this context, socio-emotional domain that includes skills and capacities in communication such as active listening, clarity, cross-cultural, non-verbal, respect, empathy, open-mindedness, and digital literacy has been set as one of the core dimensions for global citizenship education (Toh et al., 2017).

Environmental awareness created from education will induce a social behaviour to save energy for a more sustainable environment. Whitburn et al. discovered that a deeper connection to nature may result in greater engagement in pro-environmental behaviour and conservation (Whitburn et al., 2020).

In contrast, Paço et al. found that there was a lack of relationship between knowledge and attitudes, and between knowledge and environmental behaviour related to energy consumption (Paço & Lavrador, 2017). Jakučionytė-Skodienė et al. also reported that pro-environmental behaviour has limited impact on energy consumption and CO₂ emissions in the case of Lithuana (Skodienė et al., 2020). They suggested for different factors to be considered to promote energy savings and CO₂ reduction in the residential sector in relation to heating and energy consumption. Nattavudh Powdthavee found little evidence that more education could improve pro-environmental behaviour (Powdthavee, 2020). Thus, it can be said that the relationship between environmental knowledge and behaviours is complex, and that pro-environmental behaviours could be affected by other factors. In this study, the factors proposed in the theoretical framework involved environmental factors that contribute to the development of “affective”, “cognitive” and “actional” dimensions in addition to the environmental factors that affect the development of cognitive and affective dimensions which influence pro-environmental intention behaviour.

3. Methodology

The Fuzzy Delphi method was used in this study to identify the indicators of a theoretical model in explaining the education-psychological towards pro-environmental behaviour. The Fuzzy Delphi method is a combination of the Delphi method and Fuzzy Theory which aims to resolve the shortcomings of the traditional Delphi method. The subjectivity of the Delphi method and the requirement of having multiple rounds of questionnaire and feedback from experts to reach a consensus, which is quite time-consuming, are the shortcomings of the Delphi method (Zhang, 2017). Hence, the combination with Fuzzy Theory which takes into consideration the fuzziness in group decision-making in the Delphi method serves to resolve these shortcomings (Lei & Huang, 2018). The Fuzzy Delphi method is an effectual tool for gathering data generated from experts’ opinions which are subject to uncertainty and imprecision, transforming it into quasi-objective quantitative data for easier decision-making in a particular issue (Saido et al., 2018).

It is hypothesised that educational factors (formal and non-formal education), cognitive competencies and affective factors (attitudes and values) influence secondary students’ pro-environmental behaviour on energy conservation. Thus, in this study, a hypothetical model of secondary students’ behaviour towards energy conservation based on the aforementioned factors was tested. This hypothetical model was further tested based on the students’ cultural viewpoint which was proposed as the regulating effect on their energy conservation behaviour. The novelty of this research is twofold: a) an extended model of the Theory of Planned Behaviour was used to predict and explain the pro-environmental energy conservation behaviour of the secondary school students; and b) students’ culture as a moderating factor would address the practical research gap in formulating effective and realistic climate change policies which are more relevant for a diverse society like Malaysia. The identification of these factors (educational, cognitive, affective, and culture) is vital in ensuring effective transformative climate change education towards meeting the climate change challenges locally and globally.

Six steps were implemented when using the FDM technique so as to ensure the study be considered empirical. The sequence of steps performed by the researcher are (i) Determination and selection of experts; (ii) Development of the FDM questionnaire; (iii) Dissemination and data collection; (iv) Likert scale conversion to Fuzzy scale (v) Data analysis (Triangular Fuzzy Number) to determine the threshold value (d), alpha cut and expert agreement percentage; (vi) Data interpretation.

3.1. STEP 1: determination and selection of experts

In FDM research, expert selection is a vital criterion. According to (Hasson et al., 2000) FDM is used in a study to obtain expert agreement; hence, the sampling method is of the non-probability and purposive sampling type as well as based on criteria. Every result and data obtained is based on expert agreement and consensus. Therefore, the number of experts is an element that needs to be carefully considered. Literature shows that the proposed number of experts appropriate for content validation through the FDM technique ranges from 5 to 50 participants (Clayton, 1997; Fiander & Burns, 1998; Gordon, 1994). For expert validation through the FDM technique in this study, a total of 25 experts were selected and identified for examining and evaluating as well as validating the items that had been developed for the questionnaire (research instrument).

According to (Saaty & Özdemir, 2014) the addition of a large number of but inexperienced experts will weaken the accuracy of the results. Therefore, for the purpose of this study, the selected panel of experts met the following criteria:

1. Knowledgeable in fields of study related to education, psychology, sociology, environment or fields related to climate change and student behaviour.

2. Experienced in the field studied and have at least five years of research experience in the related field, and for academic practitioners (teachers), have at least ten years of teaching experience in the lower secondary Science subject and also involved in programmes related to environmental conservation in school.

3. Have no personal interest to avoid biasness in making the validation.

The panel of experts that had been identified and were involved in providing recommendations and validation of the questionnaire items for the study’s model are as illustrated in Table 1.

Table 1. List of experts and expertise field

Position	Expertise	Number
HLI Lecturers	Environmental Education, STEM Education, Science Education, Informal Education, Ethnic Studies	9
Education Officers	Science and Mathematics Curriculum Development, Environmental Education Curriculum Development, Values Education	5
Teachers	Teaching lower secondary Science for more than 10 years	11
TOTAL		25

3.2. STEP 2: developing the FDM questionnaire

The process of constructing the questionnaire was similar to the development of a regular questionnaire. However, the expert questionnaire is more structured and includes literature sources and the result of the expert interviews conducted.

In ensuring that the study conducted by the researcher is impeccable and of a high impact, the selection of the Likert scale for the FDM questionnaire also plays an important role. This is necessary to ensure that the study instrument validated by the FDM method is empirical and capable of measuring the items desired. The Likert scale for this FDM study used a 7-point Likert scale as it is believed to be capable of providing a more accurate and precise fuzzy score value as well as capable of reducing the ambiguity value for each item studied. For the 5-point Likert scale, the ambiguity value is 20% while for the 7-point Likert scale, it is 3.3%. Thus, the researcher chose to use the 7-point Likert scale for the purpose of item validation through FDM in this study.

3.3. STEP 3: dissemination and data collection

This study was conducted during the nationwide implementation of Movement Control Order (MCO) because of the spread of COVID-19 virus, when physical meeting was prohibited. Thus, the dissemination of the questionnaire was performed through emails which were sent out to the identified experts. The number of experts identified was as described in Step 1. In the email, the researcher included an attachment of the executive summary of the research which contained the research objectives, research questions, conceptual framework, and operational definition of each construct and sub-construct developed. The experts were given a period of two weeks to answer the FDM questionnaire provided.

3.4. STEP 4: conversion of the likert scale to fuzzy scale

The analysis of the study data for the FDM was based on the conditions contained in the triangular fuzzy numbers and the defuzzification process. The conditions for the triangular fuzzy numbers involved the threshold value (d) and also the expert agreement percentage. The findings obtained from the 7-point Likert scale was converted to a fuzzy scale as shown in Table 2.

Table 2. The fuzzy scale for the 7-point Likert scale variables

7-point Likert Scale Variables	Fuzzy Scale
1 (Totally disagree)	(0.0, 0.0, 0.1)
2 (Strongly disagree)	(0.0, 0.1, 0.3)
3 (Disagree)	(0.1, 0.3, 0.5)
4 (Moderately agree)	(0.3, 0.5, 0.7)
5 (Agree)	(0.5, 0.7,0.9)
6 (Strongly agree)	(0.7, 0.9, 1.0)
7 (Totally agree)	(0.9, 1.0, 1.0)

3.5. STEP 5: data analysis

There are two items in FDM that need to be given consideration, and these are Triangular Fuzzy Numbers and the Defuzzification Process. Triangular Fuzzy Numbers consist of the values $m_1,\hspace{0.17em}{m}_2,\mathrm{d}\mathrm{a}\mathrm{n}{m}_3$ where $m_1$ represents the smallest value, $m_2$ represents the most plausible value, and $m_3$ refers to the maximum value. The three values in the Triangular Fuzzy Numbers are as illustrated in Figure 1 which shows the graph of the mean triangle against the triangular value.

Figure 1. The graph of mean triangle against the triangular value.

In Triangular Fuzzy Numbers, two conditions need to be met to determine the acceptance of a developed item. The first condition is the threshold value (d), while the second condition is the expert agreement percentage for each of the items. The threshold value (d) for each item measured should be less than or equal to 0.2 (Chen, 2000; Cheng & Lin, 2002), and the expert agreement percentage must be greater than or equal to 75% (Chu & Hwang, 2008).

The threshold value (d) is calculated based on the mathematical formula shown below. For each expert, the vertex method is used to calculate the distance between the average $r_{ij}$ (Chen, 2000). Meanwhile, the distance between the two fuzzy numbers, i.e., $m=\left(m_1,m_2,m_{3\hspace{0.17em}}\right)$ and $n=\left(n_1,n_2,n_3\right)$ is calculated using the formula as shown below:

(1) $d m, n=\sqrt{\frac{1}{3}\left[{\left(m_1-n_1\right)}^2+{\left(m_2-n_2\right)}^2+{\left(m_3-n_3\right)}^2\right]}$(1)

3.5.1. Questionnaire respondents

This study invited 25 experts and scholars to complete the FDM questionnaire (Table 3). The scholars and experts comprised two groups; the first group consisted of science teachers who teach secondary students as well as environmental education curriculum developers. The second group comprised university professors and researchers who specialise in related fields, which included the fields of environmental education, lower secondary science subject, formal and informal education, civic values, and science curriculum.

Table 3. The average response and alpha-cut value of the expert consensus

Dimension	Construct	Subconstruct	Average response			Alpha-cut
Education environment	Formal education	Teaching and learning in science class	0.790	0.926	0.965	0.894
		Science textbook/ practical book	0.792	0.925	0.970	0.896
		Assessment during science class	0.754	0.902	0.971	0.876
	Non-formal education	Co-curricular activities	0.763	0.902	0.960	0.875
		School facilities	0.823	0.981	0.986	0.920
	Informal education		0.762	0.903	0.957	0.876
Cognitive competence	Knowledge	The causes of climate change	0.779	0.914	0.970	0.888
		The signs of climate change	0.809	0.940	0.986	0.912
		The effects of climate change	0.745	0.895	0.958	0.888
		Mitigation of climate change	0.766	0.902	0.954	0.874
	Critical thinking		0.791	0.923	0.973	0.896
Affective competence	Attitude		0.773	0.919	0.978	0.890
	Subjective norm	Family	0.737	0.889	0.965	0.864
		Local community	0.727	0.875	0.952	0.851
	Perceived behavioural control		0.772	0.917	0.979	0.886
	Values	Empathy	0.777	0.920	0.979	0.892
		Respect	0.788	0.929	0.972	0.899
		Co-operation	0.801	0.940	0.988	0.910
		Responsible	0.831	0.960	0.995	0.929
		Justice	0.775	0.92	0.979	0.891
		Equity	0.796	0.931	0.976	0.901
		Integrity	0.823	0.952	0.988	0.921
		Altruistic	0.766	0.911	0.969	0.882
Intention to behave pro-environmentally	Pro-environmental behaviour	Passive	0.798	0.822	0.965	0.862
		Participative	0.799	0.935	0.966	0.897
		Socially Responsible	0.798	0.926	0.968	0.897

Data were analysed according to the constructs (variables) of the study. The appropriateness of the constructs and items was based on the threshold value (d) and the expert agreement percentage discussed in Step 5. In the following subsection, expert agreement for the items is discussed based on the constructs.

3.5.2. Construct of environmental factors

Formal education (Curriculum)

Table 4 shows the threshold value (d) for each item as well as the expert agreement percentage for the construct of formal education. Table 5 Results of the analysis showed that all the items were accepted except for items A5, B4 and C5 and the factor for rejection

Table 4. The threshold value and expert agreement of items on environmental factors

Construct/Subconstruct	Item	Conditions of Triangular Fuzzy Numbers		Expert Agreement
		Threshold Value, d	Expert Agreement Percentage
Teachers’ teaching and learning in class	A1	0.097	92%	ACCEPTED
	A2	0.070	96%	ACCEPTED
	A3	0.160	76%	ACCEPTED
	A4	0.075	96%	ACCEPTED
	A5	0.140	72%	REJECTED
Content of textbook/practical book	B1	0.114	96%	ACCEPTED
	B2	0.140	92%	ACCEPTED
	B3	0.123	92%	ACCEPTED
	B4	0.097	72%	REJECTED
	B5	0.093	92%	ACCEPTED
Evaluation during science class	C1	0.133	76%	ACCEPTED
	C2	0.143	76%	ACCEPTED
	C3	0.119	84%	ACCEPTED
	C4	0.159	80%	ACCEPTED
	C5	0.123	68%	REJECTED

Table 5. Description and reasons for rejecting items in formal education

Item	Description	Reason for rejecting the item
A5	Sharing information about environmental issues.	Not clear how the information is shared in the textbook or the science practical book.
B4	Contain interesting additional information about the environment.	‘The “additional information” being referred to is misleading.
C5	Presenting about environmental issues.	The form of the presentation is unclear, and the learning period is limited.

Non-formal education (co-curriculum)

Table 6 shows the threshold value (d) for each item as well as the expert agreement percentage for the construct of non-formal education.

Table 6. The threshold value and expert agreement of items in non-formal education

Construct/Subconstruct	Item	Conditions of Triangular Fuzzy Numbers		Expert Agreement
		Threshold Value, d	Expert Agreement Percentage
Cocurricular	D1	0.151	84%	ACCEPTED
	D2	0.234	72%	REJECTED
	D3	0.171	76%	ACCEPTED
	D4	0.173	88%	ACCEPTED
	D5	0.115	96%	ACCEPTED
	D6	0.070	96%	ACCEPTED
	D7	0.079	92%	ACCEPTED
	D8	0.099	88%	ACCEPTED
	D9	0.152	84%	ACCEPTED
	D10	0.106	88%	ACCEPTED

Table 7 shows the results of the analysis showed that all items were accepted except for item D2. The factor for rejection is as follows:

Table 7. Description and reason for rejecting item in non- formal education

Item	Description	Reason for rejecting the item
D2	Organise activities related to the environment as a recycling programme together with friends.	The activity could not be carried out during the pandemic.

Informal education

Table 8 shows the threshold value (d) for each item and the expert agreement percentage for the construct of informal education. Table 9 shows the r the factors of rejections of items E2, E3 and E4

Table 8. The threshold value and expert agreement of items on environmental factors

Construct/Subconstruct	Item	Conditions of Triangular Fuzzy Numbers		Expert Agreement
		Threshold Value, d	Expert Agreement Percentage
Informal education	E1	0.108	92%	ACCEPTED
	E2	0.231	36%	REJECTED
	E3	0.212	72%	REJECTED
	E4	0.238	36%	REJECTED
	E5	0.152	76%	ACCEPTED
	E6	0.192	76%	ACCEPTED
	E7	0.174	76%	ACCEPTED
	E8	0.085	92%	ACCEPTED
	E9	0.110	92%	ACCEPTED
	E10	0.079	92%	ACCEPTED
	E11	0.136	88%	ACCEPTED

Table 9. Description and the reason for rejecting items on informal education

Item	Description	Reason for rejecting the item
E2	Radio	Students do not listen to the radio.
E3	Newspaper	Students do not read the newspaper.
E4	Magazines	Not many magazines write about the environment

However, the researcher still retained the item on radio and merged it with the items on newspaper and magazines. The researcher retained the items above because there is no evidence that state students do not listen to the radio or read the newspaper or magazines. Nevertheless, the researcher had combined the items on newspaper and magazines and added e-newspaper and e-magazines as suggested by the experts. There was also the addition of the item of agencies or institutions (science centre, planetarium, museum) on the recommendation of the experts.

3.5.3. Cognitive factors

Table 10 shows the threshold value (d) and the expert agreement percentage for the construct of knowledge.

Table 10. The threshold value and expert agreement of items on cognitive factors

Construct/Subconstruct	Item	Conditions of Triangular Fuzzy Numbers		Expert Agreement
		Threshold Value, d	Expert Agreement Percentage
Knowledge (Causes of climate change)	F1	0.057	96%	ACCEPTED
	F2	0.085	92%	ACCEPTED
	F3	0.120	84%	ACCEPTED
	F4	0.219	52%	REJECTED
	F5	0.199	76%	ACCEPTED
	F6	0.120	84%	ACCEPTED
Knowledge (Signs of climate change)	G1	0.116	84%	ACCEPTED
	G2	0.105	88%	ACCEPTED
	G3	0.155	88%	ACCEPTED
	G4	0.064	96%	ACCEPTED
	G5	0.075	96%	ACCEPTED
Knowledge (Effects of climate change)	H1	0.098	96%	ACCEPTED
	H2	0.173	80%	ACCEPTED
	H3	0.209	76%	REJECTED
	H4	0.236	36%	REJECTED
	H5	0.135	84%	ACCEPTED
Knowledge (Mitigation measures)	J1	0.154	88%	ACCEPTED
	J2	0.138	88%	ACCEPTED
	J3	0.202	84%	REJECTED
	J4	0.188	84%	ACCEPTED
	J5	0.219	76%	REJECTED

Results of the analysis showed that items F4, H3, H4, J3 and J5 were rejected. However, the researcher retained item F4 because the climate change expert stated that the water treatment process is a source of greenhouse gas emissions especially carbon dioxide although students could not make the connection as mentioned in the experts’ comment. The subconstruct of signs of climate change was discarded as it overlapped with the effects of climate change and the number of items was insufficient. Thus, the construct of knowledge only contained three subconstructs following the experts’ recommendation, namely cause and effects of global climate change as well as the climate change mitigation measures. Items J3 and J5 had the highest percentage of agreement which was above 80% and for this reason, the researcher retained the items by improving the sentences. Table 11 shows the summary of expert comments on the reasons for rejecting the items.

Table 11. Description and reason for rejecting items in knowledge construct

Item	Description	Reason for rejecting the item
F4	Water treatment process	Students are not able to relate water treatment as a source of greenhouse gas emissions.
H3	The tourist sector is affected because of the occurrence of natural disasters that destroy the beautiful environment.	Has the same meaning as the items on signs of climate change happening.
H4	Human health is affected because of the increase in aedes mosquito breeding grounds which causes dengue.	The item is not suitable because the problem of dengue fever is related to other factors.
J3	Practising water conservation in daily life.	The sentence on water conservation practices is unclear. The item requires sentence reconstruction.
J5	Using recycled paper	Item needs to provide a broader picture according to the hierarchy of importance in mitigating climate change.

3.5.4. Affective factors

Attitude

Table 12 shows the threshold value (d) for each item and the expert agreement percentage for the construct of attitude. Table 13 shows the factor of rejection of item K6.

Table 12. The threshold value and expert agreement of items on affective factors

Construct/Subconstruct	Item	Conditions of Triangular Fuzzy Numbers		Expert Agreement
		Threshold Value, d	Expert Agreement Percentage
Attitude	K1	0.105	88%	ACCEPTED
	K2	0.181	84%	ACCEPTED
	K3	0.086	96%	ACCEPTED
	K4	0.097	92%	ACCEPTED
	K5	0.096	92%	ACCEPTED
	K6	0.236	84%	REJECTED
	K7	0.095	92%	ACCEPTED
	K8	0.132	84%	ACCEPTED
	K9	0.108	88%	ACCEPTED

Table 13. Description and reason for rejecting items of construct of attitude

Item	Description	Reason for rejecting the item
K6	I need to practice water conservation to reduce global warming.	Unclear and confusing with the use of the word “global warming”.

Results of the analysis showed that item K6 was rejected because the item statement was confusing. Thus, only item K6 was dropped, making the total number of items for the construct of attitude only eight (8).

Construct of subjective norm

Table 14 shows the threshold value (d) for each item and the expert agreement percentage for the construct of subjective norm. Table 15 shows the factors of rejections of item L3 ,L4, and L5.

Table 14. The threshold value and expert agreement of items on construct of subjective norm

Construct/Subconstruct	Item	Conditions of Triangular Fuzzy Numbers		Expert Agreement
		Threshold Value, d	Expert Agreement Percentage
Subjective norm	L1	0.137	80%	ACCEPTED
	L2	0.152	76%	ACCEPTED
	L3	0.182	72%	REJECTED
	L4	0.170	68%	REJECTED
	L5	0.227	36%	REJECTED
	L6	0.165	80%	ACCEPTED

Table 15. Description and reason for rejecting items in construct of subjective norm

Item	Description	Reason for rejecting the item
L3	The school is of the opinion that each student should be involved in environmental conservation activities.	Opinions alone are not enough to influence students to conserve the environment.
L4	My family often talks about the issue of climate change or global warming.	Discussions are not enough to influence children to conserve the environment.
L5	My parents are of the opinion that I should practice water conservation and electricity saving at home as an environmental conservation measure.	The factor of water and electricity saving at home is more geared towards saving the water or electricity bills than for reasons of conserving the environment.

Results of the analysis showed that items L3, L4 and L5 were rejected based on the factor of expert agreement percentage that was less than 75%.

Construct of perceived behavioural control

Table 16 shows the threshold value (d) for each item and the expert agreement percentage for the construct of perceived behavioural control.

Table 16. The threshold value and expert agreement of items on construct of perceived behavioural control

Construct/Sub-construct	Item	Conditions of Triangular Fuzzy Number		Expert Agreement
		Threshold Value, d	Expert Agreement Percentage
Perceived behavioural control	M1	0.132	80%	ACCEPTED
	M2	0.129	80%	ACCEPTED
	M3	0.142	84%	ACCEPTED
	M4	0.120	84%	ACCEPTED
	M5	0.120	84%	ACCEPTED

Results of the analysis showed that all the items proposed were accepted by the experts.

Value construct

Table 17 shows the threshold value (d) for each item and the expert agreement percentage for the construct of value.

Table 17. The threshold value and expert agreement of items on value construct

Construct/Subconstruct	Item	Conditions of Triangular Fuzzy Numbers		Expert Agreement
		Threshold Value, d	Expert Agreement Percentage
Value of Empathy	N1	0.097	92%	ACCEPTED
	N2	0.158	76%	ACCEPTED
	N3	0.135	84%	ACCEPTED
Value of Respect	P1	0.157	80%	ACCEPTED
	P2	0.209	80%	REJECTED
	P3	0.083	96%	ACCEPTED
	P4	0.127	92%	ACCEPTED
Value of Cooperation	Q1	0.111	92%	ACCEPTED
	Q2	0.093	92%	ACCEPTED
	Q3	0.108	88%	ACCEPTED
Value of Responsibility	R1	0.083	96%	ACCEPTED
	R2	0.064	96%	ACCEPTED
	R3	0.093	92%	ACCEPTED
Value of Justice	S1	0.134	84%	ACCEPTED
	S2	0.145	84%	ACCEPTED
	S3	0.097	92%	ACCEPTED
Value of Equality	T1	0.108	88%	ACCEPTED
	T2	0.083	96%	ACCEPTED
	T3	0.155	88%	ACCEPTED
Value of Integrity	U1	0.143	92%	ACCEPTED
	U2	0.079	96%	ACCEPTED
	U3	0.070	96%	ACCEPTED
Altruistic Value	V1	0.139	88%	ACCEPTED
	V2	0.178	84%	ACCEPTED
	V3	0.093	92%	ACCEPTED
	V4	0.097	92%	ACCEPTED

Results of the analysis showed that item P2 for the value of respect was rejected. Thus, the value of respect subconstruct had only three (3) items. The factor for the rejection is shown in Table 18.

Table 18. Description and reason for rejecting items in value construct

Item	Description	Reason for rejecting the item
P2	Respect the environment because of compliance towards and respect for the rules stipulated by the government.	The sentence which contains “because” and “and” confuses the students in understanding the meaning of the sentence.

3.5.5. Pro-environmental conservation behaviour

Table 19 shows the threshold value (d) for each item and the expert agreement percentage for the construct of environmental conservation behaviour.

Table 19. The threshold value and expert agreement of items on pro-environmental behaviour

Construct/ subconstruct	Item	Conditions of Triangular Fuzzy Numbers		Expert Agreement
		Threshold value, d	Expert Agreement Percentage
Passive members	W1	0.154	88%	ACCEPTED
	W2	0.138	92%	ACCEPTED
	W3	0.142	92%	ACCEPTED
	W4	0.153	88%	ACCEPTED
	W5	0.172	88%	ACCEPTED
	W6	0.142	92%	ACCEPTED
	W7	0.142	92%	ACCEPTED
	W8	0.142	92%	ACCEPTED
	W9	0.155	88%	ACCEPTED
Active members	X1	0.154	88%	ACCEPTED
	X2	0.169	84%	ACCEPTED
	X3	0.140	92%	ACCEPTED
	X4	0.155	88%	ACCEPTED
	X5	0.140	92%	ACCEPTED
	X6	0.140	92%	ACCEPTED
	X7	0.142	92%	ACCEPTED
	X8	0.155	88%	ACCEPTED
	X9	0.138	92%	ACCEPTED
Socially responsible members	Y1	0.142	92%	ACCEPTED
	Y2	0.155	88%	ACCEPTED
	Y3	0.138	92%	ACCEPTED
	Y4	0.140	92%	ACCEPTED
	Y5	0.143	92%	ACCEPTED
	Y6	0.140	92%	ACCEPTED
	Y7	0.142	92%	ACCEPTED
	Y8	0.172	88%	ACCEPTED
	Y9	0.142	92%	ACCEPTED

4. Discussion and implications

Transformative education emphasises the importance of education in realising the aspiration of individuals and the nation as well as ensuring the universality of human beings and the global society (Padilla, 2017). This study focused on climate change mitigation in particular pro-environmental behaviours as a measure to reduce carbon dioxide emission which is the main cause of climate change. The researcher established that the behaviours to be studied are environmental conservation behaviours as well as factors that influence those behaviours. The contributing factors studied included environmental factors (curriculum and cocurricular), cognitive competence factors (knowledge), and affective competence factors (attitude towards environmental conservation, social norm, perceived behavioural control and civic values).

A person’s behavioural intention is determined and influenced by the person’s attitude towards that behaviour as well as the individual’s subjective norm (Ajzen & Fishbein, 2004). However, TPB only focuses on the aspects of the individual’s psychology or affective competence as a predictor factor of intention for certain behaviours only. Previous researchers have argued that affective competence can be developed through life experiences gained through formal, non-formal and informal education (Hammond et al., 2020). Therefore, the researcher took into consideration the social cognitive theory (SCT) in the model developed for the study. The SCT explains that an individual’s behaviour is based on two variables that act as predictors of intention for climate change behaviour, namely personal factors (cognitive competence) and also environmental factors (educational approach that contributes to cognitive competence) (Bandura, 2001). In conclusion, the factors in the proposed conceptual model are to explain the environmental factors that contribute to the cognitive, affective, and behavioural (intent) development. This model can also further strengthen the predictive power of behavioural intention proposed by the TPB.

4.1. Environmental factors—formal education (curriculum and co-curricular)

According to SCT, the behavioural factors and cognitive factors are interrelated with environmental factors. Previous researchers have argued that educational approaches through formal, non-formal and informal education are needed so that the realities of climate change can be handled through the formation of global competencies (Mochizuki & Bryan,) Formal education for environmental conservation can be shaped effectively through formal education in particular through curriculum that is local in nature and through students’ experience. Teaching and learning that involves students thinking critically and productively as well as critically discussing the problems faced in facing the challenges of the effects of climate change is greatly needed (Sonowal, 2009). Moreover, teachers who have global competencies and practice the values and behaviours will also help in educating students to face the challenges of climate change such as global warming or environmental conservation behaviour (Sonowal, 2009).

Non-formal education is education that happens outside the class or outside the school but still under the same management as the formal education (PREPARING OUR YOUTH FOR AN INCLUSIVE AND SUSTAINABLE WORLD, 2018). Non-formal education is more flexible in terms of the curriculum, delivery methods and place of learning even though the learning is still structured and planned (Shala & Grajcevci, 2016). School activities can also be implemented with the involvement and participation of the local community. Thus, it can be maintained that the cocurricular activities in formal education further complements the curriculum of the formal education.

4.2. Cognitive competence—knowledge, critical thinking

Based on the hypothesised model, cognitive competence refers to the students’ science knowledge about climate change and also environmental conservation. Scientific knowledge about environmental conservation, causes, effects, and solutions to environmental conservation is necessary (Padilla, 2017).

4.3. Affective competence—attitude, subjective norm, perceived behavioural control and values

Affective competence for global competencies is socio-emotional skills that refer to values, attitude, and social skills that are developed affectively, psychosocially and physically, enabling students to live together peacefully and harmoniously (Padilla, 2017).

According to TPB, the factors that influence a person’s behavioural intentions depend on the individual’s attitude towards the behaviour, subjective norm, and perceived behavioural control. Previous studies that are related to environmental conservation behaviours (attitude, values, and beliefs) have shown that individuals who have a positive attitude towards the environment will have the intention to conserve the environment (Steg et al., 2005). In fact, (Lee & Tanusia, 2016) also emphasised that a person’s attitude is crucial in influencing a person’s behaviour to conserve the environment. Furthermore, the study conducted by (Wells et al., 2016) managed to show that attitude has a significant relationship with specific environmental conservation behaviours and at the same time, can influence the individual to perform those behaviours at home and also at the workplace.

According to TPB, the factors that influence a person’s intention to conserve the environment is the individual’s attitude towards the environmental conservation behaviour. Attitude is a person’s tendency or inclination to be favourable or unfavourable towards an object. The object intended in this study refers to environmental conservation behaviours. The three components of attitude examined are as proposed by (Lee & Tanusia, 2016; Ostrom, 1969) which are affective components (feelings towards the object), behavioural (behaviour towards the object), and cognitive (perception and evaluation of the object).

In addition, intention for environmental conservation behaviour is also influenced by subjective norm. Subjective norm refers to social influence and the influence of the surrounding community on the environmental conservation behaviour while perceived behavioural control is the belief to perform the behaviour after considering the knowledge, skills, time and also opportunity to practice the behaviour (Lee & Tanusia, 2016). As one of the predictors in TPB, subjective norm is influenced by the belief to perform the environmental conservation behaviour and the pressures exerted by family, friends, and the society (Masud et al., 2016).

Perceived behavioural control is related to beliefs about the individual’s capability to perform the behaviour and in this study, it refers to environmental conservation behaviour after considering the positive and negative factors of the behaviour (cognitive competence). Steg et al., (2005) found that users are more predisposed to environmental conservation behaviours if they are aware that their action will have a negative impact on the environment. Past studies have also demonstrated that intention towards environmental conservation behaviour practised at home (in the household) is significantly influenced by perceived behavioural control and also positive attitude towards the behaviour.

For a student who has affective competence, they will be more critical and open as well as capable of making sound judgement in resolving the effects of climate change where the student has a sense of belonging, has shared values, is responsible, empathetic, has solidarity and is respectful of the various differences (UNESCO, 2015). Nevertheless, the developed model is based on the civic values that have been adapted according to the Malaysian context. There are eight (8) values that have achieved the consensus and recommendation of the experts, namely the values of empathy, respect, cooperation, responsibility, fairness, equity, integrity and altruism.

5. Intention for environmental conservation behaviour

Intention towards environmental conservation refers to the subjective probability dimension of an individual which is associated with environmental conservation behaviour (Lee & Tanusia, 2016). Based on the study by (Octav-Ionut & Macovei, 2015) which focused on the issue of global warming and action towards the challenges of the effects of climate change, intention towards environmental conservation behaviour is explained as an action that can reduce greenhouse gas emissions. A person’s actual behaviour is based on the intensity of the person’s intention towards the behaviour (Ajzen, 2020). Therefore, this model was developed based on TPB which takes into consideration attitude, subjective norm and perceived behavioural control towards environmental conservation. The developed model will examine in depth the intentions of students towards pro-environmental behaviour by categorising the students’ level of involvement in terms of passive behaviour, active involvement, and social responsibility. Global citizenship requires students with cognitive and affective competencies who are able to act and participate not only for themselves or the local community but also for the common good of all global citizens. Thus, this hypothesis model connects attitude towards environmental conservation in the affective domain to intention towards environmental conservation behaviour. Intention towards the behaviour examined in this study will be further divided into three categories according to the students’ level of involvement. The three categories are passive involvement, active involvement (participative) and social responsibility.

6. Conclusion

In conclusion, the findings of the study indicate that all the constructs and subconstructs for the environmental factors, cognitive factors, and affective factors that contribute to environmental conservation behaviour are accepted based on expert agreement. However, as discussed above, a few items were rejected and had to be improved before conducting the pilot study. All the experts are of the opinion that the factors are necessary in global competencies to contribute towards environmental conservation behaviour which is a mitigation to climate change.

Acknowledgements

This research was funded by the Transdisciplinary Research Grant Scheme (TRGS/1/2019/UKM/01/3/4) which is a project funded by the Ministry of Higher Education Malaysia.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Ministry of Higher Education Malaysia [TRGS/1/2019/UKM/01/3/4].

Notes on contributors

Noremy Md. Akhir

The research reported in this paper is one of the key aspects of our transdisciplinary project. The transdisciplinary research aims to establish a transformative climate change education framework focusing on a) developing civic values related to the environment, b) case-based studies on climate change, and c) augmented reality pedagogical tools for climate change scenarios. The research also aims to establish a model of “learning to action” to explain an individual’s intention to behave in a pro-environmental way. This theoretical model consists of multiple factors (educational, civic values, attitude, and perceived norms) contributing to pro-environmental behaviour. The paper reports on the development of a questionnaire that measures these factors. The questionnaire is novel since it is a combination of factors derived from Social Cognitive Theory and the Theory of Planned Behavior.