We investigate whether there are green returns to education, where formal education encourages pro-environmental behaviours using nationally representative surveys on environmental issues in Thailand. To establish the causal relationship between education and green behaviours, we exploit the instrumental variables strategy using the supply of state primary schooling i.e. the corresponding number of teachers per 1000 children, which varies over time and across regions as the instrument, while controlling for regional, cohort and income effects. We find that more years of schooling lead to a greater probability of taking knowledge-based environmentally-friendly actions a great deal, but not cost-saving pro-environmental actions. In addition, the paper finds no significant impact of formal education on concern about global warming nor the willingness to pay for environmental tax.

To help protect the environment people need to change current non-environmental behaviours into more sustainable behaviours. By understanding habits for pro-environmental behaviours, people can strive toward building a more sustainable future. The goal of this thesis was to identify different types of repeated pro-environmental behaviours and how to best foster strong habits for pro-environmental behaviours. To achieve this purpose we combined the theoretical framework of habits (Bargh, 1994; Verplanken, 2006) with the theoretical framework of self-determination theory (SDT; Deci & Ryan, 1985; 2000) to address current gaps in the habit literature. The present thesis is comprised of two articles. The first article determined if the proposed indicators of habits (i.e., behaviour frequency, habit strength, and behaviour interference) could identify different patterns of repeated pro-environmental behaviours predicted by habit and self-determination theory using cluster analysis. The three studies provided support for the three proposed types of repeated behaviours: weak habits, repeated behaviours with interference, and strong habits. These results were very robust since the same pattern was found across three studies, three samples, and three different target behaviours. The second article used the groups (i.e., weak habit, repeated behaviour with interference, and strong habit) in exploratory multinomial logistic regression analyses to identify factors related to pro-environmental behaviours which distinguished between the different types of repeated behaviours. Once again, the same three types of repeated behaviours were found across these three studies, with three new samples and two different target behaviours. Three predictors fairly consistently identified differences between the groups: autonomous motivation toward the environment, perceived importance of the environment, and the frequency of other pro-environmental behaviours. The implications of the thesis findings are discussed in relation to self-determination theory and the study of habits.

Despite the increasing attention being paid to waste recycling, there is a dearth of both empirical evidence on recycling at work and examination of any spillover effects of recycling behaviour from home to work. Situated at the confluence of three social science debates (the study of recycling set within the waste management literature; the examination of spillover in the social psychology literature, and the work on pro-environmental behaviour at work in the organisational behaviour literature), this research seeks to understand recycling at work and the relationship between recycling behaviour at home and recycling at work using a sequential mixed methods approach. Due to the complexity of human behaviours including the heterogeneity of the factors underpinning recycling, this research adopts a sequential mixed methods approach with its pragmatic philosophical assumptions to examine recycling at work. Initially, semi-structured interviews with 15 key informants from different organisations including environmental/waste organisations in the UK were conducted. The findings from the interviews were used along with the evidence from the literature to develop the conceptual model and the research hypotheses. The quantitative data were collected, using a web-based questionnaire survey, from 367 respondents representing 43 different organisations across the UK. The collected quantitative data were analysed using SPSS for windows and IBM AMOS for path and causal analyses. Based on the findings, this research demonstrates that contextual factors such as organisational support are better determinants of recycling at work than personality/psychological factors such as attitudes that have dominated empirical and theoretical studies on pro-environmental behaviours for decades. Also, the findings of this research suggest that the concept of spillover of recycling from home to work is complex and inconsistent. Whilst there is a tendency for spillover of recycling behaviour, there is a significant difference between recycling at home and at work with regards to the volume of materials, the range of materials, and frequency of recycling. Nonetheless, the PROCESS macro allows the identification of various conditions that are likely to facilitate spillover of recycling from home to work. As a result, factors that are likely to determine recycling at work including the possible spillover of recycling from home to work are classified into personal/psychological and situational factors. These findings contribute to the existing bodies of knowledge on recycling behaviour, spillover effects, and organisational citizenship behaviour for the environment (OCBE). Also, the findings could assist businesses in finding proactive measures to increase recycling within their organisations. This would consequently reduce the total amount of resources being disposed of in the UK landfill sites.

The degradation of the environment and climate change represent some of the most important environmental issues affecting our society today, and we need to better understand what can be done in order to mitigate the negative effects of human activity on the environment. The present program of research proposes to examine, through three studies, how the frequency of pro-environmental behaviours (PEB) and harmful-environmental behaviours (HEB) are related to self-determined (SDM) and non-self-determined motivation (NSDM), stages of change (SOC), the satisfaction versus the frustration of basic psychological needs as defined by Self-Determination Theory, and nature relatedness. In Study 1 (N = 377), a scale was created to measure two distinct types of environmental behaviours, PEB and HEB, and the relationships between these types of environmental behaviours and SDM and NSDM were examined. Results demonstrated that both types of behaviours loaded on two distinct factors when conducting an exploratory factor analysis, that they were associated differently with environmental motivation, and that levels of motivation were different according to the reported frequency of adoption of PEB and HEB. In Study 2a (N = 266) and 2b (N = 529), the role of stages of change (SOC) was introduced to determine whether environmental motivation and behaviours were different across SOC and to examine whether SOC played a mediating or moderating role in the relationship between environmental motivation and behaviours. Results indicated that people in the later SOC displayed higher levels of SDM and PEB while people in the earlier stages reported higher levels of NSDM and HEB. It was also observed that SOC partially mediated the relationship between environmental motivation and behaviours. In Study 3 (N = 507) nature relatedness was compared to the satisfaction and frustration of basic psychological needs proposed by SDT to examine their respective roles in the association of motivation, SOC, PEB and HEB. Results revealed that nature relatedness was incremental to need satisfaction and frustration with regard to levels of SDM. Also, nature relatedness was an important determinant of SOC and both types of environmental behaviours. As for the proposed model, it was demonstrated that high levels of nature relatedness were associated with higher PEB and lower HEB through partial mediation by SDM and SOC, while need frustration was linked to a higher reported frequency of HEB through mediation by NSDM. Overall this thesis expands on SDT research by highlighting the importance of including HEB in conjunction with PEB and by demonstrating the prominent roles that SOC and nature relatedness could play in the motivational process associated with environmental action.